Sei sulla pagina 1di 623

Private Equity

FINANCIAL MARKETS AND INVESTMENTS SERIES


H. Kent Baker and Greg Filbeck, Series Editors

Portfolio Theory and Management


Edited by H. Kent Baker and Greg Filbeck

Public Real Estate Markets and Investments


Edited by H. Kent Baker and Peter Chinloy

Private Real Estate Markets and Investments


Edited by H. Kent Baker and Peter Chinloy

Investment Risk Management


Edited by H. Kent Baker and Greg Filbeck

Private Equity: Opportunities and Risks


Edited by H. Kent Baker, Greg Filbeck, and Halil Kiymaz
Private Equity
OPPORTU NI TI E S AND RIS KS

EDITED BY H. KENT BAKER

GREG FILBECK

and

HALIL KIYMAZ

1
1
Oxford University Press is a department of the University of Oxford.
It furthers the Universitys objective of excellence in research, scholarship,
and education by publishing worldwide. Oxford is a registered trade mark
of Oxford University Press in the UK and in certain other countries.

Published in the United States of America by


Oxford University Press
198 Madison Avenue, New York, NY 10016,
United States of America

Oxford University Press 2015

All rights reserved. No part of this publication may be reproduced,


stored in a retrieval system, or transmitted, in any form or by any means,
without the prior permission in writing of Oxford University Press,
or as expressly permitted by law, by license, or under terms agreed with the
appropriate reproduction rights organization. Inquiries concerning reproduction
outside the scope of the above should be sent to the Rights Department,
Oxford University Press, at the address above.

You must not circulate this work in any other form


and you must impose this same condition on any acquirer

Library of Congress Cataloging-in-Publication Data


Private equity : opportunities and risks / edited by H. Kent Baker, Greg Filbeck, and Halil Kiymaz.
p. cm. (Financial markets and investments series)
Includes bibliographical references and index.
ISBN 9780199375875 (alk. paper)
1. Private equity. I. Baker, H. Kent (Harold Kent), 1944 II. Filbeck, Greg, 1963. III. Kiymaz, Halil, 1964
HG4751.P753 2015
332.632044dc23
2014046431

987654321

Printed in the United States of America on acid free paper


Contents

List of Figures viii


List of Tables xi
Acknowledgments xv
About the Co-Editors xvi
About the Contributors xviii
Abbreviations xxviii

Part One INTRODUCTION

1. Private Equity: An Overview 3


h. kent baker, greg filbeck, and halil kiymaz

2. Economics of Private Equity 16


s h a n ta n u d u t ta , a r u p g a n g u ly , a n d l i n g e

3. Private Equity in the United States and Europe: Market


and Regulatory Developments 32
a l e x a n d r o s s e r e ta k i s

Part Two MAJOR TYPES OF PRIVATE EQUITY

4. Venture Capital in Europe: Myths and Reality 49


jean-michel sahut and eric braune

5. Leveraged Buyouts 66
christian rauch and marc p. umber
6. Mezzanine Capital and Commercial Real Estate 84
j. dean heller

7. Distressed Debt Investments 100


stephen g. moyer and john d. martin

Part Three HOW PRIVATE EQUITY WORKS

8. Valuing Private Equity 123


h a r l a n p l at t a n d e m e r y a . t r a h a n

9. Cost of Capital for Private Equity 147


alain con and aurlie desfleurs

10. Liquidity Issues in Private Equity: Lessons Learned


in the PostFinancial Crisis 166
e l i f a k b e n s e l c u k a n d ay s e d i l a r a a l t i o k y i l m a z

11. Private Equity Portfolio Management: Challenges, Approaches,


and Implementation 181
thomas meyer and tom weidig

12. The Role of Private Equity in Initial Public Offerings:


The Case of Venture Capital Firms 200
s h a n ta n u d u t ta , a r u p g a n g u ly , a n d l i n g e

13. Exit Strategies in Private Equity 215


didier folus and emmanuel boutron

Part Four PERFORMANCE AND MEASUREMENT

14. Private Equity Returns: Cross-Country Evidence 239


kenneth small and jeffrey s. smith

15. Benchmark Biases in Private Equity Performance 257


robert spliid

16. Return Persistence: Finding Top Quartile Managers 274


adrian oberli

17. Private Equity Due Diligence 290


manu sharma and esha prashar
Part Five PRIVATE EQUITY: USES AND STRUCTURE

18. Institutional Investors and Private Equity 311


pa r v e z a h m e d

19. Private Equity and Value Creation 330


pter harbula

20. Compensation Structure 360


ji-woong chung

21. Global Regulatory and Ethical Framework 377


henry ordower

Part Six TRENDS IN PRIVATE EQUITY

22. Private Investment in Public Equity 397


na dai

23. Listed Private Equity 419


douglas cumming, grant fleming, and sofia a. johan

24. Private Equity Growth in International and Emerging Markets 441


alexander peter groh

25. Diversification Benefits of Private Equity Funds-of-Funds 463


axel buchner and markus kuffner

26. Publicly Traded Private Equity 483


henry lahr

27. The Future of Private Equity: A Global Perspective 503


dianna c. preece

Discussion Questions and Answers 523


Index 561
List of Figures

2.1 Number of Closed or Effective Leveraged Buyouts or Management Buyouts,


1970 to 2013 18
2.2 Firm Value 19
2.3 Poorly Performing Publicly Listed Companies in Different Financial
Markets, January 2014 20
2.4 Mechanics of a Leveraged Buyout 21
2.5 Typical Structure of a Private Equity Fund 22
4.1 Fund Return and the J-Curve Effect 55
5.1 Fundraising Levels in 2013 by Fund Type 68
5.2 Median Leveraged Buyout Debt Levels and EBITDA Multiples over Time
in the United States 72
5.3 Total Volume and Average Tranche Size of Syndicated Loans used in
Leveraged Buyouts 73
5.4 The Value Creation Principle in Leveraged Buyouts 75
5.5 Choice of Exit Channels by Leveraged Buyouts, 1990 to 2013 79
5.6 Cash Multiples of Leveraged Buyouts by Exit Channel, 1990 to 2013 79
7.1 Size of the Early High Yield Market 102
7.2 Median Distressed Fund Return 103
7.3 Distressed Private Equity Funds Raised in North America 111
7.4 Merger and Acquisition Transaction Value, 1995 to 2009 112
7.5 Low Rated High Yield Issuance, 1995 to 2009 112
7.6 Acquisition Debt Multiples 113
7.7 Leveraged Finance Default Rates 113
7.8 European Leveraged Buyout Volume 114
7.9 Fundraising for European Distressed-Asset Private Equity 115
9.1 Beta: GRO Index vs. the S&P 500 Index 152
9.2 Beta: MEZ Index vs. the S&P 500 Index 153
9.3 Beta: Buyouts vs. the S&P 500 Index 153
9.4 Beta: Private Equity Index vs. the S&P 500 Index 154
10.1 Annual Median Net Internal Rates of Return by Vintage 170
10.2 Historical Private Equity Fundraising, 2008 to 2013 171
10.3 All Private Equity Annual Amount Called Up and Distributed,
2003 to 2013 172

viii
l i s t o f f i g u r e s ix

11.1 The Risk Profile of Venture Capital: Direct, Funds, and Fund-of-Funds 193
11.2 Average Fund Performance per Vintage Year, 1990 to 2000 193
12.1 Number of Financial Buyer Exits via Public Offerings in Different
Geographic Regions and Industries 201
12.2 Number of Venture-Capital-backed Initial Public Offerings in the United
States, 1980 to 2013 202
12.3 Comparison of the Short- and Long-Term Returns of Venture-Capital-
Backed and Non-Venture-Capital-Backed Initial Public Offerings in the
United States, 1980 to 2013 202
13.1 Number of Global Buyout Exits, 2009 to 2013 216
13.2 Number of Global Buyout-Backed Exits by Region, 1995 to 2013 216
13.3 Number of Global Buyout-Backed Exits by Channel, 1995 to 2013 219
13.4 Private Equity-Backed Exits by Type, 2006 to 2013 219
13.5 Volume and Aggregate Value of PE-Backed IPOs: January 2005 to June
2014 225
13.6 PE-Backed IPOs (Volume) and Market Conditions: January 2005 to June
2014 225
13.7 Breakdown of the Aggregate Value of PE-Backed IPOs by the Region of
Portfolio Company: January 2005 to June 2014 227
13.8 Breakdown of the Yearly Deal Value at the Completed Date per Industry
Sector 229
14.1 Graph of the Top Three and Bottom Three Countries for PE
Attractiveness 245
14.2 Global Private Equity Buyouts as a Proportion of Their 2006 Value 247
14.3 Proportion of Buyout Deals by Geographic Region, 2006 to 2013 247
14.4 Global Buyout Deals by Geographic Regions, 2006 vs. 2013 248
14.5 Correlation Coefficient for Global Listed Private Equity Compared with
Other Assets 252
14.6 Correlation Coefficient for PowerShares Global Listed Private Equity
Compared with Other Assets 254
16.1 Top-to-Bottom Quartile Private Equity Internal Rate of Return Ranges by
Vintage Year 277
16.2 Transition Matrix: Probability of Transition from One Quarter to
Another 282
17.1 Types of Due Diligence 296
17.2 Summary of the Process of Due Diligence 304
18.1 Private Equity Deal Flows, 2004 to 2013 312
18.2 Model Investment Structure of Private Equity 315
19.1 An Overview of the Key Data Statistics of the Sample 336
20.1 Illustration of Profit-Sharing between Limited Partners and General
Partners 365
22.1 Size of the Private Investment in Public Equity Market 399
22.2 The Distribution of Private Investments in Public Equity by Industry 400
22.3 Security Structure of Private Investments in Public Equity 402
22.4 Investors in the Private Investment in Public Equity Market 404
22.5 Alternative Financings for Small Firms: Unseasoned Issuers 414
x l i s t o f f i g u r e s

23.1 Post Initial Public Offering Performance of Blackstone and Fortress, 2007
to 2014 421
23.2 Worldwide Distribution of Listed Private Equity Companies, December
2013 421
23.3 Listed Private Equity Market Capitalization, 1993 to 2013 422
23.4 Listed Private Equity Absolute Return Performance, 2004 to 2014 422
24.1 Global Private Equity Investment Volume 442
24.2 Private Equity Investment Volume Differentiated by Region 443
24.3 Private Equity Investment Volume in Brazil, Russia, India, and China 444
24.4 Country Private Equity Attractiveness Score: First Half of Sample
Countries 449
24.5 Country Private Equity Attractiveness Score: Second Half of Sample
Countries 450
24.6 Attractiveness Rank Changes 451
24.7 Key Drivers of Brazil, Russia, India, and Chinas Private Equity
Attractiveness 453
24.8 More Highly Granulated Determinants of Brazil, Russia, India, and Chinas
Private Equity Attractiveness 454
24.9 Key Drivers of Private Equity Attractiveness for Indonesia, Mexico, the
Philippines, and Turkey 455
24.10 More Highly Granulated Determinants of Private Equity Attractiveness for
Indonesia, Mexico, the Philippines, and Turkey 456
24.11 Attractiveness Scores and Private Equity Activity of 118 Countries 457
24.12 Country Maturity Scores and Historic Private Equity Returns 459
25.1 Kernel Density Estimates for Liquidated Funds and Funds-of-Funds 473
25.2 Comparing Kernel Density Estimates for Both Samples 475
25.3 Funds per Vintage Year 475
26.1 Listings and Market Capitalization of Publicly Traded Private Equity 485
26.2 Organizational Forms of Listed Private Equity 486
26.3 Legal Structure of a Publicly Traded Private Equity Fund 487
26.4 Legal Structure of a Private Traded Public Equity Fund-of-Funds 488
26.5 Legal Structure of a Publicly Traded Management Company 489
26.6 Legal Structure of a Publicly Traded Investment Company 489
26.7 Net Asset Value Premium in Calendar Time 492
26.8 Premium in Event Time, Years from Initial Public Offer 496
26.9 Premium in Event Time, Years from Initial Public Offering by Fund Type 497
List of Tables

2.1 Top 10 Private Equity Buyouts Ranked by Inflation-Adjusted Deal Value


in 2014 Dollars 17
2.2 10 Largest Private Equity Funds 23
5.1 Compensation Structures in Leveraged Buyouts, 1990 to 2013 69
5.2 Restructuring Strategies in Leveraged Buyouts 78
6.1 Comparison of a Mezzanine Loan and Preferred Equity 92
7.1 TargetCo Debt Capital Structure 105
8.1 Components of the Four Major Cash Flow Definitions 126
8.2 Pros and Cons of Discounted Cash Flow, Adjusted Present Value, and
Multiples-Based Valuation Methods 134
8.3 Able International: Balance Sheets for Years Ended December 31 135
8.4 Able International: Discounted Cash Flow Assumptions Based on Current
Structure and Management 136
8.5 Able International: Discounted Cash Flow Assumptions with Private Equity
Ownership 138
8.6 Able International: Multiples-Based Valuation 141
8.7 Able International: Adjusted Present Value Valuation Assuming Private
Equity Ownership 143
9.1 Descriptive Statistics: January 2000 to December 2012, Using Monthly
Data 152
9.2 Growth, Venture Capital, and a Few Buyouts: GRO 157
9.3 Buyouts 159
9.4 Mezzanine: MEZ 160
9.5 Private Equity Index 161
9.6 Illiquidity and the Cost of Equity of Private Equity: GLM Adjustment 163
11.1 Distinction between Privately Held and Private Equity 182
13.1 Private-Equity-Backed IPOs and Stock Market Evolution, 2005 to
June 2014 224
13.2 Breakdown of the Aggregate Deal Value of Private-Equity-Backed IPOs by
Region of the Portfolio Companies, 2005 to June 2014 226
13.3 Breakdown of the Yearly Aggregate Deal Value at the Completed Date per
Portfolio Company by Industrial Sector, 2005 to June 2014 228
13.4 Exit Routes Overview 233

xi
xii l i s t o f ta b l e s

14.1 The 10 Most Attractive Countries for Private Equity, 2010 to 2014 244
14.2 Brazil, Russia, India, and China Country Attractiveness Index Values,
2010 to 2014 245
14.3 Aggregate Value of Private Equity-Backed Buyout Deals by Region,
2006 to 2013 246
14.4 Annual Returns by Year for Various Assets, 2008 to 2013 251
14.5 Inverse Coefficient of Variation, 2008 to 2013 252
14.6 Global Listed Private Equity Correlation Coefficients, 2008 to 2013 253
14.7 PowerShares Global Listed Private Equity Correlation Coefficients,
2008 to 2013 253
16.1 Empirical Evidence on Private Equity Return Persistence 281
18.1 Comparing Private and Public Equity 313
18.2 Financial Assets of Institutional Investors by Country 323
18.3 Largest Retirement Funds 324
19.1 Data Panel and Descriptive Key Statistics 338
19.2 Data Panel Key Statistics 341
19.3 Impact of the Main Value Creation Levers on the Performance of Leveraged
Buyout Transactions by Entry Year and Leveraged Buyout Transaction
Type 347
19.4 EV/EBITDA Multiples in Leveraged Buyout Deals: The Impact of Different
Value Creation Levers by Entry Year and Transaction Type 349
19.5 Decomposing Internal Rate of Return in Leveraged Buyout Transactions by
Entry Year 350
19.6 Analysis of the Impact of Leveraged Buyout Ownership on Operating
Performance 353
19.7 Expected versus Actual Performance of Leveraged Buyout Deals by Entry
Year 354
20.1 Summary Statistics of Fee Terms 370
22.1 Size of the Private Investment in Public Equity Market, 1995 to 2012 399
22.2 Industry Distribution of Private Investments in Public Equity 400
22.3 Security Structure of Private Investments in Public Equity,
1995 to 2012 401
22.4 Private Investments in Public Equity in the Financial Sector,
1995 to 2012 403
22.5 Total Amount in USD Billion Invested by Investor Type, 1995 to 2012 404
22.6 Ranking of Placement Agents by Market Share 410
22.7 Alternative Financings for Small Firms, 2011 to 2012 414
22.8 The Emergence of at-the-Market Offering, 2009 to 2012 415
22.9 The Emergence of Confidentially Marketed Public Offerings,
2009 to 2012 416
23.1 Definitions and Summary Statistics 428
23.2 Comparisons of Descriptive Statistics between Listed Private Equity as a Part
and not a Part of Investment Mandate 433
25.1 Calculation of Fees 469
25.2 Multiples 470
25.3 Sample 1: Liquidated Funds 472
l i s t o f ta b l e s xiii

25.4 Sample 2: Mature Funds 474


25.5 Liquidated Funds and Funds-of-Funds 476
25.6 Mature Funds and Funds-of-Funds 477
25.7 Certainty Equivalents 478
25.8 Levels of Indifference 478
25.9 Selection Ability 480
25.10 Marginal Coefficients of Risk Aversion 480
Acknowledgments

Editing is like walking across a room strewn with rose petals and
thorns. When you can walk across mostly unbloodied, youre finished.
Richard Due

As the editors of Private Equity: Opportunities and Risks, we have many people to thank
for their involvement in this book. Although the list is long, we have edited it to five:
the chapter authors, professional staff at Oxford University Press, the indexer, our aca-
demic institutions, and families. The chapter authors deserve our deep appreciation for
their excellent work and for enduring countless rounds of edits with few complaints.
In the end, we could eventually walk across the room unbloodied. The team at Oxford
University Press including but certainly not limited to Scott Parris (Editor), Cathryn
Vaulman (Assistant Editor), Cherline Daniel (Sr. Project Manager) merit special thanks
as does Claudie Peterfreund (Indexer). We also appreciate the support provided by our
respective institutionsthe Kogod School of Business at American University, the
Behrend College at Penn State Erie, and the Crummer Graduate School of Business at
Rollins College. Finally, our families have graciously allowed us to devote much of our
time to this book project instead of to them. The authors dedicate this book to their
families: Linda and Rory Baker; Janis, Aaron, Kyle, and Grant Filbeck; and Nilgun and
Tunc Kiymaz.

xv
About the Co-Editors

H. Kent Baker is a University Professor of Finance in the Kogod School of Business


at American University. Professor Baker is an author or editor of 24 books including
Investment Risk Management, Investor Behavior: The Psychology of Financial Planning
and Investing, Market Microstructure of Emerging and Developed Markets, Behavioral
Finance: Investors, Corporations, and Markets, Portfolio Theory and Management, and
Survey Research in Corporate Finance. As one of the most prolific finance academics,
he has published more than 160 articles in such journals as the Journal of Finance,
Journal of Financial and Quantitative Analysis, Financial Management, and Financial
Analysts Journal. He serves on the editorial boards of six journals and has consulting
and training experience with more than 100 organizations. Professor Baker holds a
BSBA from Georgetown University; MEd, MBA, and DBA degrees from the Uni-
versity of Maryland; and an MA, MS, and two PhDs from American University. He
also holds CFA and CMA designations.
Greg Filbeck holds the Samuel P. Black III Professor of Finance and Risk Manage-
ment at Penn State Erie, the Behrend College and serves as the Associate Director
for the Black School of Business and the Department Chair for Finance & Econom-
ics. He formerly served as Senior Vice-President of Kaplan Schweser and held aca-
demic appointments at Miami University and the University of Toledo, where he
served as the Associate Director of the Center for Family Business. Professor Fil-
beck is an author or editor of five books and has published more than 75 refereed
academic journal articles in the Financial Analysts Journal, Financial Review, and Jour-
nal of Business, Finance, and Accounting among others. Professor Filbeck conducts
training worldwide for candidates for the CFA, FRM, and CAIA designations.
Professor Filbeck holds a BS from Murray State University, an MS from Penn State
University, and a DBA from the University of Kentucky.
Halil Kiymaz is Bank of America Professor of Finance in the Crummer Graduate
School of Business at Rollins College. He held positions at Bilkent University,
University of Houston-Clear Lake, ECUST, Copenhagen Business School, Ada
University, and Kadir Has University. Professor Kiymaz has published more than
70 articles in scholarly and practitioner journals and co-edited three books. His re-
search has appeared in the Journal of Banking and Finance, Financial Review, Global
Finance Journal, Journal of Applied Finance, Journal of Economics and Finance, and

xvi
a b o u t t h e c o - e d i t o r s xvii

Review of Financial Economics among others. Professor Kiymaz also serves on the
editorial board of four journals and is the area editor of the International Journal of
Emerging Markets. He has consulting and training experience with various organiza-
tions. Professor Kiymaz received a BS from the Uluda University and an MBA,
MA, and PhD from the University of New Orleans.
About the Contributors

Parvez Ahmed is an Associate Professor of Finance at the Coggin College of Busi-


ness, University of North Florida (UNF) and serves as the Director of the Center
for Sustainable Business Practices. He was named a U.S. Fulbright Scholar in 2009.
Professor Ahmed previously taught at Penn State UniversityHarrisburg and the
University of North Carolina at Charlotte where he was the founding faculty advi-
sor for the student-managed investment fund. He has published in such journal as
the Journal of Portfolio Management, Financial Management, Journal of Banking and
Finance, Journal of Investing, Journal of Wealth Management, and Financial Review. He
recently published Mutual Funds: Fifty Years of Research Findings. Professor Ahmed
holds an MBA from Temple University and a PhD in finance from the University of
Texas at Arlington.
Emmanuel Boutron is an Associate Professor of Finance at the Management
School and member of the research center CEROS at University of Paris Ouest
Nanterre La Dfense. He is also the head of the Master Business and Manage-
ment and the General Management Program in Continuing Education (IFG
Laureate Group). Professor Boutron specializes in corporate finance and his
primary research field focuses on IPOs. He recently examined SMEs (small and
medium enterprises) financing through an applied research program funded by
LObservatoire du financement des petites et moyennes entreprises par le march. Pro-
fessor Boutron holds a PhD in finance from the University of Paris Ouest Nan-
terre La Dfense.
Eric Braune is an Assistant Professor in Management and Strategy at INSEEC in
Lyon. He served previously as the Regional Managing Director of a large French
group and he is still committed to business incubators and Science Park in Lyon
(France). His research mainly deals with innovation management, organizational
theory, and corporate governance. He received a Master of Research in Economics
Philosophy from Aix-en-Provence University and a PhD in management from IAE
Aix-en-Provence.
Axel Buchner is an Assistant Professor of Finance at the DekaBank Chair in Finance
and Financial Control at Passau University, Germany. Before joining Passau Uni-
versity, he was a post-doctoral researcher and lecturer at Technical University of
Munich and a consultant at the Center of Private Equity Research (CEPRES). At
CEPRES, he specialized in private equity and equity-related consulting projects for

xviii
a b o u t t h e c o n t r i b u t o r s xix

national and European organizations such as the German Federal Ministry of Ec-
onomics and Innovation and the European Investment Bank (EIB), in addition to
large institutional investors. His research focuses on issues in private equity, venture
capital, and asset pricing. His teaching interests are in derivatives, empirical finance,
portfolio theory, and asset pricing. He holds a masters degree in business adminis-
tration from Munich University and a PhD in finance from Technical University of
Munich.
Ji-Woong Chung is an Assistant Professor in Finance at Korea University Business
School. His research includes corporate finance and financial markets with a partic-
ular emphasis on the area of private equity, hedge funds, and international capital
flows. Before joining Korea University, he was an Assistant Professor at the Chinese
University of Hong Kong and a Research Fellow at Institute of Economics and Fi-
nance. His research won the Wharton School WRDS Award for Best Empirical
Paper in 2011 and Best PhD Paper Award from the Coller Institute of Private Equity
at London Business School in 2010. He received a BA in economics and applied
statistics from Yonsei University, Korea, and holds a PhD in finance from Fisher
College of Business at The Ohio State University.
Alain Con is a Professor of Finance at the Graduate School of Business (ESG) of
the University of Quebec in Montreal (UQAM) and an associate researcher of the
Ivanho Cambridge Real Estate Chair at ESG-UQM Graduate School of Business.
He previously was an Associate Professor of Finance at EDHEC School of Man-
agement. Professor Con has also been a Visiting Professor at Paris-Dauphine Uni-
versity, University of Paris-Ouest-Nanterre, EDHEC School of Management, Laval
University, HECUniversity of Lige, and University of Sherbrooke. His research
focuses on asset pricing, international finance, hedge funds, REITs, business cycles,
and financial econometrics. He has published in the Journal of Empirical Finance,
Journal of Financial Research, Economics Letters, Finance Research Letters, Journal of
Economics and Business, Journal of Multinational Financial Management, Finance, and
Journal of Alternative Investments. He has also written a book on financial manage-
ment. He holds an MA in economics from Laval University and an accreditation to
supervise research (HDR) from ParisDauphine University. He obtained a PhD in
finance from the University of Grenoble and a PhD in economics from the Univer-
sity of Paris I Panthon-Sorbonne.
Douglas Cumming, CFA is a Professor of Finance and Entrepreneurship and the
Ontario Research Chair at the Schulich School of Business, York University. He is
a co-editor of Entrepreneurship Theory and Practice and has been a guest editor for
12 special issues of top journals. Professor Cumming has published more than 110
articles in leading refereed academic journals in finance, management, and law and
economics, such as the Journal of Financial Economics, Review of Financial Studies,
Journal of International Business Studies, and the Journal of Empirical Legal Studies. He
is the coauthor of Venture Capital and Private Equity Contracting (Elsevier Academic
Press, 2nd Edition, 2013), and Hedge Fund Structure, Regulation and Performance
around the World (Oxford University Press, 2013). His work has been reviewed in
numerous media outlets including The Economist, The New York Times, Canadian
Business, the National Post, and The New Yorker. He holds a JD and a PhD in finance
from the University of Toronto.
xx a b o u t t h e c o n t r i b u t o r s

Na Dai is an Associate Professor of Finance at the School of Business at SUNYAlbany.


Her research primarily focuses on venture capital, private equity, hedge funds, pri-
vate investment in public equity, and small business financing. She has published in
Financial Management, Journal of Banking and Finance, Journal of Corporate Finance,
Journal of Empirical Finance, European Financial Management, Journal of Business Ven-
turing, and Entrepreneurship Theory and Practice. She is the co-author of Hedge Fund
Structure, Regulation and Performance around the World, published by the Oxford
University Press. She has a BA (Honors) and an MA from Ren Min University of
China and a PhD from the University of Kansas.
Aurlie Desfleurs is an Associate Professor in the Accounting Department at the Uni-
versity of Sherbrooke in Canada. She previously was an Assistant Professor at the School
of Accounting of Laval University. Her research focuses on financial analysts forecasts,
M&As, and international financial reporting statements. Professor Desfleurs has pub-
lished in the Journal of Economics and Business, Journal of Multinational Financial Man-
agement, Journal of Derivatives and Hedge Funds, and Bankers, Markets and Investors. She
graduated from EDHEC School of Management and obtained an MBA and a PhD in fi-
nance from Laval University. She is also a Chartered Professional Accountant in Canada.
Shantanu Dutta is an Associate Professor of Finance at the Telfer School of Manage-
ment, University of Ottawa. Before his career in academe, he served as a finance man-
ager at Lafarge, a world leader in construction materials. Professor Duttas research
focuses on M&As, media coverage and financial decisions, corporate governance,
and dividend policy. He has published in such journals as Financial Management,
Journal of Banking and Finance, Journal of Business Finance & Accounting, Journal of
Multinational Financial Management, Multinational Finance Journal, and Corporate
GovernanceAn International Review. He is a recipient of SSHRC grant and Bar-
clays Global Investors Canada Research Award (2006) for the best paper on the
Canadian security market. He has also received the Literati Network Awards for Ex-
cellence 2009 and 2014 for his published articles in the International Journal of Man-
agerial Finance. Professor Dutta received a PhD from Carleton University.
Grant Fleming is a Founding Partner of Continuity Capital Partners and has worked
in private capital markets in Asia for more than 10 years, with particular focus on
private equity and credit and distressed opportunities. Before founding Continuity
Capital Partners, he spent 10 years at Wilshire Associates Incorporated, researching
and building Asian private equity and distressed debt portfolios for a global investor
base located in Australia, Japan, the United States, and Europe. He was a member
of the firms global investment committee where he oversaw investments of more
than $6 billion in private equity, distressed debt, and special situations. He has held
academic positions at the University of Auckland and the Australian National Uni-
versity. He has also published in such academic journals as Business History, Business
History Review, Cambridge Journal of Economics, Entrepreneurship Theory and Prac-
tice, Journal of Banking and Finance, Journal of Corporate Finance, and Journal of Mon-
etary Economics. He earned his BA, MA (Honors), and PhD in economics from the
University of Auckland.
Didier Folus is a Professor of Finance at the Management School, University of Paris
Ouest Nanterre La Dfense. His primary research interests are in long-term invest-
ment exposure to risks and insurance securitization in a corporate risk management
a b o u t t h e c o n t r i b u t o r s xxi

perspective. He is the head of the research center CEROS at the University of Paris
Ouest Nanterre La Dfense, and the head of the Master Program in Banking, Fi-
nance, and Insurance at the same university (Paris and Luxemburg). Professor
Folus teaches finance courses at Paris Dauphine University and is an Associate
Professor at Institut de Formation de la Profession de lAssurance (IFPASSFrench
Insurance Companies Continuing Education School). He advises institutional in-
vestors, insurance companies, and banks on innovative financial products. He holds
a masters degree from the Ecole Nationale de la Statistique et de lAdministration
Economique (Paris) and a masters degree and a PhD in finance from the University
of Paris Dauphine.
Arup Ganguly is a doctoral student in finance at the University of Pittsburgh hold-
ing the Braskem America, Inc. Fellowship at the Joseph M. Katz Graduate School
of Business. Before joining the University of Pittsburgh, Mr. Ganguly worked as an
Associate at Scotia Capital, an investment banking and corporate lending wing of
Scotia Bank in Canada. During the same time, as a part-time instructor at the Joseph
L. Rotman School of Management, University of Toronto, he received several teach-
ing awards at both the undergraduate and MBA levels. His research interests are
primarily in empirical corporate finance, specifically in the areas of private equity,
capital structure, M&As, and cash holdings. He is an ardent fan of the Pittsburgh
Penguins. Mr. Ganguly received an M.Fin from the University of Toronto where he
graduated as the class valedictorian.
Lin Ge is a PhD student in finance at the University of Pittsburgh where she holds the
ELG Metals Fellowship at the Joseph M. Katz Graduate School of Business. Before
commencing her doctoral studies at Katz, Ms. Ge worked in the hedge fund indus-
try in Toronto. Her career objective is to explore the yet unresolved questions in cor-
porate finance and contribute to the field of financial economics. Ms. Ges research
interests are in M&As, corporate governance, venture capital, and corporate finan-
cial policies. In her spare time, she enjoys learning traditional Chinese painting and
reading autobiographies. Ms. Ge has an undergraduate degree in economics and an
MBA specializing in finance.
Alexander Peter Groh is Professor of Finance and Director of the Entrepreneurial
Finance Research Centre at EMLYON Business School, France. He has held visit-
ing positions at the University of New South Wales, Sydney, Australia, IESE Busi-
ness School, Barcelona, Spain, and INSEAD, Fontainebleau, France. His research
focuses on venture capital and private equity, and includes valuation issues, perfor-
mance measurement, and socioeconomic determinants for the development of vi-
brant venture capital and priate equity markets. He has published in such journals
as the Journal of Banking and Finance, Journal of Corporate Finance, Journal of Inter-
national Money and Finance, Corporate Governance: An International Review, Euro-
pean Financial Management Journal, Journal of Alternative Investments, Journal of Real
Estate Finance and Economics, Emerging Markets Review, and Venture Capital. He has
been involved in management training courses for the European Venture Capital
and Private Equity Association (EVCA) and has worked for Quadriga Capital, a
Frankfurt-based private equity fund, since 1996. Professor Groh earned a PhD from
Darmstadt University of Technology in Germany, where he also studied Mechanical
Engineering and Business Administration.
xxii a b o u t t h e c o n t r i b u t o r s

Pter Harbula is Executive Director of Corporate Finance at Edenred Group, a listed


group on the Paris Stock Exchange. Besides his professional work for Edenred, he
is a lecturer at HEC Paris business school. Dr. Harbula is also an Accredited Senior
Appraiser with the American Society of Appraisers, member of the French Society
of Valuation Experts (SFEV) and the French Association of Merger & Acquisition
Professionals (ARFA). He has authored various articles and chapters in books on
corporate governance, valuation, value creation, and private equity in French, Hun-
garian, and English. Dr. Harbula received an MS in business administration special-
ized in finance from the Budapest University of Economic Sciences and a PhD in
business administration from the Paris II Assas University.
J. Dean Heller is senior real estate partner in the Los Angeles office of Seyfarth Shaw
LLP, an international law firm. He has extensive experience since the 1980s in repre-
senting mezzanine lenders and preferred equity investors in all types of commercial
properties including hotels. Mr. Heller is the author of Whats in a Name: Mezza-
nine Debt Versus Preferred Equity, published in the Stanford Journal of Business &
Finance, as well as numerous articles and presentations on various aspects of real
estate financing. He received a BA, magna cum laude, from Duke University and a
JD, cum laude, from Yale University.
Sofia A. Johan is the Extramural Research Fellow at the Tilburg Law and Economics
Centre (TILEC) in The Netherlands and Adjunct Professor of Law and Finance at
the Schulich School of Business, York University. Her research primarily focuses on
law and finance, market surveillance, hedge funds, venture capital, private equity,
and IPOs. She has published in such journals as the Journal of Financial Economics,
Journal of International Business Studies, American Law and Economics Review, Jour-
nal of Banking and Finance, European Economic Review, and Entrepreneurship Theory
and Practice. She is the coauthor of Venture Capital and Private Equity Contracting:
An International Perspective and Hedge Fund Structure, Regulation and Performance
around the World. She earned her LLB from the University of Liverpool, an LLM in
international economic law from the University of Warwick, and a PhD in law and
economics from the University of Tilburg.
Markus Kuffner is a student in the MBA program at Passau University, Germany, with
a focus on finance and financial controlling as well as financial engineering and asset
pricing. From 2011 to 2013, he was a lecturer on banking and finance for the home
loan bank Schwbisch-Hall and a lecturer on financial sales for Wstenrot & Wrt-
tembergische. After completing his studies, he will join the accounting department for
foreign branches and subsidiaries at the Landesbank Baden-Wrttemberg where he
will support the implementation of the Financial Reporting Framework (FINREP).
Henry Lahr is a lecturer in finance at The Open University Business School. Before
joining The Open University, he worked as research fellow at the University of Cam-
bridges Centre for Business Research (CBR). He has been involved in several proj-
ects concerned with financing activities of innovative firms and the consequences
of venture capital and private equity financing in small and large enterprises. The
European Commissions FP7 program, the U.K. Innovation Research Centre as well
as the German Federal Ministry of Finance and the German State Media Authorities
funded his research at the CBR and Technical University Munich. He has published
on firms financial constraints to investment and innovation, innovation dynamics in
a b o u t t h e c o n t r i b u t o r s xxiii

entrepreneurial firms, financing of research and development, listed private equity,


venture capital, and private equity fund pricing. He holds an undergraduate degree
in business economics from Witten/Herdecke University and a PhD in finance from
Technical University Munich.
John D. Martin is the Carr P. Collins Chair of Finance in the Hankamer School of
Business at Baylor University where he has won numerous awards for both teaching
and research. Before moving to Baylor he was the Margaret and Eugene McDermott
Professor of Finance at the University of Texas at Austin. His research interests in fi-
nance include valuation, organization theory, and investment management. He has
authored numerous textbooks in finance including Valuation: The Art and Science of
Corporate Investment Decisions published by Prentice Hall, and more than 60 schol-
arly papers published in both academic and professional journals. He holds an MBA
from Louisiana Tech University and a PhD from Texas Tech University.
Thomas Meyer is a co-founder of LDS Partners and a Shimomura Fellow of the De-
velopment Bank of Japans Research Institute of Capital Formation. He is a member
of the EVCA Private Equity Risk Working Group, Risk Management for Alternative
Investment Funds Working Group of ALFI in Luxembourg, and Chartered Alter-
native Investment Analyst Associations (CAIA) private equity sub-committee.
Dr. Meyer co-directed the limited partner course delivered by the Private Equity
Institute at the Sad Business School, University of Oxford, that led to the EVCA
awarded certificate. He has authored and co-authored several books including
Beyond the J Curve: Managing a Portfolio of Venture Capital and Private Equity Funds,
J-Curve Exposure: Managing a Portfolio of Venture Capital and Private Equity Funds,
Mastering Illiquidity: Risk Management for Portfolios of Limited Partnership Funds, and
Private Equity Unchained: Strategic Insights for the Institutional Investor. He received
an MBA from the London Business School and holds a Dr. rer. nat. in computer sci-
ence from the University of Trier.
Stephen G. Moyer is the President of Distressed Debt Alpha, a consulting firm that
focuses on distressed debt investing. He is also an Adjunct Professor at the Univer-
sity of Southern California. Before forming his consulting firm, he spent 30 years in
investment management and banking with notable firms such as PIMCO, Tennen-
baum Capital, Imperial Capital, Drexel Burnham, and First Boston. He is the author
of Distressed Debt Analysis and other publications. He holds a JD from Stanford Law
School and an MBA from the University of Chicago.
Adrian Oberli is administrative director of the BA in business administration pro-
gram at University of St. Gallen (HSG), Switzerland, and a research fellow at Har-
vard Business School. His research focuses on private equity and venture capital.
Before joining HSG, Mr. Oberli worked in alternative investments for several years
in London and Munich. In 2012, he founded the Swiss Private Equity Confer-
ence, an annual event held in Zurich, gathering distinguished leaders in PE and VC
(http://www.pe-conference.com/swiss/). Mr. Oberli received an MA in banking
and finance from University of St. Gallen and an MSc in strategic management from
Rotterdam School of Management. He is a CFA charterholder and certified Finan-
cial Risk Manager (FRM).
Henry Ordower is Professor of Law at Saint Louis University School of Law. Be-
sides research on and teaching of corporate finance in addition to both U.S. and
xxiv a b o u t t h e c o n t r i b u t o r s

comparative taxation, Professor Ordower maintains an active consulting practice in


taxation, hedge and private equity funds, and business organizations. Previously of
counsel to major St. Louis law firms, he also served as vice president and general
counsel to an emerging markets hedge and private equity fund management com-
pany based in New York. Professor Ordower has an extensive background in Euro-
pean languages and has lectured and participated in international legal conferences
in Europe, Asia, and Africa as well as the United States. He has been elected to mem-
bership in the American College of Tax Counsel, the European Association of Tax
Law Professors, and the International Academy of Comparative Law. He holds an
AB from Washington University and an MA and JD from the University of Chicago.
Harlan Platt is a Professor of Finance at the DAmore-McKim School of Business
at Northeastern University. His current research interests include valuation, bank-
ruptcy and restructuring, crisis management, and hedge funds. He created the certi-
fication examination of the Turnaround Management Association and administered
it for 10 years. Professor Platt has published in many academic and practitioner jour-
nals including the Journal of Banking and Finance, Journal of Business Finance and Ac-
counting, Journal of Alternative Investments, Journal of Bankruptcy Law, and Journal of
Investing. He has also written 10 books including the bestseller The Phoenix Effect.
Professor Platt serves on several corporate boards of directors including Republic
Financial Corporation, Millennium Chemicals, and is the Chairman of the John
Hancock Collateral Investment Trust. He obtained a BA from Northwestern and a
PhD from the University of Michigan.
Esha Prashar holds a teaching position at Chitkara Business School, Chitkara Uni-
versity, India, where she teaches M&As, investment analysis, portfolio management,
and financial derivatives. Her research involves asset pricing and valuation with a
special focus on private equity as well as private equity indices and their relationship
with market indices. Ms. Prashar has published in the Journal of Private Equity and
Journal of Wealth Management. She obtained her MBA degree from Panjab University.
Dianna C. Preece is a Professor of Finance at the University of Louisville where she
has taught for more than 20 years. She teaches corporate finance, investments, and
financial markets and institutions at the undergraduate level and MBA courses in
corporate finance and investments. Professor Preece also teaches in several banking
schools including the Kentucky School of Banking and the Iowa School of Banking.
She has published in such journals as the Journal of Banking and Finance and Journal
of Business Finance and Accounting. Professor Preece received her DBA from the Uni-
versity of Kentucky and also holds the CFA designation.
Christian Rauch is an Assistant Professor in the finance department of Goethe Uni-
versity Frankfurt, Germany. Besides his work at Goethe University, he holds po-
sitions as guest lecturer at the Frankfurt School of Finance and Management and
several associate research positions. He researches and teaches in the fields of bank-
ing regulation and corporate finance, especially alternative investments. His work
has been presented at international conferences and published in such journals as
the Journal of Banking and Finance, European Journal of Finance, and Journal of Pri-
vate Equity. He graduated in business administration and corporate law from Euro-
pean Business School and New York University, and earned a PhD in finance from
Goethe University Frankfurt.
a b o u t t h e c o n t r i b u t o r s xxv

Jean-Michel Sahut is a Professor at IPAG Paris where he teaches corporate finance,


business plans, evaluation of firms, investments, and entrepreneurship for engi-
neering and management students. Previously, he was a Professor at the Geneva
School of Business Administration, University of Applied Sciences. He was Asso-
ciate Dean for Research at Amiens School of Management, Professor of Finance
at Telecom and Management at Paris Sud and Director of the RESFIN Labora-
tory. Professor Sahut has much international experience in executive education and
consulting in Europe. Professor Sahut is also an expert for the 7th EU Framework
Program. He has published more than 60 articles about finance, venture capital, en-
trepreneurship, and corporate governance in international peer review journals and
five books. He received a BA in econometrics from the University of Paris 1, Pan-
thon Sorbonne, an MA in management, and a PhD in finance from Aix-Marseille
University.
Elif Akben Selcuk is an Instructor in the Business Administration Department of the
Faculty of Economics, Administrative and Social Sciences at Kadir Has University,
Turkey. She teaches courses on financial and managerial accounting. Before joining
Kadir Has University, she worked as a research and teaching assistant at Bogazici
University and Yeditepe University. Her research has been published in such journals
as the Business and Economics Journal, International Business Research, and Journal of
American Academy of Business, Economics Bulletin. Professor Selcuks current research
interests include financial literacy, corporate finance, merger and acquisitions, corpo-
rate governance, and emerging markets. She obtained a BA in business administra-
tion, an MA in economics, and a PhD in finance from Bogazici University.
Alexandros Seretakis is a PhD candidate and teaching and research assistant at the
University of Luxembourg where he is completing his doctoral thesis on hedge fund
regulation and teaching courses in the field of European company law. His main re-
search interests involve leveraged buyouts, hedge funds, and corporate law and gov-
ernance. He has published in the New York University Journal of Law and Business,
Fordham Journal of Corporate and Financial Law, Brooklyn Journal of Corporate, Fi-
nancial and Commercial Law, and Corporate Finance and Capital Markets Law Review.
Mr. Seretakis served as a research fellow at New York Universitys Pollack Center for
Law and Business, a joint venture of the School of Law and Stern School of Business.
He is admitted to practice law in Greece and New York. He has a BA from Aristotle
University of Thessaloniki, an LLM in banking and finance from the University Col-
lege London, and an LLM in corporate law from New York University.
Manu Sharma is a Professor of Finance at Panjab University in India. Before joining
academia, he worked in investment banking, venture capital, and debt capital mar-
kets. His experience encompasses M&A advisory and private equity valuations, ven-
ture development and business plan writing, credit analysis, lease syndication, and
model building. His research focuses on private equity. Professor Sharma analyzes
the market variants that contribute to the overall growth of private equity indexes
as well as investment, risk analysis, and valuation. He has published in the Journal of
Private Equity, Journal of Wealth Management, and Journal of Index Investing.
Kenneth Small is the Marshall Butler Professor of Finance an Assistant Dean of
Students in the Craig E. Wall College of Business at Coastal Carolina University. He
teaches wealth management and has developed and taught various courses at the
xxvi a b o u t t h e c o n t r i b u t o r s

graduate and undergraduate levels. Professor Small has won several teaching,
advising, and service awards at both the university and college levels including a
Governors Distinguished Professor in 2012. He has published in such journals
as the Journal of Behavioral Finance, Journal of Economics and Finance, Journal
of Investing, and Journal of Wealth Management. Professor Small received a PhD
in finance from the University of Tennessee, is a Certified Financial Planner
(CFP), and a CFA.
Jeffrey S. Smith is an Assistant Professor of Economics and Finance at the Virginia
Military Institute (VMI). His primary teaching and research interests include in-
vestments and portfolio management. He previously served in the United States Air
Force for more than 20 years, teaching both at the Air Force Institute of Technology
and the U.S. Air Force Academy. He has published in such journals as the Journal of
Behavioral Finance, Journal of Economics and Finance, Journal of Investing, and Jour-
nal of Wealth Management. He has a BA in economics from the University of South
Carolina, an MS in applied economics from Wright State University, and a PhD in
economics from the University of Tennessee.
Robert Spliid is an external lecturer at the Copenhagen Business School, focusing
on private equity. He is the author of the book Private Equity: Raw Capitalism or
Active Ownership on the development of private equity in Denmark. He has pub-
lished in the Journal of Private Equity. He has been a columnist at Danish business
paper Brsen since 2000 and a frequent commentator on Danish Television on busi-
ness and financial affairs. Since 1982, Dr. Spliid has held leading capital market po-
sitions in banking in Denmark, Germany, Switzerland, Luxembourg, and Portugal
and is currently a Senior Vice President at Nykredit Bank in Denmark. He received
a Master of Economics from the University of Copenhagen and a PhD from Copen-
hagen Business School.
Emery A. Trahan is a Senior Associate Dean of Faculty and Research and Professor
of Finance at the DAmore-McKim School of Business at Northeastern University.
His current research interests include valuation, financial strategy, and M&As. His
research is widely published in academic and practitioner journals including the Fi-
nancial Analysts Journal, Financial Services Review, Journal of Investing, Journal of Al-
ternative Investments, Financial Management, and Journal of Financial Research, and
is cited in various media outlets. He is also the author of three books on M&As
and strategic planning. Professor Trahans teaching spans undergraduate and MBA
programs and includes courses in executive programs. He is also active in various
corporate consulting engagements. He holds a BS in accounting from the State Uni-
versity of New York at Plattsburgh and MA and PhD degrees from the University at
Albany. He is also a CPA and a CFA charterholder.
Marc P. Umber is an Assistant Professor of Corporate Finance at the Frankfurt School
of Finance and Management. His research interests in finance include corporate val-
uation, alternative investments, and the market for corporate control. His work has
been published in such journals as the Journal of Banking and Finance and Journal
of International Money and Finance. He holds an MSc in economics and a PhD in
finance from Goethe University in Frankfurt, Germany.
Tom Weidig, CFA, works for the Luxembourg insurance regulator Commissariat Aux
Assurances, where he reviews credit, market, natural catastrophe, and operational
a b o u t t h e c o n t r i b u t o r s xxvii

risk models of two of the largest reinsurers worldwide, including model governance.
He is involved in drafting technical specifications at the European Union level for
the new Solvency II regime. Dr. Weidig started out modeling the physical world
during a PhD program at Durham University and postdoctoral research in theoret-
ical physics using computational tools at Imperial College London and the Univer-
sity of Cambridge. He then worked in the financial industry on equity derivatives at
Bear Stearns. His specialist area is private equity and venture capital, where he wrote
a book Exposed to the J-curve and many articles, participated in raising a $160 million
innovative venture capital fund called Correlation Ventures, provided consultancy
such as to the European Investment Fund, and developed software for portfolio
management of private equity and venture capital fund-of-funds.
Ayse Dilara Altiok Yilmaz is an Assistant Professor in the Business Administra-
tion Department of the School of Economics and Administrative Sciences at Bah-
cesehir University, Turkey. She teaches undergraduate and graduate courses on
financial management, international finance, asset pricing, and financial markets
and institutions. Before joining Bahcesehir University, she worked as a research and
teaching assistant at Bogazici University, an account officer at PFS Finance, and an
assistant manager at Turk Dis Ticaret Bankasi. Her research has been published in
such journals as the Journal of Advanced Studies in Finance, International Business Re-
search, and World of Accounting Science. Professor Yilmazs current research interests
include banking, corporate finance, capital structure, and SME finance. She received
a BA in political science and international relations from Bogazici University, an
MBA from Galatasaray University, and a PhD in finance from Bogazici University.
Abbreviations

ADR American depository receipt


AIF alternative investment fund
AIFM alternative investment fund manager
AIFMD Alternative Investment Fund Managers Directive
AIM alternative investment market
ALM asset liability management
AMH adaptive market hypothesis
APV adjusted present value
ARD American research and development
ATM at the market
AUM assets under management
BDC business development company
BRIC Brazil, Russia, India, and China
CAGR compound annual growth rate
CAPEX capital expenditures
CAPM capital asset pricing model
CARs cumulative abnormal returns
CCF capital cash flow
CDO collateralized debt obligation
CFaR cash-flow-at-risk
CLO collateralized loan obligation
CM cash multiple
CMBS commercial mortgage-backed securities
CMPO confidentially marketed public offering
COGS costs of goods sold
COMI center of main interest
CRE commercial real estate
CRM customer relationship management
CRRA constant relative risk aversion
CSRC China Securities Regulatory Commission
DCF discounted cash flow
DDHME Dagenais and Dagenaiss higher moment estimator
DPI distribution to paid-in-capital

xxviii
a b b r e v i at i o n s xxix

EBITDA earnings before interest, tax, depreciation, and amortization


EIVs errors-in-variables
EMEA Europe, Middle East, and Africa
EMH efficient market hypothesis
EMPEA Emerging Markets Private Equity Association
ESG environment, social, and governance
ETFs exchange-traded funds
EVCA European Private Equity and Venture Capital Association
FATCA Foreign Account Tax Compliance Act
FCF free cash flow
FINRA Financial Industry Regulatory Authority
FSLIC Federal Savings and Loan Insurance Corporation
FSOC Financial Stability Oversight Council
GDP gross domestic product
GFC global financial crisis
GLPEI Global Listed Private Equity Index
GP general partner
HIRE hiring incentives to restore employment
HNWI high net worth individual
IARCP International Association of Risk and Compliance Professionals
ILPA Institutional Limited Partners Association
IPO initial public offering
IRC index return comparison
IRR internal rate of return
IRS Internal Revenue Service
IV instrumental variables
KKR Kohlberg, Kravis, and Roberts
LBD Longitudinal Business Database
LBO leveraged buyout
LP limited partner
LPA limited partnership agreement
LPE listed private equity
LPEQ Listed Private Equity Association
LTV loan-to-value
M&A mergers and acquisition
MBO management buyout
MIRR modified internal rate of return
MPT modern portfolio theory
MSCI Morgan Stanley Capital International
NAV net asset value
NWC net working capital
NOPAT net operating profit after-tax
NPV net present value
NVCA National Venture Capital Association
OACF operating activity cash flow
OCC Office of the Comptroller of the Currency
xxx a b b r e v i at i o n s

OCIE Office of Compliance Inspections and Examinations


PEGCC Private Equity Growth Capital Council
PEPI Private Equity Performance Index
PER price-to-earnings ratio
PIK payment in kind
PIPE private investment in public equity
PME public market equivalent
PPM private placement memorandum
PrEQIn Private Equity Quarterly Index
PRIVEX Socit Gnrale Private Equity Index
PTP public-to-private
PTPE publicly traded private equity
RCR relative compounded return
REIT real estate investment trust
RIC regulated investment company
RLBO reversed leveraged buyout
ROC return on capital
ROE return on equity
RTC Resolution Trust Corporation
RV residual value
RVPI residual value to paid-in
S&Ls savings and loans
SBA Small Business Administration
SBO secondary buyouts
SBICs small business investment companies
SCF simple cash flow
SEC Securities and Exchange Commission
SEO seasoned equity offering
SG&A selling, general and administrative expenses
SPAC special purpose acquisition corporation
SPE special purpose entity
TFP total factor productivity
TVE Thomson Venture Economics (mentioned in chapter 16)
TVEX Thomson Reuters VentureXpert (mentioned in chapter 15)
TVPI total value to paid-in
TWR time weighted return
UCC Uniform Commercial Code
UNI unlevered net income
UPMIFA Uniform Prudent Management of Institutional Funds Act
VaR value-at-risk
VC venture capital
VCs venture capitalists
VCT venture capital trust
VDD vendor due diligence
WACC weighted average cost of capital
WKSI well-known seasoned issuers
Private Equity
Part One

INTRODUCTION
1
Private Equity
An Overview
H. KENT BAKER
University Professor of Finance, Kogod School of Business, American University

GREG FILBECK
Samuel P. Black III Professor of Finance and Risk Management, Penn State Erie, the Behrend College

HALIL KIYMAZ
Bank of America Professor of Finance, Crummer Graduate School of Business, Rollins College

Introduction
Warren Buffet, the Oracle of Omaha, has never hidden his disdain for the private equity
(PE) crowd (Lenzner 2012). Buffett told Time magazine that I dont like what private
equity firms do in terms of taking every dime they can and leveraging (companies) up so
that they really arent equipped, in some cases, for the future (Foroohar 2012). Buffet is
critical of PE for its lopsided compensation structure in which investors, called limited
partners (LPs), pay 2 percent of their principal each year to the manager or general part-
ner (GP) even if the person accomplishes nothing or loses a bundle plus an additional
20 percent of the LPs profit if the manager succeeds. He is also critical of the amount
of debt that the managers pile on investment companies. However, others do not share
Buffets scorn for PE. To appreciate why requires an understanding of the nature of PE.
What is PE? Private equity is an asset class consisting of equity securities and debt in
companies not quoted on a public exchange. Major types of PE include venture capital,
buyouts, mezzanine capital, and distressed (turnaround) investments. Venture capital refers
to equity investments made, typically in less mature companies, for the launch of a seed
or start-up company, early stage development, or expansion of a business. By contrast, a
buyout involves investments in mature companies that require financing to pursue growth
opportunities. A buyout involves a group of investors acquiring a target company from its
current owners with the help of equity finance from a PE provider and debt finance from
financial institutions. Mezzanine capital refers to subordinated debt or preferred equity se-
curities that often represent the most junior portion of a companys capital structure that
is senior to the companys common equity. Distressed investments refer to investments in
equity or debt securities of financially stressed companies (Cumming 2010, 2012).
3
4 i n t r o d u c t i o n

A private equity fund invests in equity and to a lesser extent debt securities based
on investment strategies associated with PE. PE funds can also invest in publicly held
companies. Another recent innovation involves listed private equity, which refers to
publicly traded companies that invest capital in privately held enterprises. In 2013, Pro-
Shares launched Global Listed Private Equity ETF (BATS:PEX). PEX is the first glob-
ally diversified ETF focused on companies that invest mainly in private enterprises.
Jensen (1989) once described PE firms as lean, decentralized organizations with rel-
atively few investment professionals and employees. Although PE firms have grown
substantially larger over time, they still tend to be small relative to the firms in which
they invest.
As Michael L. Sapir, Chairman and CEO of ProShare Advisors LLC, notes, private
equity has long been a staple of institutional and high-net-worth individuals portfolios;
for other investors, its been difficult to access (ProShares 2013, p. 1). Traditionally, PE
investing has been available through limited partnerships, which can have high mini-
mums and other restrictions. Both accredited (qualified) and institutional investors,
particularly pension funds, endowments, and insurance companies, provide funds to
PE firms. PE firms typically try to generate returns by identifying private enterprises
with potential and providing them with long-term capital, which can used for such pur-
poses as funding new technologies, introducing new products or services, expanding
working capital within an owned company, making acquisitions, strengthening balance
sheets, and restructuring. PE firms try to improve a companys financial results and
prospects in hopes of allowing for a turnaround of a distressed company, reselling it to
another firm or cashing out via initial public offerings (IPOs). Thus, PE investments
often demand long holding periods.
Tien, Ho, and Chiu (2008) attribute the popularity of PE among investors to
these factors: (1) PE firms only invest the money of qualified investors; (2) the
investors of PE firms primarily have been institutions and not individuals; and (3)
PE firms provide investors with access to detailed information about their invest-
ment decisions. Most PE funds are closed-end funds that investors commit to
provide funds to pay for both investments in companies and management fees to
the PE firm.

G R O W T H I N P R I VAT E E Q U I T Y
The PE market grew steadily from the 1970s until 2007. For example, transactions in
the PE market increased from less than $ 30 billion in 1995 to more than $ 750 billion
in 2007 (Haarmeyer 2008). The amount of funds raised also increased sharply from
$3billion in 1985 to $ 314 billion in 2007. However, the height of the fundraising boom
was in 2007 with the creation of more than 450 PE firms. Since then, the drop in fund-
raising has affected the number of new firms entering the PE market. PE assets under
management reached to $ 2.5 trillion at the end of December 2013 (Preqin 2014). In
2014, the Private Equity Growth Capital Council (2014) reports 3,300 PE firms head-
quartered in the United States. The percent of total mergers and acquisitions (M&As)
based on PE increased from 7 percent in 1995 to 25 percent in 2011 in the United States.
Similar patterns are observed for the United Kingdom, Japan, and Germany during the
same period (Fruhan 2012).
P riv at e E qu it y : An Ov e rv ie w 5

T H E P R I VAT E E Q U I T Y P R O C E S S
The PE process starts with raising large pools of equity from investors. These funds
are typically used to buy operating firms using the equity and borrowings. The
amount of debt borrowed for such undertakings is often many times larger than
the amount of equity investment. PE funds are invested in a wide range of firms
from start-ups to mature firms. PE funds are also used for product and technology
development, geographical and product expansion of firms, or strengthening the
capital structure of firms, among others. The next step is to change managements
incentive structure to provide the management team with a larger portion of any
wealth created. The general partners (GPs) of the PE firms typically replace the
target firms board. GPs also manage the acquired firm for optimum cash-flow
generation and finally resell the acquired firm with the goal of generating a profit.
A typical PE firm is organized as a partnership or limited liability corporation (LLC)
where the investors are passive partners. Fund managers serve as general managers who
select firms to invest and manage those investments. Once the capital is committed, the
LPs have little power over how the firms operate. The PE firm managers are compen-
sated through an annual management fee, share of the funds profits, and some monitor-
ing fees. Martin and Schrum (2007) describe the structure of fees in detail and provide
empirical evidence on those fees.

Purpose and Scope


During the past few decades, PE has attracted considerable attention of investors, prac-
titioners, and academics. In fact, a substantial literature on PE has emerged. The pur-
pose of this book is to provide a synthesis of the theoretical and empirical literature
on PE in both emerging and developed markets. This book contains chapters written
by scholars and practitioners in their areas of expertise. The scope of the book is broad
examining such topics as the types of PE, how PE works, performance and measure-
ment, uses and structure, and trends in PE. Readers of Private Equity: Opportunities
and Risks can gain an in-depth understanding about PE from experts in this field from
around the world.
In todays financial environment, PE emerges as a dynamic area that contin-
ues to evolve at a rapid pace. This book takes readers through not only the core
topics and issues of PE but also examines the latest trends and cutting-edge de-
velopments. Additionally, discussion of research on PE permeates the book. The
coverage extends from discussing basic concepts and their application to increas-
ingly complex and real-world situations. Thus, this volume spans the gamut from
theoretical to practical, while attempting to offer a useful balance of detailed and
user-friendly coverage. Those interested in a broad survey will benefit as will those
looking for more in-depth presentations of specific areas within this field of study.
In summary, as the fifth book in the Financial Markets and Investments Series, Pri-
vate Equity: Opportunities and Risks offers a fresh look at the intriguing but complex
subject of PE.
6 i n t r o d u c t i o n

Distinctive Features
Private Equity: Opportunities and Risks has several distinguishing features.

The book provides a detailed look at one of the most dynamic areas in finance. It
skillfully blends the contributions of scholars and practitioners from around the
globe into a single review of important topics in this area. The varied backgrounds of
the contributors assure different perspectives and a rich interplay of ideas. The book
also reflects the latest trends and research involving PE in a global context.
While retaining the content and perspectives of the many contributors, the book
follows an internally consistent approach in format and style. Similar to a choir that
contains many voices, this book has many authors with their own separate voices. A
goal of both a choir and this book is to have the many voices sing together harmoni-
ously. Thus, the book is much more than simply a collection of chapters from differ-
ent authors.
When discussing the results of empirical studies that link theory and practice, the
objective is to distill them to their essential content so they are understandable to a
wide array of readers.
The end of each chapter contains four to six discussion questions that help to reinforce
key concepts. Guideline answers are presented at the end of the book. This feature
should be especially important to faculty and students using the book in classes.

Intended Audience
Given its broad scope, this practical and comprehensive book should be of interest to
professionals, investors, academics, and others interested in PE. For example, profes-
sionals and qualified investors can use this book to provide guidance in helping them
navigate through the key areas in PE. For academics the book provides the basis for
gaining a better understanding about various aspects of PE and as a springboard for
future research. They can also use the book as a stand-alone or supplementary resource
for advanced undergraduate or graduate courses in investments. Others including stu-
dents and libraries should find this book suitable as a reference. Thus, the book should
be essential reading for anyone who needs a better understanding of PE from seasoned
professionals to those aspiring to enter the demanding world of finance.

Structure of the Book


The remaining 26 chapters of the book consist of six parts. A brief synopsis of each chap-
ter by part follows.

PA R T O N E I N T R O D U C T I O N
Besides Chapter 1, the first part contains two other chapters. These chapters discuss the
economics of PE and market and regulatory developments of PE.
P riv at e E qu it y : An Ov e rv ie w 7

Chapter 2 Economics of Private Equity (Shantanu Dutta, Arup Ganguly,


andLin Ge)
One of the most controversial recent debates in finance among politicians, policymak-
ers, financial economists, and the media concerns the contribution of PE in society. The
PE industry has grown dramatically over the last few decades despite critics challeng-
ing the fabric of the economic model by portraying PE companies as heartless, profit-
seeking barbarians. This chapter investigates this controversial topic, going beyond just
the story-telling to examine the economics and benefits of PE. Overall, PE firms miti-
gate agency costs between the managers and shareholders, infuse good governance and
better management practices, improve operating performances, and create new jobs,
enabling value creation for both the investee companies and their investors.

Chapter 3 Private Equity in the United States and Europe: Market


andRegulatory Developments (Alexandros Seretakis)
A transformation of the PE industry has occurred in recent years. The financial crisis of
20072008 and the sovereign debt crisis caused a substantial decline in PE activity while
a new wave of rules has substantially strengthened the oversight and regulation of the PE
industry. Analysis suggests that the effects of market and regulatory developments will be
particularly adverse for the European PE industry. In contrast, PE activity in the United
States is likely to continue its growth aided by the revival of the U.S. economy and the
measured approach of U.S. regulators toward the risks posed by the industry. This chapter
offers an analysis of market and regulatory developments in the PE industry in the United
States and Europe in the postfinancial crisis era and provides an assessment of the future
of the PE industry.

PA R T T W O M A J O R T Y P E S O F P R I VAT E E Q U I T Y
The second part focuses on four major types of PE: venture capital, leveraged buyouts,
mezzanine capital, and distressed investments.

Chapter 4 Venture Capital in Europe: Myths and Reality (Jean-Michel Sahut


and Eric Braune)
The European venture capital (VC) industry is still young. This chapter examines this
industry and focuses on European VC practices. It also discusses the performance of
European VC financing and the convergence of U.S. and European practices. The Eu-
ropean VC industry has matured and European investors are becoming increasingly in-
volved in the companies management. Further, the performance gap between U.S. and
European VC firms has almost disappeared. Now, the performance of IPOs of European
VC firms is comparable to that of their U.S. counterparts. However, the practices of
European VC investors still differ widely from those of their U.S. colleagues. The Euro-
pean VC industry follows a trajectory based on its own experience. As public funding
predominates in Europe, this experience partially overlaps that of U.S. VC investors.

Chapter 5 Leveraged Buyouts (Christian Rauch and Marc P. Umber)


Leveraged buyouts (LBOs) represent the largest category of PE investments. With
an aggregate fundraising of $28 billion worldwide in the first quarter of 2014, buyout
8 i n t r o d u c t i o n

fundsare among the most important alternative investment vehicles for institutional
investors. In a typical LBO, a buyout fund acquires a company using the funds capital
with external debt funding. The main goal of an LBO is to generate returns to the funds
equity investors. Buyout funds typically restructure the acquired company using vari-
ous value enhancement strategies such as capital structure adjustments, changes to the
companys governance, or changes to its operational business to sell the company at a
premium. The intricacies of LBOs are many and constantly evolving. This chapter offers
a holistic overview on the environment of LBO transactions, including both the me-
chanics of the deal and the institutional characteristics of buyout funds.

Chapter 6 Mezzanine Capital and Commercial Real Estate (J. Dean Heller)
This chapter explores the salient features, varied modes, sources and users, and history of
mezzanine investment in commercial real estate (CRE). While mezzanine capital is also
important in corporate finance, the chapter focuses on CRE financing, where mezzanine
loans, in particular, became a routine component in larger debt financings before the
Great Recession in late 2008. Among other topics, the chapter traces a brief history of
real estate mezzanine financing from the late 1980s until the present and analyzes the dif-
ferences and similarities between equity and debt forms of mezzanine financing. As real
estate emerges from the shadow of that event, mezzanine capital appears ready to resume
its place on the stage, if perhaps more often in the guise of preferred equity than before.

Chapter 7 Distressed Debt Investments (Stephen G. Moyer and John D. Martin)


As an asset class, distressed debt is largely a by-product of the expansion of LBOs that
create corporations with high debt levels that are more prone to financial distress or
default. PE firms have grown to be important participants in the sector. Unlike the tradi-
tional buyout investment in which PE firms purchase a business by making a relatively
small equity investment and then borrowing the rest of the capital, in a distressed strat-
egy the investor purchases the firms distressed debt securities and then seeks to gain
control by converting those securities into a controlling stake of the equity of the post-
restructured entity. This chapter contains an overview of the evolution of PE firms in the
distressed debt sector, a summary of the primary strategies and techniques they employ,
and observations on how their activities have evolved. Finally, the chapter discusses the
future outlook of the distressed debt sector regarding PE firms.

PA R T T H R E E H O W P R I VAT E E Q U I T Y W O R K S
The third part focuses on valuation, cost of capital, and liquidity issues in PE. The sec-
tion further explores exit strategies for PE and the role of IPOs in PE.

Chapter 8 Valuing Private Equity (Harlan Platt and Emery Trahan)


PE firms generate returns by following an investment process similar to that of any in-
vestor. That is, they buy assets at a price for less than what they can be sold in the future.
Sound valuation analysis is a key part of this process. PE firms must employ methods
that reflect the current value of target companies as well as the potential value of these
companies after realizing expected operating and financing improvements. This chapter
P riv at e E qu it y : An Ov e rv ie w 9

presents a detailed discussion of the key issues involved in a valuation analysis and pro-
vides a comprehensive numerical example. Topics covered include measures of cash
flow, using cash flow to estimate value, discounted cash flow (DCF), adjusted present
value (APV), and multiples valuation methods, as well as estimating the cost of capital
for PE valuation.

Chapter 9 Cost of Capital for Private Equity (Alain Con and Aurelie
Desfleurs)
After a short literature review on the cost of capital for PE, this chapter focuses on esti-
mating the cost of equity for PE. First, unbiased estimators are used to correct for econ-
ometric bias induced by errors-in-variables in linear asset pricing models. Second, an
adjustment method is used to deal with the problem of stale valuation and illiquidity
observed in the PE industry. Third, with these valuation improvements, the chapter
provides new evidence on the importance of the liquidity premium on PE returns. The
results show that risk factors related to the market premium, size effect, book-to-market
effect, and liquidity premium may be useful in computing the cost of capital of PE. Ad-
justed multifactor models should be considered in the process of PE valuation with spe-
cial attention and adaptation for the different classes of PE.

Chapter 10 Liquidity Issues in Private Equity: Lessons Learned in the


PostFinancial Crisis (Elif Akben Selcuk and Ayse Altiok Yilmaz)
From 2003 to mid-2007, liquidity in the financial markets was high and PE firms made
large distributions to their investors, who reinvested these distributions into new PE
funds. With the onset of the financial crisis of 20072008, PE fund managers and inves-
tors faced severe liquidity problems. This chapter examines liquidity issues in PE within
the context of the financial crisis. The crisis caused a large increase in investors requests
to liquidate their PE interests due to the denominator effect, which is how investors
become superficially over-allocated in alternative investments such as PE due to the de-
clining investment values in other asset classes. Since PE investments were illiquid and
long term, effective exit strategies were difficult to achieve. The liquidity problems expe-
rienced during the crisis revealed the risks of the endowment model of investing. The
crisis revealed the need for investors to consider liquidity risk when designing portfolios.

Chapter 11 Private Equity Portfolio Management: Challenges, Approaches,


andImplementation (Thomas Meyer and Tom Weidig)
PE does not fit easily into the standard risk-return Markowitz portfolio optimization
framework used in the investment portfolios of institutional and other investors. The
concepts of return, risk, and correlation cannot be properly defined because no liquid
market with daily observable prices exists. Therefore, good portfolio management re-
quires a deep understanding of all the idiosyncrasies of PE investments and how they
interfere with standard portfolio management. Alternative frameworks such as behav-
ioral finance and adaptive market hypothesis are discussed. While integrating PE into a
wider portfolio is problematic, designing a portfolio on its own is more straightforward.
Different approaches are discussed especially on how to attain diversification through
various dimensions.
10 i n t r o d u c t i o n

Chapter 12 The Role of Private Equity in Initial Public Offerings: The Case
ofVenture Capital Firms (Shantanu Dutta, Arup Ganguly, and Lin Ge)
This chapter focuses on the role played by VC firms in IPOs. The literature shows the
characteristics and strategic objectives of VC firms can affect a firms IPO decision and
performance. The chapter also explores the relevant issues in VC-backed IPOs, starting
with the various conflicts of interest and then examines the three distinct roles played by
VC firms in IPOs: certification, underpricing, and monitoring. Research suggests that
VC-backed IPOs that bear an implicit certification from a venture capitalist have lower
costs of going public. A venture capitalists presence can also influence the underpricing
phenomenon of an IPO. Although most early studies contend that VC-backed IPOs
experience lower underpricing, the debate continues. The literature also shows that ven-
ture capitalists monitor their investments directly or indirectly to alleviate the problem
of information asymmetry and control the behavior of entrepreneurs.

Chapter 13 Exit Strategies in Private Equity (Didier Folus and Emmanuel Boutron)
The main goal of a PE fund manager is to receive a return in excess of the price paid for
the companies in the portfolio at the time of exit. Various exit strategies are available to
fund managers including a trade sale, which is the sale of the company to another PE
firm or a secondary buyout for a medium or large portfolio company. Another way to
exit is an IPO. A more recent exit strategy is for the portfolio company to pay a preferred
dividend to the PE fund in order to repay the initial invested amount. This strategy is
also known as a dividend recapitalization, which is sometimes financed with additional
debt. Financial economics can help inform the PE funds GPs about the different exit
routes. Pecking order theory, agency costs, and information asymmetry each offer rele-
vant scientific arguments explaining the observed behaviors.

PA R T F O U R P E R F O R M A N C E A N D M E A S U R E M E N T
The fourth part examines historical performance and benchmark biases in PE. The sec-
tion also discusses return performance and due diligence in PE.

Chapter 14 Private Equity Returns: Cross-Country Evidence (Kenneth Small


and Jeffrey Smith)
This chapter reviews returns for international PE funds and covers six primary factors
that drive PE investment across countries: (1) economic activity, (2) well-functioning
capital markets, (3) taxes, (4) corporate governance and investor protection, (5) the
social environment, and (6) the existence of entrepreneurial opportunities. Buyout deal
flow data show that most activity occurs in North America followed by Europe and
Asia. The chapter also examines the ability of listed PE to serve as a diversifying asset.
Evidence shows that listed PE returns are highly correlated with the returns of the S&P
500, MSCI World, and MSCI Europe equity indexes and may serve as a poor diversify-
ing asset if added to an all equity portfolio.

Chapter 15 Benchmark Biases in Private Equity Performance (Robert Spliid)


Studies dispute how PE as an asset class performs relative to public equity. These diverse
assessments occur because different benchmarks, samples, and measurement methods
P riv at e E qu it y : An Ov e rv ie w 11

are used. This chapter discusses the challenges of estimating the performance of PE
funds and PE as an asset class. It also examines the problems of using the internal rate of
return and the advantage of using a public market equivalent. Other challenges are find-
ing the appropriate data and benchmark and adjusting performance-to-risk and liquidity
differences. Performance can be measured either gross or net of fees with each method
relevant for different purposes. When choosing the sample for the performance analysis,
selection biases due to survivorship, vintage year, size of the fund, lack of update, and
skill of the investment manager should also be considered. Finally, the chapter describes
how selecting a data-weighting method depends on the purpose of the analysis.

Chapter 16 Return Persistence: Finding Top-Quartile Managers


(Adrian Oberli)
Recent empirical research shows mixed evidence of return persistence for PE. Still, the
conventional rationale that many PE investors employ during fund manager selection
focuses on prior fund returns as a means of predicting future top-quartile performance.
Finding fund managers based on past top-quartile performance is not straightforward.
First, considerable leeway exists for managers to define top-quartile performance, which
explains why much more than one quarter of all funds claim to be classified as top-
quartile. Second, PE groups often raise subsequent funds before the previous funds
performance can be accurately measured and classified as top-quartile. This process hin-
ders selecting a manager based on the assumption of return persistence. Thus, using a
holistic due diligence process might offer insights into identifying future top-quartile
PE managers.

Chapter 17 Private Equity Due Diligence (Manu Sharma and Esha Prashar)
PE due diligence is important because when done properly, it can create a clear opinion
about a future transaction for a PE firm and save the firm money by revealing potential
risks. The objective of due diligence is to provide investors with greater assurance that
they will realize the value of investments. Due diligence is no longer an isolated step but
is now integrated throughout the investment process. The chapter highlights the impor-
tance that due diligence plays in a PE transaction, the roles of both GPs and LPs, and the
involvement of different parties in the due diligence process. The chapter also describes
the steps in the due diligence process, challenges facing LPs investing in PE funds and
PE firms investing in operating (target) companies, and different types of due diligence.

PA R T F I V E P R I VAT E E Q U I T Y : U S E S A N D S T R U C T U R E
The fifth part focuses on using PE to create value and the managerial compensation struc-
ture. The global regulatory, ethical, and institutional framework of PE is also explored.

Chapter 18 Institutional Investors and Private Equity (Parvez Ahmed)


PE has experienced explosive growth over the past two decades, growing at an annual rate
of 20 percent and reaching 2.4 percent of U.S. stock market capitalization. When the U.S.
economy rebounded from the financial crisis of 20072008, institutional investors flocked
back to PE, contributing to levels not seen since 2008. Much of the capital invested in PE
12 i n t r o d u c t i o n

comes from wealthy private investors and institutional investors such as endowments,
foundations, pension funds, and insurance companies. Successful PE investment man-
agement requires selecting an effective PE firm targeted for investment, managing that
investment, and finally exiting the direct PE investment. Search costs, human capital, and
liquidity time preferences influence the decision to invest in PE. The herding behavior of
institutional investors with their propensity to hold large numbers of securities in their
portfolio also affects their choice of investment in PE. Compared to individual investors,
institutional investors are better positioned to mitigate the agency costs inherent in PE.

Chapter 19 Private Equity and Value Creation (Peter Harbula)


Although leverage can be a major factor contributing to the success of LBOs through its
positive impact on corporate governance, it can also be abused as evidenced by the PE led
M&A market in 2006 and 2007. Using data composed of continental European transac-
tions between 1997 and 2013, this chapter examines the relationships among leverage,
value creation, asset prices, and LBO returns and operating performance. The analysis
suggests that LBOs can create value. This takes place under two conditions: (1) when all
relevant key success factors are consistently applied and (2) when leverage is not used as
the only lever to abnormally boost investment returns but also as a means of enhancing
the corporate governance of the target companies, in conjunction with the other value cre-
ation levers such as cash-flow yield improvement and undervalued target of LBO funds.

Chapter 20 Compensation Structure (Ji-Woong Chung)


This chapter explores compensation structure of PE funds. Actual fees that GPs earn
and the economic incentives that they face are more subtle and complex than the tradi-
tional 2-and-20 rule suggests. This chapter examines the sources of these complexities
especially focusing on the potential conflicts of interest that may arise between LPs and
GPs due to particular forms of fee structure. The chapter also discusses the detailed
procedures for calculating various fees under different fund terms and examines issues
involving portfolio company fees. Finally, it reviews academic studies documenting the
empirical magnitude of incentives for GPs, determinants of fee structure, and cross-
sectional as well as time-series variations of compensation structure.

Chapter 21 Global Regulatory and Ethical Framework (Henry Ordower)


This chapter reviews statutory and court-sanctioned private regulatory frameworks af-
fecting the creation of PE funds and their primary activity of corporate acquisitions.
The chapter reviews U.S. legislation regulating securities, investment companies, and
tender offers. It also examines state antitakeover legislation, state court decisions on
hostile takeovers and poison pill defenses, as well as European Union directives on
takeovers and alternative investment fund managers. The chapter concludes that reg-
ulation in the United States has shifted the balance of power in corporate acquisitions
to incumbent management. Finally, the chapter examines the diametrically opposed
ethical views of PE funds as investment entities that either (1) acquire and destroy
corporations, harm communities, and eliminate employment or (2) enhance corpo-
rate value by rooting out corporate inefficiency, increasing employment, and enhanc-
ing community value.
P riv at e E qu it y : An Ov e rv ie w 13

PA R T S I X T R E N D S I N P R I VAT E E Q U I T Y
The sixth part investigates trends existing in PE including public outlets for PE invest-
ment through private investment and the development of a secondary market through
listed and publicly traded PE. Diversification opportunities through international and
emergency markets as well as the PE funds-of-funds structure are also covered. This
section ends with a look into the future for PE on a global basis.

Chapter 22 Private Investment in Public Equity (Na Dai)


Private investment in public equity (PIPE) has become an important financing alterna-
tive for U.S. public companies. This chapter provides an overview of the development
of the PIPE market between 1995 and 2012 including its market size, industry distribu-
tion, major investors, contractual design, and the role of placement agents. Respond-
ing to concerns about the propriety of so-called death-spiral PIPE transactions entered
by hedge funds, the Securities and Exchange Committee (SEC) began a series of en-
forcement actions in 2003. The chapter discusses these enforcement actions through
regulation and the resulting structural changes in the PIPE market. Finally, the chapter
describes the most recent innovations in the PIPE market such as the confidential mar-
keted public offering (CMPO) and the at-the-market (ATM) offering.

Chapter 23 Listed Private Equity (Douglas Cumming, Grant Fleming,


andSofia Johan)
This chapter examines a relatively unknown asset class of listed PE. Based on data cov-
ering 171 institutional investors in Europe between 2008 and 2010, the chapter notes
institutional investors propensity to invest in listed PE is mainly a function of size, type,
location, decision-making authority, and liquidity preferences. Investment in listed PE
is more commonly made by smaller, private (not public) pension institutions, those pre-
ferring liquidity, quick access, administrative and cash-flow management simplicity, and
institutions based in the United Kingdom, Switzerland, Sweden, and the Netherlands.
Institutions are also less likely to invest in listed PE when investment decision-making is
empowered to an alternative asset class team.

Chapter 24 Private Equity Growth in International and Emerging Markets


(Alexander Peter Groh)
This chapter presents international PE activity data dating back to 2000, grouped into
several traditional and emerging regions. No clear global development pattern exists
for the PE market. After the historic high in 2000, activity peaked again in 2007 with
high volatility since 2007. While the trend in traditional markets is relatively flat, emerg-
ing economies are increasingly attracting investments. Additionally, the number of
countries whose economic development is sufficiently mature to allow PE activity has
increased. This maturity is measured in a research project discussed within this chap-
ter that ranks 118 countries in terms of their PE attractiveness. However, the current
ranking is not the only important factor; a countrys medium-term development pro-
vides additional information to investors. Malaysia, Finland, Chile, Turkey, Colombia,
Mexico, Russia, the Philippines, Oman, Indonesia, Peru, Morocco, and the Baltics stand
out in terms of developing their PE attractiveness between 2009 and 2014.
14 i n t r o d u c t i o n

Chapter 25 Diversification Benefits of Private Equity Funds-of-Funds


(Axel Buchner and Markus Kuffner)
The double layer of fees in funds-of-funds is controversial, especially during periods
of high economic performance in which the extra fees can be large. This chapter takes
an investors view when examining the conditions in which an investment in funds-of-
funds pays off despite the fees-on-fees. The chapter considers an investors degree of
risk aversion and applies a certainty equivalent level of final wealth. Thus, determining
whether the benefits outweigh the additional costs requires knowledge of the investors
degree of risk aversion. This chapter shows that funds-of-funds substantially reduce the
risk of PE investments especially if management adds value by carefully selecting the
underlying funds leading to a higher acceptance of fees.

Chapter 26 Publicly Traded Private Equity (Henry Lahr)


Simultaneously with the PE boom of the 1990s, the sub-segment of publicly traded firms that
buy and sell private companies expanded almost as quickly as the entire asset class. Publicly
traded PE mirrors the structures of conventional non-listed PE but offers added advantages
to investors. This chapter first describes organizational forms and legal structures of publicly
traded vehicles and then analyzes their risk profiles and pricing characteristics. Systematic
risk of listed vehicles can be substantial while being highly diverse across internally and exter-
nally managed entities. Empirical research uses market prices of traded PE funds to estimate
the pricing relationship between net asset values (NAVs) and share prices. Evidence suggests
that NAV discounts mainly depend on liquidity, fund type, and investor sentiment.

Chapter 27 The Future of Private Equity: A Global Perspective (Dianna Preece)


This chapter discusses several factors that are likely to shape the future of PE. It explores
the impact of globalization on the industry and investment trends that may affect where
PE funds are deployed in the next decade. PE investments are being made in previously
overlooked areas such as Sub-Saharan Africa and East Asia. Investors are becoming more
demanding of firms, requiring greater transparency in an industry long recognized for its
secrecy and lack of regulatory oversight. This chapter provides a broad discussion of regu-
latory changes that are likely to influence PE. Many regulatory changes will improve over-
sight and transparency but will increase compliance costs. Finally, the chapter considers
investor trends. In the wake of low returns and underfunded pension plans, institutional
investing in PE continues to rise. Additionally, PE firms will continue to look to operational
gains at both the portfolio firm and PE firm level to improve returns in the coming decade.

Summary and Conclusions


The rapid structural, technological, and regulatory changes affecting the securities
industry worldwide have also affected PE. Furthermore, the causes of these changes
are complex and thus require a level of depth and breadth that this book provides. Pri-
vate Equity: Opportunities and Risks endeavors to interweave the contributions of both
scholars and practitioners into a single review of important but selective topics.
PE offers benefits for institutional and private wealth management clients including di-
versification and enhancement of risk-adjusted returns. However, the lack of transparency,
P riv at e E qu it y : An Ov e rv ie w 15

regulatory restrictions, and liquidity concerns that exist for some PE options limit their
attractiveness for some investors. Major PE types include VC, LBOs, mezzanine capital,
and distressed investments. PE valuation and determination of an appropriate cost of cap-
ital is complex due to the lack of an active secondary market. Most investments in PE are
for a finite period and investors must consider appropriate exit strategies.
Unlike public equity markets, benchmarking PE performance is challenging because
of the voluntary nature of contributors to vendor-based indexes. In assessing perfor-
mance, investors must understand the biases that exist within these indexes. While asset
allocation often drives performance differences in the public equity markets, managerial
selection and the possibility of return persistence in PE markets make the due diligence
process a vital consideration in PE selection. Managerial compensation frequently con-
tains an incentive fee component in addition to the management fees associated with
traditional markets. Recent innovations point to further development of a fledgling sec-
ondary market for PE and new strategies to enhance opportunity sets in the future.
So, while Warren Buffet offers valid points in his criticism of PE, empirical evidence
supports opportunities for diversification and enhanced risk-adjusted performance de-
spite higher fees and leverage that may be associated with some strategies. This book
endeavors to help clarify ways in which PE can be an appropriate outlet for some inves-
tors but provides cautionary notes for others.

References
Cumming, Douglas. 2010. Private Equity: Fund Types, Risks and Returns, and Regulation. Hoboken,
NJ: John Wiley & Sons, Inc.
Cumming, Douglas. 2012. The Oxford Handbook of Private Equity. New York: Oxford University
Press.
Foroohar, Rana. 2012. Warren Buffett Ready to Take Republicans Tax Challenge. Time, Janu-
ary 11. Available at http://swampland.time.com/2012/01/11/warren-buffett-to-mitch-
mcconnell-put-up-or-shut-up/.
Fruhan, William E., Jr. 2012. Role of Private Equity Firms in Merger and Acquisition Transactions.
Harvard Business School Background Note 206101. April.
Haarmeyer, David. 2008. Private Equity: Capitalisms Misunderstood Entrepreneurs and Catalysts
for Value Creation. The Independent Review 13:2, 245288.
Jensen, Michael C. 1989. Eclipse of the Public Corporation. Harvard Business Review 67:5, 6174.
Lenzner, Robert. 2012. Why Warren Buffet Disdains the Private Equity Crowd. Forbes.com., Jan-
uary 14. Available at http://www.forbes.com/sites/robertlenzner/2012/01/14/why-warren-
buffett-loathes-the-private-equity-crowd/.
Martin, James A., and Janice Lynn Schrum. 2007. Private Equity: The Leveraged Buyout Model
Revisited with a Dash of Clustering. Problems and Perspectives in Management 5:4, 7783, 91.
Preqin. 2014. The 2013 Preqin Private Equity Report. New York: Preqin. Available at https://www.
preqin.com/item/2013-preqin-global-private-equity-report/1/6131.
Private Equity Growth Capital Council. 2014. PE by the Numbers Quick Facts. Available at
http://www.pegcc.org/education/pe-by-the-numbers/.
Proshares. 2013. ProShares Launches Listed Private Equity ETF. Press Release, February 28. Available
at http://finance.yahoo.com/news/proshares-launches-listed-private-equity-154700158.html.
Tien, Chengli, Yvonne Yo, and Hongjen Chiu. 2008. Does the Involvement of Private Equity In-
vestments Matter to Firm Performance and Internationalization? Journal of American Acad-
emy of BusinessCambridge 13:2, 102109.
2
Economics of Private Equity
S H A N TA N U D U T TA
Associate Professor, University of Ottawa

A R U P GA N G U LY
PhD Candidate, University of Pittsburgh

LIN GE
PhD Candidate, University of Pittsburgh

Introduction
The general publics interest in private equity (PE) has recently burgeoned with the
medias scrutiny of Mitt Romneys past work experience in PE in the presidential elec-
tion in 2012. Yet, PE as a business model goes back to the beginning of the twenti-
eth century. According to financial historians, J. P. Morgan & Company conducted the
first formal PE deal, a leveraged buyout (LBO), in 1901. It bought a Pittsburgh-based
floundering steel company, Carnegie Steel Company, for nearly half a billion dollars and
turned it into the largest company in the world at that time, called the United States
Steel Corporation, which still exists today. Since then, thousands of PE deals have oc-
curred. Such growth in PE was fueled in the 1980s due to the popularity of high-yield
junk bonds. The largest PE buyout to date took place in February 2007 when KKR &
Co. L.P., TPG Capital, and Goldman Sachs Capital Partners jointly bought Energy
Future Holdings (TXU), a Texas-based electric utility company for $43.22 billion.
Table 2.1 highlights the 10 largest PE buyouts, ranked according to their inflation
adjusted deal value in 2014 dollars. This table reveals several key points. First, most of
these deals were initiated in 2006 and 2007 when the unprecedented credit bubble was
at its peak and debt was cheap. This finding makes sense from an economic perspective
because most PE buyouts are LBOs in which the PE firms use large amounts of debt to
acquire companies. Second, just a few companies in this trillion-dollar industry con-
ducted most of these large deals. Third, some of the biggest deals eventually resulted in
failures such as the buyouts of RJR Nabisco and more recently Energy Future Holdings.
This evidence shows that even the big players are vulnerable.
Following the debacle of RJR Nabisco in 1999, the general public and media lashed
out at PE firms often depicting them as heartless, profit-seeking barbarians. Critics

16
The Econ om ics of P riv at e E qu it y 17

Table 2.1Top 10 Private Equity Buyouts Ranked by Inflation-Adjusted Deal


Value in 2014 Dollars

Target Leading PE Year Deal Value Inflation-Adjusted


Acquirers ($billion) Deal Value in 2014
($ billion)
RJR Nabisco KKR 1989 30.06 57.67
Energy Future KKR, TPG, and 2007 43.22 49.59
Holdings Goldman Sachs
Equity Office Blackstone 2007 38.90 44.48
Properties
Hospital Corp. Bain, KKR, and 2006 32.19 37.99
ofAmerica Merrill Lynch
Harrahs Apollo Global 2006 27.40 32.22
Entertainment Management and
TPG
First Data KKR, TPG, and 2007 27.50 31.55
Goldman Sachs
Alltel Goldman Sachs 2007 27.15 31.15
andTPG
Clear Channel KKR, Bain, and 2006 25.70 30.22
Thomas H. Lee
Hilton Hotels Blackstone 2007 26.24 30.11
Alliance Boots KKR 2007 23.34 26.78

Source: Bloomberg Database.

strongly contend that PE firms kill jobs because their priority is to maximize profits and
the quickest way to do so is to cut costs through layoffs. Kosman (2009) claims that PE
firms cut 3.6 percent more jobs than their peers in the first two years of their ownership.
Another common criticism is that PE firms do not pay enough taxes. Specifically, the
carried interest for PE firms is taxed at capital gains rates, which is lower than the tax
rate on ordinary income (The New Yorker 2012). Some suggest changing this preferen-
tial tax treatment. For example, the Obama administration proposed a change in their
budget proposal for fiscal 2013, but nothing concrete has emerged.
A key difference between PE-owned firms and other publicly or privately traded
corporations is that PE-run firms are highly leveraged, usually taking on more than
70 percent debt. This capital structure gives them a huge advantage of debt tax shields,
which is a reduction in taxes that emanates from the tax deductibility of interest pay-
ments. Critics of PE have not only fulminated about PE firms exploitation of debt tax
shields in such a manner but also argued that the likelihood of PE-managed firms filing
bankruptcy increases due to their high debt levels. According to Kosman (2009), more
than half the companies that PE firms bought in the 1980s by borrowing more than
18 i n t r o d u c t i o n

$1 billion in junk bonds went bankrupt. PE firms are also blamed for being disruptive in
nature. They often bring in new management, streamline the company by selling not so
profitable units, perform across-the-board cuts, and cause a substantial cultural change
in the companies that they manage. These changes disrupt the normal functioning of
such firms and often result in certain inevitable social and economic costs.
Detractors of PE have also attacked the compensation schemes in the PE industry.
They claim that the fee structure consisting of both advisory and management fees is
too high and unreasonable. More recently, the financial press has castigated the PE
industry for raising more money than it can actually spend. The Wall Street Journal
(2014) notes that the dry powder, which is the amount of money raised by PE firms
but not yet invested, reached an all-time high of $1.141 trillion globally at the start
of June 2014. This trend raises a bigger concern that although PE firms can raise bil-
lions of dollars from investors, they cannot find worthy investments for some of these
funds.
Nevertheless, the growth of PE industry has been phenomenal in the last few de-
cades. As Figure 2.1 shows, PE exhibited strong growth since the early 1980s except
during the financial crisis of 20072008. Post-2009, investment in PE increased and
was a $3 trillion industry in 2013. Contrary to claims by the popular media that PE firms
are vulture funds that rapaciously strip firms and cash out, destroy jobs, and increase
the likelihood of bankruptcy for target companies, academic evidence draws an entirely
different picture.
The purpose of this chapter is to explore the economics of PE by investigating its
background and benefits. The chapter has the following organization. The next section
explores the economics behind the PE business structure followed by academic evi-
dence on PEs potential benefits. Finally, the chapter concludes with some closing ideas
and questions.

5000
4720

4500
Number of LBO or MBO Transactions

4183
4037
4000 3854
3739 3751

3500
3208
3074
3000 2906

2500 2360

2047
2000 1751
1541
1500 1364 1363
1275

952
1000
612
492
500 352
144 208 199 166 207 257 268
69 75 132
2 1 1 1 0 5 3 5 5 13 16 38 36 54
0
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
2099
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
13
19

Years

Figure 2.1 Number of Closed or Effective Leveraged Buyouts or Management


Buyouts, 1970 to 2013 The figure shows the number of LBOs or MBOs each year from
1970 to 2013. Source: Capital IQ (2014).
The Econ om ics of P riv at e E qu it y 19

Private Equity Background


The economics behind the PE business model begins with a seminal paper by
Modigliani and Miller (1958). They show that under certain perfect capital market
assumptions such as no taxes, no financial distress costs, no asymmetry in infor-
mation, pre-determined investment decisions, and the absence of transaction costs,
firms are indifferent toward their capital structure decisions.
As Figure 2.2A shows, in a perfect capital market world, how the pie is sliced be-
tween debt and equity holders does not affect the firms total value. As Figure 2.2B shows,
however, when incorporating real world frictions such as taxes, capital structure begins
to matter because now the size of the slice that goes to the debt and equity holders also
depends on what fraction goes to the government as taxes. Moreover, debt provides a
benefit known as the debt tax shield, which means that a firm can reduce its tax burden
by deducting its interest payments as an expense. Several academic studies starting with
Kaplan (1989) note that such tax shields are a crucial source of value creation in PE deals.
Finally, as Figure 2.2C shows, there is a likelihood of squandering wealth in the pres-
ence of operational inefficiencies, mismanagement, and other avoidable costs. The slice
that is wasted often results in making such firms attractive targets to PE buyers because
they invest in mature but poorly managed companies hoping to turn them around and
eventually make a profitable exit. A natural question is: How many poorly run com-
panies exist that can be lucrative targets for PE firms? The answer to such a question
might help to understand the phenomenal growth of the PE industry despite the nega-
tive press involving the PE business model.
Based on the data extracted from Aswath Damodarans database (Damodaran 2014),
more than 45 percent of publicly traded companies in the United States as of January
2014 have returns on capital (ROC) that are less than their costs of capital as shown in

Avoidable
Government Costs Equity
Debt
Equity
Government
Equity
Debt
Debt

Total value of the firm Size of the debt can Operational inefficiencies,
remains the same affect the portion going to mismanagement, and other
irrespective of the source of the government. avoidable costs can affect the
financing. portion going to the equity
holders.

Figure 2.2 Firm Value Figure 2.2A shows that under the perfect capital market
assumptions, financial structure is irrelevant (Modigliani and Miller 1958). As Figure 2.2B
shows, financial structure begins to matter for the firm in the presence of taxes. Figure 2.2C
illustrates that costs such as operational inefficiencies and mismanagement could affect the
portion of value going to the stakeholders.
20 i n t r o d u c t i o n

Figure 2.3. This finding provides a simple litmus test for distinguishing between well-
run and poorly run companies. The percentage of publicly traded firms that are poorly
performing in Europe, the emerging markets, and Japan are about 44, 46, and 46 per-
cent, respectively. While these percentages are comparable to the numbers in the United
States, other developed countries such as Australia, New Zealand, and Canada jointly
have nearly 68 percent of their publicly listed companies performing poorly with ROC
less than their costs of capital. Such figures also provide an indication of why PE has
become such a global phenomenon in the last few decades. Although Figure 2.3 only
considers the publicly listed companies worldwide, the number of poorly performing
firms is likely to be much higher by also including private firms.
In his seminal paper, Jensen (1989) argues that the avoidable waste in many public
corporations, which is reflected as the fraction of the slice titled avoidable costs in
Figure 2.2C, originates due to the conflict between the managers and the shareholders
over free cash flow (FCF) and can be avoided in a PE set-up. This solution is availa-
ble because being private enables firms to alleviate their agency problems through in-
centive alignment and better governance. Based on the trends observed at that time,
Jensen predicts that LBOs by PE firms would eventually overtake public corporations
in terms of being the most dominant corporate organization form. He thinks that the

80%

70%

60%

50%

40%

30%

20%

10%

0%
United States Europe Emerging Markets Japan Australia, New
Zealand, Canada
ROC < Cost of Capital ROC Cost of Capital

Figure 2.3 Poorly Performing Publicly Listed Companies in Different


Financial Markets, January 2014 This figure shows that the percentages of the
poorly performing publicly listed companies in the United States, Europe, emerging
markets, Japan, and other developed nations such as Australia, New Zealand, and Canada.
Europe includes the European Union, the United Kingdom, Switzerland, and Scandinavia.
Emerging markets include Asia, Latin America, Eastern Europe, Middle East, Africa,
China, and India. The cost of capital is the weighted average of the cost of equity and after-
tax cost of debt, weighted by the market value of equity and debt. Return on capital (ROC)
is estimated by dividing the after-tax operating income by the book value of the invested
capital. Source: Damodaran (2014).
The Econ om ics of P riv at e E qu it y 21

better governance structure, stronger managerial incentives, and the disciplining effect
of debt enable the PE business model to be more efficient.
Understanding what an LBO model is, which is sometimes known as going private
is important. As Figure 2.4 shows, the mechanics of an LBO are surprisingly simple in
theory. The PE firms screen thousands of potential targets generally focusing on those
that underperform their peer group, are relatively mature in their business with a stable
product demand, and have much unused debt capacity. After identifying a public com-
pany that meets its criteria, a PE firm buys the target company and takes it private by
heavily financing the purchase with debt, often borrowing against the target companys
assets. As Figure 2.4 shows, the target firm is often leveraged up in the LBO process to
more than 70 percent debt in the capital structure. During the private phase of the target
company, the PE firm generally brings in new expertise and enhances the profitability
of the target firm by reducing costs, improving operational efficiency, and aligning the
managerial incentives by making them active owners. According to Kaplan (1991), the
median LBO-target remains in the private phase for nearly seven years. More recent
estimates put the average holding period for PE-backed portfolio companies globally
between 4.5 and 5.6 years (Preqin 2014).
PE buyers are motivated by financial, not strategic reasons. Hence, after the private
phase during which they try to flip the company and at the same time service the as-
sumed debt, the PE buyer often exits by taking the target company public again and re-
ducing the debt level as Figure 2.4 shows. Other exit strategies in PE are also available.
The presence of the private phase in any PE deal is often a key to success. However, this
raises the question as to why the public status of the target company needs to be changed?
Financial economists ( Jensen 1989; Lehn and Poulsen 1989) contend that privately
owned companies should perform better than their listed counterparts because being
private helps firms reduce Jensens FCF problem and hence mitigates agency problems.
The same argument also applies when a PE firm acquires a public firm. Taking the target
firm private gives the PE firm the ability to make the required changes without the inter-
ference of shareholders or regulators and to avoid the associated costs of being public,
such as additional regulatory burdens.
Fully grasping the economics behind PE requires an understanding of the typical
structure of a PE fund as Figure 2.5 shows. The general partners (GPs) and the limited
partners (LPs) invest in a PE fund, which may consist of multiple portfolio compa-
nies. The idea is to diversify investments by investing in several different companies.
GPs are either a single PE firm or a consortium of PE firms. GPs not only manage

Net Debt Net Debt


Net Debt
LBO EXIT
Equity Equity
Equity

Figure 2.4 Mechanics of a Leveraged Buyout This figure depicts the mechanics of
a LBO showing how a low leveraged firm is made high leveraged, and then brought back to
the normal leverage level at the end of the LBO process.
22 i n t r o d u c t i o n

Limited Limited Limited


General Partner 1 Partner 2 Partner 3
Partners

PE FUND

Portfolio Portfolio Portfolio Portfolio


Company 1 Company 2 Company 3 Company 4

Figure 2.5 Typical Structure of a Private Equity Fund This figure shows the
typical structure of a PE fund, which includes the GPs, LPs, and their acquired portfolio
companies.

the fund on a daily basis but also invest their own capital. LPs are typically wealthy
individuals or institutions such as pension funds, endowments, and sovereign wealth
funds. LPs are investors who aim to earn returns without having to actively manage
the PE fund. They usually remain invested in a PE fund for 8 to 10 years. GPs typically
charge LPs a management fee, which is usually around 2 percent of the capital, for
their services in addition to 20 percent of the profits once they can cross the hurdle
rate, which is generally around 8 to 12 percent. Finally, the remaining 80 percent of
the profits goes to the LPs.
Such a PE business model has been successful not only because of the reduced
agency problems due to the private status of the portfolio companies but also because
of closer monitoring, better management expertise, and the disciplining effect of higher
debt levels in an LBO process. Some recent studies also provide evidence of a signifi-
cant positive impact on governance mechanisms when a PE firm takes over a publicly
traded firm in such a set-up (Ivashina and Kovner 2011; Acharya, Gottschalg, Hahn,
and Kehoe 2013). Furthermore, the key differentiator between a good and bad PE firm
is that the successful PE firm stays ahead of the game by raising new PE funds every
three to five years.
As Table 2.2 shows, the PE fund model has been immensely successful in raising
money. The top 10 funds by fund size have raised more than $10 billion each, with the
largest fund, J. P. Morgan Strategic Property Fund, having a fund size of $23 billion.
Such fund sizes reflect the willingness of LPs to invest in PE funds.

Private Equity Benefits


This section examines the benefits of PE. Undeniably, PE is a controversial financial
model with both advocates and critics. This section explores this contentious issue and
provides robust empirical findings from the extant literature.
The Econ om ics of P riv at e E qu it y 23

Table 2.2 10 Largest Private Equity Funds

Fund Name Fund Size Company Country


($million) Headquarters
J. P. Morgan Strategic $23,000 J. P. Morgan Investment United States
Property Fund Management Inc.
GS Mezzanine 20,000 Goldman Sachs Group, United States
Partners V, L.P. Merchant Banking
Division
TPG Partners VI, L.P. 19,000 TPG Capital, L.P. United States
Apollo Investment 18,400 Apollo Global United States
Fund VIII, L.P. Management, LLC
Blackstone Capital 16,700 The Blackstone Group, United States
Partners VI, L.P. PE Group
Performa 16,243 Desenvolve SP, Brazil
Investment Arm
Apax Europe VII 15,400 Apax Partners LLP United Kingdom
Warburg Pincus 15,000 Warburg Pincus LLC United States
PE X, L.P.
CVC European Equity 14,354 CVC Capital Partners United Kingdom
Partners VI LP Limited
PRISA LP 14,100 Prudential Real Estate United States
Investors

Source: Capital IQ (2014).

P R I VAT E E Q U I T Y F I R M S A N D VA L U E C R E AT I O N
Do PE firms create or destroy value for their target companies? Value creation is possi-
ble by reducing costs, improving operating performance and asset utilization, and gen-
erating growth. In an early study, Bull (1989) analyzes LBOs of 25 companies using
seven different accounting measures. He finds that the financial performance of these
companies two years post-buyout is better than the financial performance two years
pre-buyout. This superior performance post-buyout goes beyond the benefits of tax sav-
ings from interest tax shields.
Other studies provide evidence of value creation. Kaplan (1989) examines 76 large
management buyouts (MBOs) of public companies completed between 1980 and 1986.
He finds that their operating income increases three years after the buyouts. Kaplan
further confirms the enhancements in operating performances are due to improved in-
centives and reduced agency costs rather than layoffs or managerial exploitation of the
stockholders. Smith (1990) documents significant and sustained improvements in op-
erating returns per employee and per dollar of operating assets for 58 MBOs of public
companies that occurred between 1977 and 1986. Muscarella and Vetsuypens (1990)
24 i n t r o d u c t i o n

study 72 reverse LBOs from January 1976 to July 1987. They find significant operational
improvements in these firms due to improved asset utilization. Using the data from the
Longitudinal Business Database (LBD) of the U.S. Census Bureau, Lichtenberg and
Siegal (1990) analyze plant-level data for 131 LBOs between 1981 and 1986. They doc-
ument that total factor productivity (TFP) increased significantly more than the indus-
try average following an LBO. In a more recent study involving the United Kingdom,
Harris, Siegel, and Wright (2005) complement the results of Lichtenberg and Siegal.
Based on their analysis of 35,753 manufacturing units, the authors find the TFP of these
plants increases substantially after buyouts.
Growth provides another route through which PE firms create value for their target
companies. Bernstein, Lerner, Srensen, and Strmberg (2010) examine the impact of
PE involvement in 20 different industries across 26 different countries between 1991
and 2007. They find that industries in which the PE funds have invested in the previous
five years grew faster both in productivity and employment. The authors document this
pattern for both common law nations and continental Europe.
If financial markets are reasonably efficient, LBO announcement returns should be a
good sign of whether investors think PE creates value in their target companies. Analyz-
ing a sample of companies that went private between 1980 and 1987, Lehn and Poulsen
(1989) find that the average cumulative abnormal returns associated with going private
announcements are 16.3, 19.9, and 20.5 percent for the windows [1,1], [10,+10],
and, [20,+20], respectively. Brown, Fee, and Thomas (2009) document a similar result
and find that the average abnormal announcement return to the LBO firms is about 18.6
percent. Moreover, Hege, Lovo, Slovin, and Sushka (2011) also document higher an-
nouncement returns for target firms if they are acquired by a PE firm versus a strategic
buyer. Overall, the empirical evidence strongly suggests that PE firms create value for
their target companies.

P R I VAT E E Q U I T Y F I R M S A N D VA L U E C R E AT I O N F O R L I M I T E D
PA R T N E R S
Another important consideration is whether PE firms also create value for their inves-
tors (i.e., LPs), many of whom are financial institutions such as pension funds and insur-
ance companies. The evidence on this question is mixed. Analyzing the equity returns
for 58 LBOs, Muscarella and Vetsuypens (1990) show that the median annualized rate
of return on equity is 268.4 percent. They further find that the median return on equity
in the case of divisional reverse LBOs is even higher at 301.2 percent. Ljungqvist and
Richardson (2003) use the cash flow data of PE funds raised between 1981 and 2001
and document that PE firms generate risk-adjusted excess returns of more than 5 per-
cent annually.
Yet, other studies document that PE funds underperform the general market. Using
the Thomson VentureXpert database, Kaplan and Schoar (2005) evaluate the returns of
169 LBO funds between 1980 and 2001. They find that after deducting fees, returns on
the median fund are less than those of the S&P 500 index. Other researchers using the
same database but different time periods and econometric techniques reach a similar
conclusion as Kaplan and Schoar, that PE funds on average underperform the market
(Phalippou and Gottschalg 2009; Driessen, Lin, and Phalippou 2012).
The Econ om ics of P riv at e E qu it y 25

Although the evidence on PE fund performance is mixed, a lack of good data pres-
ents major problems when measuring their performance. Because researchers often use
different sources of commercial and proprietary data, they frequently find conflicting re-
sults. Harris, Jenkinson, and Stucke (2010) illustrate that a 10-year internal rate of return
(IRR) for venture capital (VC) funds given by two leading PE data providers (Thomson
Reuters and Cambridge Associates) between the first quarter of 1998 and the first quar-
ter of 2008 was 17.20 and 32.83 percent, respectively. Harris, Jenkinson, and Kaplan
(2014) document that Thomson VentureXpert, also known as Venture Economics, un-
derstates the PE fund returns. Using verified data on 1,400 PE funds received from more
than 200 institutional investors, they document that PE funds consistently outperform
the stock market. Based on the recent trend in research on PE fund performance using
higher quality PE fund level data will allow PE firms to create more value for their LPs.

P R I VAT E E Q U I T Y F I R M S A N D G O V E R N A N C E S T R U C T U R E
As mentioned previously, one reason LBOs might become the dominant corporate
organization form is that they create better governance structures. Specifically, Jensen
(1989) argues that debt mitigates the agency problem due to FCF in the same manner
as dividends. Further, debt is preferable to dividends because debt can be a stricter
commitment device forcing managers to pay out excess cash to the shareholders when
limited investment opportunities are available ( Jensen 1986). Besides the FCF agency
problem, the author argues that LBO firms have more concentrated ownership, usually
controlled by a few buyout sponsors. Therefore, they tend to be more actively involved
in monitoring the management compared to the similar public firms with relatively dis-
persed ownership. Due to the large amount of debt taken during LBOs, firms are forced
to improve and optimize their structures to meet those debt obligations. Further, the
compensation systems in LBOs usually have high pay-performance sensitivity. Taken
together, LBOs are believed to lead to a more efficient governance structure.
Following Jensen (1989), much research examines the impact of LBOs by PE firms
on a target firms governance structure. So far, most of the evidence shows a positive re-
lationship between LBOs and the efficiency and effectiveness of corporate governance
for firms both within and outside of the United States.
Baker and Wruck (1989) conduct a case study on O. M. Scott after its LBO in 1986.
They find that the firm experienced better operating performance and improved invest-
ment policies following its LBO. The authors attribute these changes to the pressure
of heavy debt obligations and management equity ownership that induces a strong in-
centive compensation structure. Analyzing the case of Safeway after its LBO in 1986,
Denis (1994) concludes that improved incentive structure and monitoring from the
LBO resulted in a more productive cash-generating process.
To generalize the results of these two case-based studies, other researchers empiri-
cally examine the active monitoring role of LBO sponsors. Using a sample of U.K. LBOs,
Acharya, Kehoe, and Reyner (2009) interview 20 executives in the United Kingdom.
They find that PE boards are becoming more successful at aligning the interests of man-
agement and other stakeholders and hence are more effective. In another U.K.-based
study, Cornelli and Karaka (2012) find that after LBOs, outside directors tend to be re-
placed by LBO sponsors who are much more active in their supervisory responsibilities.
26 i n t r o d u c t i o n

They also observe that CEOs pay-performance sensitivity and firm operating perfor-
mance increases following an LBO. Similarly, using a sample of U.S. LBOs, Gong and
Wu (2011) find that after LBOs, boards are much more likely to replace CEOs in firms
with high agency costs and low pre-LBO return on assets. They further report that post-
LBOs, the boards tend to replace entrenched managers. These results are consistent for
firms worldwide.
To complement empirical evidence, Grinsteins model (2006) shows that commit-
ment to financial claims helps investors take disciplinary actions against management.
According to Edmans model (2011), leverage concentrates stakeholders stakes and
induces them to learn more about the firms cash flows, resulting in better investment
decisions. Overall, the evidence suggests that LBOs by PE firms induce more active
monitoring of management and accordingly create better corporate governance and
eventually better firm performances.
Although more active monitoring by LBO sponsors leads to better governance in
LBO firms, LBOs have other benefits. For example, researchers document that using
equity-based incentives has become increasingly popular for top management to align
their incentives with those of the shareholders (Kaplan and Strmberg 2009; Acharya
et al. 2013). Performance-based compensation/incentive structures tend to work better
in aligning the interests of managers and shareholders.
LBOs also increase the pressure from higher debt obligations, which encourages
management to improve the firms efficiency to meet those obligations. Although Jen-
sens (1989) argument is intuitive, Cotter and Peck (2001) find that with actively moni-
tored managers, tighter debt obligations do not significantly improve firm performance.
In other words, the monitoring by LBO sponsors substitutes for tighter debt terms in
motivating managers. However, debt obligations still constrain LBO firms.
Another benefit of PE firms is that they provide knowledge and management exper-
tise. Bloom, Sadun, and van Reenen (2009) use survey data on management practices
in more than 4,000 mid-size manufacturing companies from countries in Asia, Europe,
and the United States. They document that PE-owned firms are significantly better
managed than their peers run by the government or other privately owned firms. Their
management practices are also superior on average compared to publicly listed firms
although the difference is not statistically significant.
One of the biggest empirical challenges that researchers face in determining the as-
sociation between LBOs and corporate governance is the issue of endogeneity. This
issue might be a problem of self-selection where only certain types of firms choose to
undergo an LBO, which in turn leads to improved governance. Overall, however, LBOs
seem to improve a firms governance structure via more active monitoring from LBO
sponsors, higher pay-performance sensitive compensation contracts, and greater pres-
sure due to the high leverage.

P R I VAT E E Q U I T Y F I R M S A N D B A N K R U P TC Y
Rappaport (1990) criticizes LBOs by PE firms for their lack of financial flexibility,
which stems from their high leverage. Given the high leverage, some argue that LBOs
are likely to increase the target firms bankruptcy rate but evidence does not support
this claim.
The Econ om ics of P riv at e E qu it y 27

According to Kaplan and Strmberg (2009), among the 17,171 worldwide LBO
transactions between 1970 and 2007, only 6 percent eventually resulted in bankruptcy.
The bankruptcy rate increases to 7 percent excluding post-2002 LBOs. The authors
assume an average holding period of six years, which gives them an annual default rate of
1.2 percent per year. Surprisingly, this rate is actually lower than the average default rate
of all U.S. corporate bond issuers between 1980 and 2002. Using a sample of 830 LBO
deals in France, Boucly, Sraer, and Thesmar (2011) find no difference in bankruptcy
rates between target firms and control firms (non-LBO firms, but similar otherwise),
either within three years following LBOs or at any point of time.
Tykvov and Borell (2012) study the bankruptcy risks of European companies
around LBOs from 2000 to 2008. Their evidence shows that distress risk increases after
LBOs but it does not lead to higher bankruptcy rates than comparable non-LBO firms.
They further report that with experienced PE firms, the bankruptcy rates are even lower,
which reflect the capability of PE firms in terms of financial distress management. Con-
sistent with these results, Wilson and Wright (2013) use unique, hand-collected data
consisting of a large sample of U.K. firms. The authors find that PE-backed LBOs are
not riskier than the non-LBO firms and are more likely to avoid insolvency when they
are distressed.
Hotchkiss, Smith, and Strmberg (2014) report similar results using a U.S. sample.
They study the role of PE firms in financial distress using a sample of more than 2,000
firms that had LBOs between 1997 and 2010. The authors conclude that controlling
for leverage, PE-backed firms are not more likely to go bankrupt than other leveraged
firms. On the contrary, when firms do default, PE-backed firms recover faster via effec-
tive restructuring.
Overall, LBOs do not increase the likelihood of bankruptcy despite their high lever-
age structure. Harford and Kolasinski (2014) confirm this conclusion by showing that
among the large U.S. LBO transactions from 1993 to 2001, bankruptcies and other fi-
nancial distress related restructuring only account for 15 percent of all exits. Further-
more, PE firms apparently help with the efficient restructuring when firms do default.

P R I VAT E E Q U I T Y F I R M S A N D U N E M P L O Y M E N T
A hotly debated issue is whether PE firms eliminate jobs in their acquired companies
in an attempt to cut costs. Labor unions, public media, and policymakers are quick to
point out and to accentuate any news of layoffs post-LBO. Surprisingly, empirical evi-
dence on whether PE firms reduce employment is not clear-cut.
Muscarella and Vetsuypens (1990) find that employment decreases by 0.6 percent
between the time the buyout takes place and the firm goes public again. By contrast,
using a sample of 48 MBOs, Kaplan (1989) finds that median employment increased
by 0.9 percent. Opler (1992) studies 44 LBOs in the latter half of the 1980s and finds a
slight increase of 0.3 percent in employment.
In an attempt to settle this debate, Davis, Haltiwanger, Handley, Jarmin, Lerner,
and Miranda (2013) perform a comprehensive study and collect data on 3,200 target
firms of LBOs and their 150,000 establishments between 1980 and 2005. Using U.S.
Censusdata and controlling for size, age, prior growth rate, and industry, Davis et al.
(2013, p. 1) find that, the sum of gross job creation and destruction at target firms
28 i n t r o d u c t i o n

exceeds that of controls by 14 percent of employment over two years. Such a finding
also indicates that even though the PE firms cut jobs after they take their target compa-
nies private, they also create more new jobs at new establishments. So any valid criticism
needs to consider both the job reduction and new job creation in order to provide a
holistic picture.

P R I VAT E E Q U I T Y F I R M S A N D I N N O VAT I O N
A common belief is that PE firms as financial buyers focus more on short-term profits
than on long-term growth. Substantial evidence shows improvements in operational
efficiency following LBOs. However, recognizing whether negative changes occur in
long-run investments or innovations (e.g., research and development (R&D) expen-
ditures) after going private is important. The empirical evidence on this issue is mixed.
One measure of long-run investments or innovation is R&D expenditures. Lichten-
berg and Siegel (1990) study the effects of LBOs occurring in the early 1980s. They
find that LBO targets are less R&D-intensive compared to other firms. However, this
observation may be because LBO targets tend to be in non-R&D-intensive industries. A
more important finding is that no significant change occurs with R&D intensity before
and after LBOs. Consistent with these results, using a sample of LBOs between 1977
and 1988, Hall (1991) finds no significant decrease in R&D after an LBO. Furthermore,
using a sample of U.K. LBOs, Wright, Thompson, and Robbie (1992) report that PE
firms appear to increase product development and asset purchases after LBOs. Yet,
using a sample of LBOs between 1981 and 1987, Long and Ravenscraft (1993) find
that R&D expenditures decrease post-LBOs.
Another group of studies use patents as a measure of long-run investment and inno-
vation. Lerner, Sorensen, and Strmberg (2011) do not find any evidence that LBOs
sacrifice long-term investments based on patents filed by 472 PE-backed firms be-
tween1986 and 2005. Ughetto (2010) examines a sample of Western European firms
undergoing LBOs between 1998 and 2004. She finds that the number of patents actu-
ally increases after LBOs. The impact of LBO on patents depends more on the type of
LBO investors than on LBOs in general.
Although the evidence is mixed on the impact of LBOs on R&D, the findings are
positive on the impact of LBOs on patent activity. Overall, the issue of whether PE firms
tend to sacrifice long-term growth for short-term profitability needs further research.

P R I VAT E E Q U I T Y F I R M S A N D D I V I D E N D S
Another common misconception is that PE firms strip their target companies by taking
on high levels of debt in order to award themselves dividends. For example, Appelbaum
and Batt (2014, p. 9) note:

They refinanced many of their loans through amend and extend agreements;
relied more on management fees than profits from the sale of their companies;
and made greater use of dividend recapitalizations, loading portfolio compa-
nies with more debt in order to pay dividends to themselves and their limited
partners.
The Econ om ics of P riv at e E qu it y 29

Harford and Kolasinski (2010) analyze a comprehensive sample of 788 large U.S.
PE buyout transactions from 1993 to 2001. They find that special dividends, paid to
GPs and LPs, are rare, occurring in only 42 instances out of 2,435 firm-years in their
entire sample. The authors further confirm that those 42 instances are uncorrelated with
financial distress in the future.

Summary and Conclusions


The PE industry has witnessed a staggering growth rate since the early 1990s. Such
ubiquity of PE ownership can be overwhelming at times and intimidating to some. The
critics of PE and the financial press often portray PE firms as corporate raiders, focus on
specific case studies, and downplay the benefits of PE.
This chapter examines the empirical evidence on PE firms and their performances
both at the PE portfolio firm level and the PE fund level to explore the economics and
the benefits of PE. The goal of this chapter is to provide a synthesis of the empirical ev-
idence. This review suggests that the benefits of PE are real. On average, PE firms not
only improve the operating performance of the investee companies but also enhance
their corporate governance. They bring in new skills, new jobs, innovation, and better
knowledge of management practices besides infusing new capital. In summary, the PE
firm business model reduces agency costs enabling firms to create value for both their
investee companies and investors. Thus, PE has become an integral part of the portfolios
of many institutional and high net worth individual investors.

Discussion Questions
1. Explain the nature and functions of PE firms.
2. Explain the difference between GPs and LPs.
3. Identify the criticisms and benefits of PE.
4. Discuss how PE firms improve corporate governance in their investee companies.
5. Discuss how PE firms create value in their portfolio companies.

References
Acharya, Viral V., Oliver F. Gottschalg, Moritz Hahn, and Conor Kehoe. 2013. Corporate Govern-
ance and Value Creation: Evidence from Private Equity. Review of Financial Studies 26:2,
368402.
Acharya, Viral V., Conor Kehoe, and Michael Reyner. 2009. Private Equity vs. PLC Boards in the
UK: A Comparison of Practices and Effectiveness. Journal of Applied Corporate Finance 21:1,
4556.
Appelbaum, Eileen, and Rosemary Batt. 2014. Private Equity at Work: When Wall Street Manages
Main Street. New York: Russell Sage Foundation.
Baker, George P., and Karen H. Wruck. 1989. Organizational Changes and Value Creation in Lev-
eraged Buyouts: The Case of the OM Scott & Sons Company. Journal of Financial Economics
25:2, 163190.
30 i n t r o d u c t i o n

Bernstein, Shai, Josh Lerner, Morten Srensen, and Per Strmberg. 2010. Private Equity and Indus-
try Performance. Working Paper, National Bureau of Economic Research.
Bloom, Nick, Rafaella Sadun, and John van Reenen. 2009. Do Private Equity-Owned Firms Have
Better Management Practices? In Anuradha Gurung and Josh Lerner, eds., The Globalization
of Alternative Investments Working Papers, Volume 2: The Global Economic Impact of Private
Equity Report 2009, 2543. Geneva: World Economic Forum.
Bloomberg. 2014. Bloomberg Database.
Boucly, Quentin, David Sraer, and David Thesmar. 2011. Growth LBOs. Journal of Financial Eco-
nomics 102:2, 432453.
Brown, David T., C. Edward Fee, and Shawn E. Thomas. 2009. Financial Leverage and Bargaining
Power with Suppliers: Evidence from Leveraged Buyouts. Journal of Corporate Finance 15:2,
196211.
Bull, Ivan. 1989. Financial Performance of Leveraged Buyouts: An Empirical Analysis. Journal of
Business Venturing 4:4, 263279.
Capital IQ. 2014. Capital IQ Database. Available at https://www.capitaliq.com.
Cornelli, Francesca, and Oguzhan Karaka. 2012. Corporate Governance of LBOs: The Role of
Boards. Working Paper, London Business School and Boston College. Available at http://
ssrn.com/abstract=1875649.
Cotter, James F., and Sarah W. Peck. 2001. The Structure of Debt and Active Equity Investors: The
Case of the Buyout Specialist. Journal of Financial Economics 59:1, 101147.
Damodaran, Aswath. 2014. Database. Available at http://pages.stern.nyu.edu/~adamodar.
Davis, Steven J., John C. Haltiwanger, Kyle Handley, Ron S. Jarmin, Josh Lerner, and Javier Miranda.
2013. Private Equity, Jobs, and Productivity. Working Paper, National Bureau of Economic
Research.
Denis, David J. 1994. Organizational Form and the Consequences of Highly Leveraged Transac-
tions: Krogers Recapitalization and Safeways LBO. Journal of Financial Economics 36:2,
193224.
Driessen, Joost, Tse-Chun Lin, and Ludovic Phalippou. 2012. A New Method to Estimate Risk and
Return of Non-traded Assets from Cash Flows: The Case of Private Equity Funds. Journal of
Financial and Quantitative Analysis 47:3, 511535.
Edmans, Alex. 2011. Short-term Termination Without Deterring Long-term Investment: A Theory
of Debt and Buyouts. Journal of Financial Economics 102:1, 81101.
Gong, James J., and Steve Y. Wu. 2011. CEO Turnover in Private Equity Sponsored Leveraged
Buyouts. Corporate Governance: An International Review 19:3, 195209.
Grinstein, Yaniv. 2006. The Disciplinary Role of Debt and Equity Contracts: Theory and Tests.
Journal of Financial Intermediation 15:4, 419443.
Hall, Bronwyn H. 1991. The Impact of Corporate Restructuring on Industrial Research and Devel-
opment. Working Paper, National Bureau of Economic Research.
Harford, Jarrad, and Adam Kolasinski. 2010. Evidence on How Private Equity Sponsors Add Value
from a Comprehensive Sample of Large Buyouts and Exit Outcomes. Working Paper, Univer-
sity of Washington.
Harford, Jarrad, and Adam Kolasinski. 2014. Do Private Equity Returns Result from Wealth Trans-
fers and Short-termism? Evidence from a Comprehensive Sample of Large Buyouts. Manage-
ment Science 60:4, 888902.
Harris, Richard, Donald S. Siegel, and Mike Wright. 2005. Assessing the Impact of Management
Buyouts on Economic Efficiency: Plant-level Evidence from the United Kingdom. Review of
Economics and Statistics 87:1, 148153.
Harris, Robert S., Tim Jenkinson, and Steven N. Kaplan. 2014. Private Equity Performance: What
Do We Know? Journal of Finance 69:5, 18511882.
Harris, Robert S., Tim Jenkinson, and Rudiger Stucke. 2010. A White Paper on Private Equity Data
and Research. Working Paper, UAI Foundation Consortium.
Hege, Ulrich, Stefano Lovo, Myron B. Slovin, and Marie E. Sushka. 2011. How Does Private Equity
Bid in Corporate Asset Sales? Working Paper, HEC Paris.
The Econ om ics of P riv at e E qu it y 31

Hotchkiss, Edith, David C. Smith, and Per Strmberg. 2014. Private Equity and the Resolution of
Financial Distress. Working Paper, Boston College, Swedish House of Finance, and University
of Virginia. Available at http://ssrn.com/abstract=1787446.
Ivashina, Victoria, and Anna Kovner. 2011. The Private Equity Advantage: Leveraged Buyout
Firms and Relationship Banking. Review of Financial Studies 24:7, 24622498.
Jensen, Michael C. 1986. Agency Costs of Free Cash Flow, Corporate Finance, and Takeovers.
American Economic Review 76:2, 323329.
Jensen, Michael C. 1989. Eclipse of the Public Corporation. Harvard Business Review 67:5, 6174.
Kaplan, Steven N. 1989. The Effects of Management Buyouts on Operating Performance and
Value. Journal of Financial Economics 24:2, 217254.
Kaplan, Steven N. 1991. The Staying Power of Leveraged Buyouts. Journal of Financial Economics
29:2, 287313.
Kaplan, Steven N., and Antoinette Schoar. 2005. Private Equity Performance: Returns, Persistence,
and Capital Flows. Journal of Finance 60:4, 17911823.
Kaplan, Steven N., and Per Strmberg. 2009. Leveraged Buyouts and Private Equity. Journal of Ec-
onomic Perspectives 23:1, 121146.
Kosman, Josh. 2009. The Buyout of America. New York: Portfolio.
Lehn, Kenneth, and Annette Poulsen. 1989. Free Cash Flow and Stockholder Gains in Going Pri-
vate Transactions. Journal of Finance 44:3, 771787.
Lerner, Josh, Morten Sorensen, and Per Strmberg. 2011. Private Equity and LongRun Invest-
ment: The Case of Innovation. Journal of Finance 66:2, 445477.
Lichtenberg, Frank R., and Donald Siegel. 1990. The Effects of Leveraged Buyouts on Productivity
and Related Aspects of Firm Behavior. Journal of Financial Economics 27:1, 165194.
Ljungqvist, Alexander, and Matthew Richardson. 2003. The Cash Flow, Return and Risk Charac-
teristics of Private Equity. Working Paper, National Bureau of Economic Research.
Long, William F., and David J. Ravenscraft. 1993. LBOs, Debt and R&D Intensity. Strategic Man-
agement Journal 14:S1, 119135.
Modigliani, Franco, and Merton H. Miller. 1958. The Cost of Capital, Corporation Finance and the
Theory of Investment. American Economic Review 48:3, 261297.
Muscarella, Chris J., and Michael R. Vetsuypens. 1990. Efficiency and Organizational Structure: A
Study of Reverse LBOs. Journal of Finance 45:5, 13891413.
The New Yorker. 2012. Private Inequity. January 30. Available at http://www.newyorker.com/
magazine/2012/01/30/private-inequity.
Opler, Tim C. 1992. Operating Performance in Leveraged Buyouts: Evidence from 19851989.
Financial Management 21:1, 2734.
Phalippou, Ludovic, and Oliver Gottschalg. 2009. The Performance of Private Equity Funds.
Review of Financial Studies 22:4, 17471776.
Preqin. 2014. Private Equity Spotlight, May. Available at https://www.preqin.com/docs/
newsletters/pe/Preqin_Private_Equity_Spotlight_May_2014.
Rappaport, Alfred. 1990. The Staying Power of the Public Corporation. Harvard Business Review
68:1, 96104.
Smith, Abbie J. 1990. Corporate Ownership Structure and Performance: The Case of Management
Buyouts. Journal of Financial Economics 27:1, 143164.
Tykvov, Tereza, and Mariela Borell. 2012. Do Private Equity Owners Increase Risk of Financial
Distress and Bankruptcy? Journal of Corporate Finance 18:1, 138150.
Ughetto, Elisa. 2010. Assessing the Contribution to Innovation of Private Equity Investors: A Study
on European Buyouts. Research Policy 39:1, 126140.
Wall Street Journal. 2014. PE Has More Than It Can Spend. June 15. Available at http://online.wsj.
com/articles/private-equity-has-more-than-it-can-spend-1402670650.
Wilson, Nick, and Mike Wright. 2013. Private Equity, Buyouts and Insolvency Risk. Journal of
Business Finance & Accounting 40:78, 949990.
Wright, Mike, Steve Thompson, and Ken Robbie. 1992. Venture Capital and Management-led,
Leveraged Buy-outs: A European Perspective. Journal of Business Venturing 7:1, 4771.
3
Private Equity in the United
States and Europe
Market and Regulatory Developments
A L E X A N D R O S S E R E TA K IS
Teaching and Research Assistant, University of Luxembourg

Introduction
The private equity (PE) industry gained mainstream attention during the U.S.
takeover battles of the 1980s epitomized by the iconic leveraged buyout (LBO)
of RJR Nabisco (Burrough and Helyar 2010). The controversial tactics used by
private equity firms during the takeover boom such as hostile takeovers and the ag-
gressive use of leverage created a negative image of the industry. The public started
to view PE firms as corporate raiders seeking to make quick profits by stripping
companies of their assets. Nonetheless, academics recognized the benefits of the
LBO model in improved corporate governance and operating performance of target
firms leading Jensen (1989) to predict that LBOs would eventually eclipse publicly
held corporations. Regulatory changes, the mild recession of the early 1990s, and
a decline in the availability of credit stopped the first buyout wave (Cheffins and
Armour 2008).
Buyout activity resumed in the late 1990s with LBOs spreading rapidly in Europe
(Wright, Renneboog, Simons, and Scholes 2006; Kaplan and Strmberg 2009).
The bursting of the dot-com bubble in 20002001 severely affected private equity
activity, but the market quickly recovered and entered its most robust period in his-
tory. Between 2004 and 2007, the value and number of LBOs increased exponen-
tially with PE firms completing 7 out of the 10 largest LBOs in history (Gaughan
2011). Some credit the second LBO wave to record levels of capital raised by PE
firms, the abundant liquidity in the financial system, and the growing recognition
among public company chief executive officers (CEOs) of the benefits of going pri-
vate (Kaplan 2007).
The crash of the U.S. housing market in 2007 and the resulting financial crisis of
20072008, which saw freezing credit markets and widespread failures of financial

32
Pr iv ate E qu it y in t h e U.S . an d E u rope 33

intermediaries, caused the collapse of the PE market. Cain, Davidoff, and Macias
(2012) estimate that during 2007 and 2008 the total transaction value of takeover
terminations by PE bidders in the United States reached $168 billion. The sover-
eign debt crisis and the turmoil created in financial markets in the United States and
Europe during the period 20102011 further constrained the ability of PE bidders to
finance takeovers. However, the unprecedented actions of central banks aimed at low-
ering interest rates resulted in rising stock prices and buoyant debt markets boosting
the revival of PE markets in the United States and Europe. PE sponsors took advan-
tage of robust credit markets in 2013 to complete new deals, refinance existing ones,
and cash out their investments by dividend recapitalizations while record high stock
prices made the initial public offering (IPO) exit route attractive (Dezember 2013).
Deal activity remained robust in 2014. 2014 was also a record year for exits via sales
with corporate acquirers utilizing their cash balances to acquire PE portfolio compa-
nies (Canada 2014).
The financial crisis of 20072008 and the deep flaws revealed in the regulatory
design of the financial system prompted a forceful regulatory response by European
Union (EU) and U.S. regulators. Privately organized pools of capital previously
outside the regulatory reach were one of the first targets of regulatory action. The
result was the adoption in 2010 of the Dodd-Frank Wall Street Reform and Con-
sumer Protection Act (Dodd-Frank) in the United States and the Alternative Invest-
ment Fund Managers Directive (AIFM Directive) in the European Union that went
into effect in 2011. The Dodd-Frank Act brings PE firms under the regulatory radar
for the first time, mandating their registration with the Securities and Exchange
Commission (SEC), and disclosure of various types of information. Further, on the
determination of the Financial Stability Oversight Council (FSOC), systemically
important private equity managers and/or funds may be brought under the Federal
Reserves supervision. Overall, the Dodd-Frank Act adopts a measured approach
toward the risks posed by the PE industry having as a primary focus the protection
against systemic risk. According to Caruana (2010), systemic risk is the risk of dis-
ruption to financial services that is caused by an impairment of all or parts of the
financial system and has the potential to have serious negative consequences for
the real economy. In contrast, some forcefully criticize the AIFM Directive, which
regulates both hedge fund and PE fund managers, for its burdensome and restric-
tive terms. The AIFM Directive contains complex provisions aimed at protecting
investors in PE funds and tackling the systemic risks the industry poses to the fi-
nancial system.
The purpose of this chapter is to offer an analysis of market and regulatory develop-
ments in the PE industry after the financial crisis of 20072008. The chapter is divided
in four sections. The first section provides an introduction to PE and the LBO model
and introduces the sources of value creation in buyouts. The next section discusses
market developments in the PE market in the United States and European Union in the
post-financial crisis era. Section three contains an analysis of major regulatory develop-
ments focusing on the Dodd-Frank Act and the AIFM Directive. Section four concludes
by offering an assessment of market and regulatory developments in the PE industry
during the post-financial crisis era.
34 i n t r o d u c t i o n

Leveraged Buyouts and Private Equity: An Overview


PE covers various forms of financing catering to different investors and development
stages of companies. Major types of PE include venture capital, development capital,
mezzanine capital, distressed investments, and LBOs (Metrick 2006). Venture capital
funds extend staged financing to early stage firms using complex financial contracts re-
sponding to the problems of uncertainty, information asymmetry, and potential oppor-
tunism of entrepreneurs (Gilson 2003). Development capital provides growth equity
financing to already established and profitable companies seeking to expand their oper-
ations (Temple 1999). Mezzanine capital involves the provision of subordinated debt
or preferred equity to support expansions, acquisitions, and recapitalizations of compa-
nies, and LBOs (Silbernagel and Vaitkunas 2012). Distressed debt investors invest in
securities of companies undergoing extraordinary situations such as bankruptcy, reor-
ganization, liquidation, or debt restructuring ( Jain 2011).
LBOs are the most well-known PE investments. LBOs acquire an existing and
mature company by a PE firm funded by using equity and high levels of debt (Kaplan
and Strmberg 2009). The debt component of the price offered to buy the target com-
pany ranges from 60 to 90 percent with the rest of the price funded by PE investors
funds. If the target company is a public company, which is subsequently transformed
into a private one, the LBO is defined as a public-to-private transaction. Buyouts of com-
panies initiated by the incumbent management in partnership with a PE firm are termed
management buyouts (MBOs). In case of management buy-ins, an outside management
team with the support of a PE investor leads the acquisition. PE firms lead institutional
buyouts without involving the incumbent management. However, the PE buyers will
retain incumbent management in the post-buyout company compensating it with an
equity stake. PE firms usually seek to avoid buyouts without the consent and coopera-
tion of target management to avoid bidding wars (Cheffins and Armour 2008).
The PE firms orchestrating LBOs raise the capital necessary to execute transactions
by establishing individual PE funds (Fenn, Liang, and Prowse 1995). The funds are
structured as limited partnerships with PE firms serving as general partners (GPs) and
investors who contribute capital becoming limited partners (LPs). As a GP, the PE firm
is responsible for managing the fund, selecting, executing, and monitoring the invest-
ments. GPs are rewarded with an annual management fee usually set at 2 percent of
assets under management and a share in the profits of the fund set at 20 percent, widely
known as carried interest. Further, GPs charge monitoring and transaction fees on the
companies in which they invest. According to Metrick and Yasuda (2010), PE firms
earn most of their revenue by fixed revenue components.
PE funds have a fixed life, usually 10 years during which investors cannot withdraw
their capital. Typically during the first three to five years of the investment period, the
GP identifies promising targets for LBOs and executes acquisitions. During the next
seven to five years, defined as the holding period, the PE firm seeks to maximize the value
of the investments and eventually exit them through an IPO, a secondary buyout of the
target company to another PE buyer, or a sale to a strategic buyer.
LPs are prohibited from managing the fund. Nonetheless, partnership agreements con-
tain specialized provisions seeking to curb potential opportunism of GPs such as restrictions
Pr iv ate E qu it y in t h e U.S . an d E u rope 35

on the total amount a fund can invest in a company and periodic disclosures about the funds
performance (Gompers and Lerner 1996). Also, to ensure aligning interests with their in-
vestors, GPs invest a portion of their own capital in the PE fund (Kaplan and Strmberg
2009). To avoid applying heavy regulations, PE funds are open only to sophisticated inves-
tors. Participation in PE funds is subject to a high minimum subscription (Payne 2011). As
a result, investors in PE are high net worth individuals and institutional investors such as
pension funds, university endowments, insurance companies, and fund of funds.
To exit their investments profitably and reap the associated profits, PE buyers apply
their skills seeking to increase firm value. A major source of value creation is reducing
agency costs between managers and shareholders in widely held corporations (Masulis
and Thomas 2009). The PE investor may obtain majority control of the target company
thus emerging as a major shareholder. Shleifer and Vishny (1986) note the enhanced
incentives of blockholders to discipline and monitor management. As a major share-
holder, the PE investor appoints most of the target companys board of directors. Fur-
thermore, to ensure an alignment of managerial interests, the PE investor compensates
post-buyout executives with a large equity stake requiring them to invest a part of their
personal wealth in the company. As Jensen (1986) notes, the highly indebted struc-
ture of the target companies after the buyout motivates the management team to reduce
costs, operate the company efficiently, and pay out cash flows more than that required
to fund positive net present value (NPV) projects.
Apart from reducing agency costs, PE firms can increase the value of target compa-
nies by applying their industry and operating expertise. Kaplan (1989a) and Harris,
Siegel, and Wright (2005) find that post-buyout firms in the United States and United
Kingdom experience an increase in operating performance. Additionally, Kaplan
(1989b) documents the large tax benefits emanating from the tax deductibility of in-
terest on debt used for financing buyouts. In his study of public-to-private buyouts be-
tween 1980 and 1986, he estimates the use of leverage creates tax benefits ranging from
21 to 143 percent of the premium paid by the PE bidders.

Market Developments in the United


States and Europe
The period between 2003 and 2007 can be described as the golden age of PE both in the
United States and Europe. The availability of cheap debt substantially boosted buyout
activity enabling PE firms to finance multi-billion buyout deals (Axelson, Jenkinson,
Strberg, and Weisbach, 2013). However, the financial crisis of 20072008 and the fol-
lowing sovereign debt crisis severely affected PE activity. Deal volumes declined sharply
while PE bidders experienced severely curtailed access to financing. Nonetheless, PE
activity started to revive in 2012 aided by the extraordinary measures of central banks in
the United States and Europe aimed at boosting credit and the stock markets. Favorable
market conditions continued in 2013 allowing PE bidders to return record amounts of
cash to their investors and complete new acquisitions. Nonetheless, PE activity still has
not recovered to its pre-crisis levels. Furthermore, the nascent recovery of the industry
crucially depends on continuing favorable market conditions in the future.
36 i n t r o d u c t i o n

M A R K E T D E V E L O P M E N T S I N T H E U N I T E D S TAT E S
The U.S. buyout market remains the largest in the world accounting for 54 percent of
global buyout deals and 59 percent of global deal value. The attractiveness of the U.S.
market is based on the presence of large institutional investors and renowned PE firms
(Preqin 2012a). Besides, a deep and liquid stock market provides PE bidders with an
attractive exit opportunity from their investments through an IPO (Black and Gilson
1999). Jerome Kohlberg, Henry Kravis, and George Roberts engineered the modern
LBO model in the United States in the late 1960s while working at Bear Stearns
(Kaufman and Englander 1993). They would later leave Bear Stearns to establish KKR,
which remains one of the largest PE firms in the world.
The exponential growth in buyout activity between 2003 and 2007 came to an abrupt
halt with the onset of the financial crisis of 20072008. Frozen credit markets and the
slump in stock prices forced PE bidders to end or renegotiate their pending acquisitions
and shy away from new deals (Davidoff 2009). The total value of LBOs fell from more
than $400 billion in 2007 to about $100 billion in 2008 falling further in 2009 when the
leveraged buyout market collapsed (Private Equity Growth Capital Council 2013). The
unwillingness of banks to finance LBOs and tightened credit markets forced PE bidders
to increase their equity contribution to more than 50 percent of the total purchase price
in 2009 (Bain and Company 2010). Besides, mega-deals completed during the boom
years such as the $48 billion buyout of TXU, the largest buyout completed so far, had
to be renegotiated with creditors suffering steep losses (Anderson and Creswell 2010).
Buyout activity in 2010 started to recover from its 2009 lows aided by accommo-
dating credit markets (Bain and Company 2011). The benign conditions in financial
markets deteriorated in the second half of 2011 with the escalation of the sovereign debt
crisis. Fears of a disorderly break-up of the Eurozone threw markets into turmoil. Vola-
tile markets and economic uncertainty halted the rebound of LBO activity (Bain and
Company 2012). Nonetheless, U.S. buyout activity accelerated in 2012 despite uncer-
tainty caused by the clash between the two major political parties in the United States
over raising the federal debt limit. Record low-interest rates, investor optimism about
the growth prospects of the U.S. economy, and the stability of U.S. financial markets
spurred the increase in buyout activity (Bain and Company 2013).
Continuing loose monetary policy by the Federal Reserve in 2013 led to buoyant
credit and stock markets with valuations of companies hitting record levels and stock
prices witnessing a spectacular rise. PE firms took advantage of market conditions exit-
ing their investments through IPOs, refinancing the debt of portfolio companies, and
adding debt to companies to fund payouts to themselves (Dezember 2013). As a result,
PE firms could return a record amount of cash to their investors. Also, 2013 saw the
return of PE mega-deals such as the $24 billion buyout of computer maker Dell Inc. by
its founder, Michael Dell, and PE firm Silver Lake Management LLC and the $23 bil-
lion buyout of food company Heinz by Warren Buffett and Brazilian PE firm 3G. Even
though 2014 was notable for the absence of mega-deals, PE activity remained robust
with investors piling into PE funds (Primack 2014). Furthermore, PE firms massively
exited their investments via sales to corporate acquirers (Canada 2014).
Overall, several general trends in the U.S. PE market are worth noting. An important
development is the transformation of the largest PE firms into more broad-based asset
Pr iv ate E qu it y in t h e U.S . an d E u rope 37

management firms. PE firms have expanded their line of business apart from LBOs
into investments in real estate, hedge funds, credit extension, and financial advisory
services (Roumeliotis and Meads 2012). Expanding their product offerings allows
PE firms to deliver superior returns to their investors and to diversify their source of
income and benefit from economies of scale. Also, more PE firms choose to become
or are considering becoming listed on a public stock exchange after Blackstones suc-
cessful IPO in 2007.
Some still criticize the public listing of PE firms for allowing them to raise perma-
nent capital, which makes continuously raising of funds in the market unnecessary and
weakens market discipline ( Jensen 2007). A more widely criticized development is the
increase in secondary buyouts as a means of exiting PE investments. In a secondary
buyout, the PE investor owning the firm sells it to another PE investor. Some criticize
secondary buyouts for aggravating the agency costs between investors and managers.
Toward the end of the investment period, PE managers who usually receive manage-
ment fees on the invested portion of the funds capital have an incentive to burn cash
and invest in deals contrary to the interests of investors. Consistent with this hypothesis,
Degeorge, Martin, and Phalippou (2013) report that secondary buyouts made late in a
funds investment period underperform similar primary buyouts.

MARKET DEVELOPMENTS IN EUROPE


Europe has traditionally been the second biggest market for PE investments after the
United States (Preqin 2012b). Buyout activity in Western Europe between 2000 and
2004 surpassed activity in the United States reaching about 48.9 percent of transaction
value worldwide (Kaplan and Strmberg 2009). European countries show a wide di-
versity in developing their buyout markets (Andres 2012). The United Kingdom, the
first European country to experience a buyout boom during the 1980s, represents the
biggest buyout market in Europe with London serving as a global hub for PE firms and
professionals (Gilligan and Wright 2010). Germany and France are the next most im-
portant markets for buyout activity in Europe distantly following the United Kingdom
(Center for Management and Buyout Research 2013). Spain and Italy also witnessed
large growth in buyout activity after 1996 (Wright et al. 2006). Nonetheless, the finan-
cial crisis of 20072008 put an abrupt end to the rise of LBO activity with deal value
and volume suffering an almost 50 percent drop compared to its pre-crisis level (Mac-
Farlane 2013).
Europe experienced its own buyout boom between 2003 and 2007. Some of the
biggest European buyouts completed during the height of the LBO boom in 2006 and
2007 include the management buyout of the U.K. pharmacy company Alliance Boots,
the leveraged acquisition of TDC (a Danish telecommunications company), and the
sale of Philips Semiconductors (the semiconductor business of Royal Philips Electron-
ics) to a consortium of PE buyers. The financial crisis of 20072008 caused a steep de-
cline in buyout activity across Europe as credit markets froze and risk aversion took
hold. Although buyout activity reached its lowest level in 2009 (Preqin 2012b), it sub-
stantially picked up in 2010 and continued growing in the first half of 2011 despite the
gradual escalation of the sovereign debt crisis (PricewaterhouseCoopers 2012). Market
conditions worsened sharply during the second half of 2011 as fears spread about the
38 i n t r o d u c t i o n

ability of European governments to honor their debt obligations. As a result, the Euro-
pean buyout recovery came to a standstill with buyout activity remaining weak during
2012 especially in southern European countries (Bain and Company 2013).
Signs of stabilization in the Eurozone economy and a revival of credit markets due
to the aggressive monetary policy of the European Central Bank led to a sharp rise in
buyout activity in 2013. Further, a vibrant IPO market allowed PE sponsors to exit
their investments while the availability of credit led to a surge in refinancing and divi-
dend recapitalizations (Husband 2013). The revival of buyouts was strong in Germany
reaching to the levels witnessed before the financial crisis of 20072008 and reflecting
Germanys position as the dominant and most resilient economy in Europe (Pritchard
2013). Furthermore, PE firms are aggressively raising funds for potential buyouts in
southern European countries such as Spain and Italy lured by improving fundamentals
and dispositions of corporate noncore assets (MacFarlane 2013).
On the future of buyout activity in Europe, market participants consider that op-
portunities in Europe lie in areas outside the LBO market such as real estate, consumer
loans, distressed assets, and dispositions of assets by European banks seeking to down-
size their operations due to regulatory requirements and market pressure (Pritchard
2013). Indeed, banks in Europe sold approximately $75 billion of commercial and res-
idential property loans in 2014, with PE firms being among the biggest buyers (Pat-
naude 2015). The sovereign debt crisis and the resulting recession have substantially
impeded buyout activity in Europe. Besides, the tremendous growth of buyout activity
in the Asian-Pacific region and Latin America is threatening Europes long-standing po-
sition as the second most important market globally for PE buyouts (Preqin 2012b).
The growing importance of emerging markets such as Brazil, India, and China in the
world economy has attracted PE investors seeking to capitalize on the growth potential
in these markets.

Regulatory Developments after the Financial Crisis


A complete overhaul of financial regulation in the United States and Europe followed
the financial crisis. Despite the absence of evidence about its contribution to the fi-
nancial crisis, U.S. and European regulators quickly determined that the previously
unregulated and opaque alternative investment funds sector posed a major threat to
financial stability and aggravated systemic risk in the financial markets. The result was
the adoption of the Dodd-Frank Act in the United States and the AIFM Directive
in Europe, regulations that substantially affect the PE industry. The Dodd-Frank Act
brings the PE industry under the regulatory radar forcing PE fund managers to regis-
ter with the SEC for the first time and introduces the Volcker Rule banning banking
entities from investing or sponsoring PE funds. The AIFM Directive seeks to create a
harmonized regulatory framework for alternative investment fund managers operat-
ing in Europe. It requires their registration with a supervisory authority and compli-
ance with various requirements, which seek to ensure the protection of investors and
to tackle the issue of systemic risk. European regulators adopted an interventionist
stance toward the PE industry raising its compliance costs and altering the structure
of PE deals in Europe.
Pr iv ate E qu it y in t h e U.S . an d E u rope 39

R E G U L ATO R Y D E V E L O P M E N T S I N T H E U N I T E D S TAT E S
The financial crisis of 20072008 started in the U.S. subprime market, spread through-
out the financial system, and led to the adoption in 2010 of the Dodd-Frank Act, an am-
bitious effort to overhaul U.S. financial regulation. The Act mainly targets systemic risk.
According to Skeel (2010, p. 4), the Dodd-Frank Act has two primary objectives: its
first objective is to limit the risk of contemporary finance . . . and the second is to limit
the damage caused by the failure of a large financial institution. The Act contains terms
seeking to improve transparency in the PE industry and reduce concerns about the po-
tential contribution of PE to systemic risk. Title IV of the Dodd-Frank Act requires PE
firms to register with the SEC under the Investment Advisers Act of 1940 (hereinafter
called the Advisers Act) and comply with heightened disclosure requirements and
provisions seeking to protect investors in PE funds. Section 619 of the Act, the so-called
Volcker Rule, forbids banking entities from sponsoring or investing in a PE fund subject
to limited exceptions. Finally, systemically important PE firms or funds may be brought
under the supervision of the Federal Reserve on their designation as systemically im-
portant financial institutions by the FSOC. Systemically important financial institutions
are those institutions whose failure could significantly jeopardize financial stability and
adversely impact the real economy.
Title IV abolishes section 203(b)(3) of the Advisers Act, which allowed PE fund
managers to avoid registration as investment advisers with the SEC. Section 203(b)(3)
provided an exemption from registration under the Advisers Act for an investment ad-
viser who had fewer than 15 clients, did not hold itself out to the public as an investment
adviser, and did not serve as an investment adviser to a registered investment company
or business development company. Because of abolishing 203(b)(3), fund managers
who previously relied on this exemption are now required to register with the SEC.
However, the Dodd-Frank exempts from registration advisers to family offices, venture
capital funds, investment advisers advising private funds with less than $150 million
assets under management in the United States, and foreign private advisers. A foreign
private adviser is any investment adviser who has no place of business in the United
States, has fewer than 15 clients and investors in the United States in private funds ad-
vised, and has less than $25 million assets under management invested in private funds
advised by the adviser by clients in the United States and investors in the United States.
Furthermore, the adviser must not hold itself out to the public as an investment adviser
or act as an investment adviser to any registered investment company or a business de-
velopment company.
Section 113 of the Dodd-Frank Act introduces a novel regulatory framework for
nonbank systemically important financial institutions aimed at safeguarding financial
stability. In response to the failures of the previous regulatory regime, which mainly
focused on micro-prudential regulation (i.e., the regulation of individual financial in-
stitutions), the Dodd-Frank Act establishes FSOC to monitor and respond to systemic
risks in U.S. financial markets. The FSOC may designate nonbank financial companies
including PE firms and/or their funds as systemically important financial institutions.
In making such designations, the FSOC considers various factors including the compa-
nys degree of leverage, its size and interconnectedness with the rest of the U.S. financial
system, and the liquidity risk and maturity mismatch between the companys assets and
40 i n t r o d u c t i o n

liabilities. Other factors include whether the company is already subject to regulatory
oversight and whether it is a dominant provider of services in that such a loss of access
to its services could cause financial distress. Once designated as a systemically impor-
tant financial institution, a nonbank financial company is brought under the supervision
of the Federal Reserve Board, which has the authority to develop and impose prudential
standards.
The Volcker Rule introduced by section 619 of the Dodd-Frank Act bans banking en-
tities from sponsoring or investing in PE funds. The definition of sponsorship includes
serving as a GP; managing member or trustee of a fund; selecting or controlling the
funds directors, trustees, or management; or sharing the same name as the fund. Bank-
ing entities are allowed, however, to organize and offer a PE fund with the provision of
bona fide trust, fiduciary, or investment advisory services provided the fund is offered
solely to customers of such services. The investment should not exceed 3 percent of the
outstanding ownership interests in the fund one year after its establishment. Banking
entities are permitted to invest in such funds up to 3 percent of their Tier 1 capital, which
refers to a banks core equity capital composed mainly of common stock and retained
earnings. Regulators may prohibit organizing and offering of PE funds if doing so poses
a threat to the financial stability of the banking entity or involves material conflicts of
interests or results in a material exposure of the banking entity to high risk assets or trad-
ing strategies. Nonbank financial companies sponsoring or investing in PE funds and
designated by the FSOC as systemically important financial institutions may be subject
to additional capital requirements and quantitative limits with respect to such activities.
Overall, regulating PE in the United States is premised on the potential contribu-
tion of the industry to systemic risk. The PE industry can be a source of systemic risk
through the widespread failure of PE-backed companies and its effects on the banking
system, which finances LBOs and the real economy. Nonetheless, no widespread failure
of PE-backed companies occurred during the financial crisis of 20072008 and the fail-
ure of these companies did not jeopardize the real economy. Also, even a comprehensive
study by the European Central Bank (2007) recognized the debt exposures of banks
to the EU leveraged buyout market are sufficiently covered by their capital buffers. Al-
though the PE industry is unlikely to be a source of systemic risk, the registration, and
reporting requirements introduced by Dodd-Frank Act will only have a minor impact
on the PE industry in compliance costs (Kaal 2012). A PE firm or fund is unlikely to
fulfill FSOCs criteria for designation as a systemically important financial institution.
Although the failure of standalone PE firms and funds is unlikely to pose a threat to
the financial system, systemic risk may emanate from banks ownership and sponsor-
ship of PE funds. The failure of internal PE funds may adversely affect the reputational
capital of the parent banking organization and result in its failure, which may destabilize
the financial system if the parent is systemically important. Further, bank-affiliated PE
funds may be able to take advantage of the explicit and implicit government guarantees
of their parent companies to finance their investments at a lower cost. Consistent with
this hypothesis, Fang, Ivashina, and Lerner (2012) find that deals completed by bank-
affiliated PE funds and financed by the parent bank are financed at substantially better
terms than deals completed by standalone funds even though they do not show better
performance. As a result, adopting the Volcker Rule by U.S. regulators is based on a
sound rationale and responds adequately to the systemic risk of internal PE funds.
Pr iv ate E qu it y in t h e U.S . an d E u rope 41

R E G U L ATO R Y D E V E L O P M E N T S I N T H E E U R O P E A N U N I O N
The financial crisis of 20072008 and the failures of the EU financial regulatory frame-
work resulted in an overhaul of EU financial regulation. One of the first targets of Eu-
ropean regulators was the opaque alternative investment fund industry. EU politicians
regularly criticized the PE industry for breaking-up companies, slashing jobs, and
promoting a short-term thinking inside corporate boardrooms at the expense of long-
term value creation. The result was the adoption of the AIFM Directive in November
2010 after a lengthy and heated negotiation. The Directives main goals are protecting
investors in alternative investment funds and tackling systemic risk. The AIFM Direc-
tive seeks to achieve these goals by creating a harmonized EU regulatory framework
for alternative investment funds (AIFs). An AIF is any collective investment undertak-
ing that raises capital from investors for investing it according to a defined investment
policy and does not require authorization under Article 5 of Directive 2009/65/EC,
commonly known as the UCITS Directive.
The AIFM Directive regulates alternative investment fund managers (AIFM) estab-
lished in the European Union that manage AIFs, whether established in the European
Union or not, and non-EU-based AIFMs that manage EU funds or market funds in
the European Union. An AIFM is any entity managing AIFs as a regular business. As
a result, managers of PE funds, hedge funds, commodity funds, and real estate funds
fall within the ambit of the Directive. PE fund managers covered by the Directive are
required to become authorized by the competent authorities of their home Member
States. Nonetheless, the Directive creates an exemption for PE fund managers of un-
leveraged AIFs and does not grant investors redemption rights for five years and whose
assets do not exceed EUR 500 million.
Covered fund managers must comply with modest initial and continuing capi-
tal requirements, devise appropriate risk and liquidity management systems, and im-
plement procedures to identify and manage conflicts of interest that could adversely
affect the funds managed or their investors. To curb excessive risk-taking, the Directive
requires fund managers to adopt sound remuneration policies and introduces remu-
neration restrictions for staff whose activities may adversely affect the risk profile
of the funds managed. Furthermore, the AIFM Directive introduces depositary and val-
uation requirements. A fund manager must appoint a single depositary for each fund
managed that will be responsible for safekeeping the funds assets and monitoring its
cash flows. Additionally, an independent valuation of fund assets must take place at least
once per year.
To increase the transparency of the AIF industry, the AIFM Directive introduces
mandatory reporting requirements toward investors and national supervisors. Fund
managers must make available to investors specific information both before and peri-
odically after their investment in the fund. Fund managers must also produce an annual
audited report for each fund and provide it to the competent national authority and
investors on request. These managers must disclose more information to supervisory
authorities for assessing systemic risk. The disclosures includes the primary markets in
which the fund manager trades, principal exposures, concentrations of each fund man-
aged, the risk profile of the funds managed, and main categories of assets in which the
funds managed are invested.
42 i n t r o d u c t i o n

The AIFM Directive also imposes disclosure obligations at the portfolio company
level. Acquisitions of major holdings in non-listed EU companies above certain thresh-
olds (starting at 10 percent) must be disclosed to national regulatory authorities. Also,
the Directive introduces provisions aimed directly at LBOs of EU companies. If a PE
fund acquires control of a non-listed company (control for a non-listed company is de-
fined as a 30 percent ownership interest or more), the PE fund manager must notify
the company, the shareholders and its regulators of gaining control. A PE fund man-
ager who acquires control of a non-listed company or control of a listed company must
disclose certain information to the company, its shareholders, and its regulator. These
disclosures include the policy for preventing and managing conflicts of interest and the
policy for external and internal communication about the company in particular on
employees. For a listed company, the control is defined by reference to the EU Takeo-
ver Directive and varies between Member States but a substantial number of Member
States defines control as a 30 percent or more ownership interest. Furthermore, in case
of an acquisition of control of a non-listed company, the fund manager must disclose its
intentions on the companys future business and the likely effects on employment. The
fund manager must also disclose information on financing the acquisition of the non-
listed company. The annual reports of a non-listed company controlled by a PE fund or
the annual report of the fund itself must contain a fair review of the development of the
companys business.
Finally, the Directive seeks to protect companies against short-term investment
strategies used by PE investors. The most notable strategy involves depleting the target
companys assets for repaying the debt incurred to finance the acquisition, a practice
commonly referred to as asset stripping. A fund manager who acquires control of a non-
listed or listed EU company shall not for two years after the acquisition facilitate, sup-
port, instruct, or vote in favor of any distribution, capital reduction, share buyback, or
acquisition of own shares by the portfolio company. The restrictions are applicable only
if the distributions made to shareholders would cause net assets to fall below the sub-
scribed capital or would exceed available net profits. The asset stripping prohibitions
substantially affect exits and deal structuring in Europe and limit the options availa-
ble for returning value to PE investors. For instance, dividend recapitalizations and re-
demptions of shares including preference shares granted to the PE investor would be
restricted during the first two years after the acquisition.
As Payne (2011) notes, adopting the AIFM Directive reflected the desire of Eu-
ropean legislators to regulate the hedge fund industry. However, in the general cli-
mate of mistrust and hostility toward the opaque AIFs sector after the financial crisis,
EU regulators decided to extend the application of the AIFM Directive to the PE
industry. The premise underlying the Directive was on the need to improve inves-
tor protection and tackle the systemic risk posed by the AIFs industry including the
PE industry. As previously mentioned, the PE industry is unlikely to be a source of
systemic risk. Also, investors in PE funds are sophisticated market players able to
protect themselves and enter mutually favorable bargains with PE firms. The AIFM
Directive is expected to substantially increase compliance costs for PE firms oper-
ating in Europe (Malcom, Tilden, Wilsdon, Resch, and Xie 2009). Moreover, the
restrictions on distributions to shareholders are likely to have a profound impact on
deal structuring and exits.
Pr iv ate E qu it y in t h e U.S . an d E u rope 43

Summary and Conclusions


In the aftermath of the financial and the sovereign debt crises, the PE industry has un-
dergone a deep transformation. Market and regulatory developments are challenging
the continuous growth of the industry. The effects of these developments will be ad-
verse in Europe. The European banking sector is still recovering from the sovereign
debt crisis with severely curtailed sources of LBO financing. Furthermore, although the
European economy has managed to exit from the recession caused by the sovereign
debt crisis, growth in the real economy is likely to remain sluggish for the coming years.
These adverse market developments are complemented by a particularly interventionist
and burdensome regulatory approach toward the PE industry. In contrast, U.S. regula-
tors have adopted a measured approach toward the risks posed by the PE industry and
have recognized its crucial role in rejuvenating the U.S. economy. Moreover, the U.S.
real economy has been recently showing signs of revival indicating the largest economy
of the world is finally exiting the recession caused by the financial crisis of 20072008.
Finally, as central banks around the world exit the extraordinary measures adopted to
combat the financial and sovereign debt crises, PE investors will faced tighter capital
markets and a more modest increase in stock prices.

Discussion Questions
1. Discuss how actions taken by central banks in the late 2000s and early 2010s in the
United States and Europe that influenced PE activity.
2. Discuss general trends in the U.S. PE industry.
3. Compare the regulatory approach of the United States and the European Union
toward the PE industry.
4. Discuss the future of PE in the United States and Europe.

References
Anderson, Jenny, and Julie Creswell. 2010. For Buyout Kingpins, the TXU Utility Deal Gets
Tricky. New York Times, February 27. Available at http://www.nytimes.com/2010/02/28/
business/energyenvironment/28txu.html?pagewanted=alland_r=0.
Andres, Christian. 2012. Buyouts around the World. In Douglas Cumming, ed., The Oxford Hand-
book of Private Equity, 639667. Oxford: Oxford University Press.
Axelson, Ulf, Tim Jenkinson, Per Strmberg, and Michael S. Weisbach. 2013. Borrow Cheap, Buy
High? The Determinants of Leverage and Pricing in Buyouts. Journal of Finance 68:6,
22232267.
Bain and Company Inc. 2010. Global Private Equity Report. Available at http://www.bain.com/
publications/articles/global-private-equity-report-2010.aspx.
Bain and Company Inc. 2011. Global Private Equity Report. Available at http://www.bain.com/
publications/articles/global-private-equity-report-2011.aspx.
Bain and Company Inc. 2012. Global Private Equity Report. Available at http://www.bain.com/
publications/articles/global-private-equity-report-2012.aspx.
Bain and Company Inc. 2013. Global Private Equity Report. Available at http://www.bain.com/
publications/business-insights/global-private-equity-report.aspx.
44 i n t r o d u c t i o n

Black, Bernard, and Ronald Gilson. 1999. Does Venture Capital Require an Active Stock Market?
Journal of Applied Corporate Finance 11:4, 3648.
Burrough, Bryan, and John Helyar. 2010. Barbarians at the Gate. London: Arrow Books.
Canada, Hillary. For Private Equity, 2014 Was All About that Exit. Wall Street Journal, December 31.
Available at http://blogs.wsj.com/privateequity/2014/12/31/2014-was-all-about-that-exit/.
Cain, Matthew D., Steven N. Davidoff, and Antonio J. Macias. 2012. Broken Promises: The Role of
Reputation in Private Equity Contracting and Strategic Default. Available at http://papers.
ssrn.com/sol3/papers.cfm?abstract_id=1540000.
Caruana, Jaime. 2010. Systemic Risk: How to Deal with It? Bank for International Settlements.
Available at https://www.bis.org/publ/othp08.htm.
Center for Management and Buyout Research. 2013. European Buyouts Report: First Half of
2013. Report, Imperial College Center for Management and Buyout Research, London.
Cheffins, Brian R., and John Armour. 2008. The Eclipse of Private Equity. Delaware Journal of
Corporation Law 33:1, 265.
Davidoff, Steven M. 2009. The Failure of Private Equity. Southern California Law Review 82:3,
482543.
Degeorge, Francois, Jens Martin, and Luc Phalippou. 2013. Agency Costs and Investor Returns in
Private Equity: Consequences for Secondary Buyouts. ECGI Working Paper No. 384.
Dezember, Ryan. 2013. Private Equity Enjoys a Record Year. The Wall Street Journal, December 30.
Available at http://online.wsj.com/news/articles/SB1000142405270230436160457929079
3412549248.
European Central Bank. 2007. Large Banks and Private Equity-Sponsored Leveraged Buyouts in
the EU. Report, Frankfurt am Main, European Central Bank.
Fang, Lily, Victoria Ivashina, and Josh Lerner. 2012. Combining Banking with Private Equity In-
vesting. Working Paper, Harvard Business School.
Fenn, George W., Nellie Liang, and Stephen Prowse, 1995. The Economics of the Private Equity
Market. Staff Studies, Board of Governors of the Federal Reserve System.
Gaughan, Patrick A. 2011. Mergers, Acquisitions, and Corporate Restructurings. Hoboken, NJ: John
Wiley and Sons, Inc.
Gilligan, John, and Mike Wright. 2010. Private Equity Demystified: An Explanatory Guide.
ICAEW Corporate Finance Faculty. Available at http://papers.ssrn.com/sol3/papers.
cfm?abstract_id=1598585.
Gilson, Ronald J. 2003. Engineering a Venture Capital Market: Lessons from the American Expe-
rience. Stanford Law Review 55:4, 10671103.
Gompers, Paul, and Josh Lerner. 1996. The Use of Covenants: An Empirical Analysis of Venture
Partnership Agreement. Journal of Law and Economics 39:2, 463498.
Harris, Richard, Donald S. Siegel, and Mike Wright. 2005. Assessing the Impact of Management
Buyouts on Economic Efficiency: Plant-Level Evidence from the United Kingdom. Review of
Economics and Statistics 87:1, 148153.
Husband, Sarah. 2013. European LBO Deals Return Amid Private Equity Friendly Financing
Market. Forbes, July 16. Available at http://www.forbes.com/sites/spleverage/2013/07/16/
european-lbo-deals-return-amid-private-equity-friendly-financing-market/.
Jain, Sameer. 2011. Investing in Distressed Debt. UBS Alternative Investments. Available at http://
papers.ssrn.com/sol3/papers.cfm?abstract_id=1865378.
Jensen, Michael C. 1986. Agency Costs of Free Cash Flow, Corporate Finance, and Takeovers.
American Economic Review 76:2, 323329.
Jensen, Michael C. 1989. Eclipse of the Public Corporation. Harvard Business Review 67:5, 6174.
Jensen, Michael C. 2007. The Economic Case of Private Equity (and Some Concerns). Harvard
NOM Working Paper No. 0702.
Kaal, Wulf. 2012. Hedge Fund Manager Registration under the Dodd-Frank Act. San Diego Law
Review 50:2, 243323.
Kaplan, Steven N. 1989a. Management Buyouts: Evidence on Taxes as a Source of Value. Journal
of Finance 44:3, 611632.
Pr iv ate E qu it y in t h e U.S . an d E u rope 45

Kaplan Steven N. 1989b. The Effects of Management Buyouts on Operating Performance and
Value. Journal of Financial Economics 24:2, 217254.
Kaplan, Steven N. 2007. Private Equity: Past, Present and Future. Journal of Applied Corporate
Finance 19:3, 816.
Kaplan, Steven N., and Per Strmberg. 2009. Leveraged Buyouts and Private Equity. Journal of
Economic Perspectives 23:1, 121146.
Kaufman, Allen, and Ernest J. Englander. 1993. Kohlberg Kravis Roberts and Co. and the Restruc-
turing of American Capitalism. Business History Review 67:1, 5297.
Macfarlane, Alec. 2013. Private Equity Returns to Spain, Italy. Wall Street Journal, November 25. Avail-
able at http://blogs.wsj.com/privateequity/2013/11/25/private-equity-returns-to-spain-italy/.
Malcom, Kyla, Mark Tilden, Tim Wilsdon, Jessica Resch, and Charles Xie. 2009. Impact of the
Proposed AIFM Directive across Europe. Charles Rivers Associates.
Masulis, Ronald W., and Randall S. Thomas. 2009. Does Private Equity Create Wealth? The Effects
of Private Equity and Derivatives on Corporate Governance. University of Chicago Law Review
76:1, 219261.
Metrick, Andrew. 2006. Venture Capital and the Finance of Innovation. Hoboken, NJ: John Wiley and
Sons, Inc.
Metrick, Andrew, and Ayako Yasuda. 2010. The Economics of Private Equity Funds. Review of Fi-
nancial Studies 23:6, 23032341.
Patnaude, Art. 2015. Loan Firms Boom as Banks Shift Debt in Europe. Wall Street Journal, January
21. Available at http://www.wsj.com/articles/loan-firms-boom-as-banks-shift-debt-in-
europe-1421844087.
Payne, Jennifer. 2011. Private Equity and Its Regulation in Europe. European Business Organiza-
tional Law Review 12:4, 559585.
Preqin Ltd. 2012a. Global Private Equity Report. Available at https://www.preqin.com/docs/
samples/preqin_global_private_equity_report_2012_sample_pages.pdf.
Preqin Ltd. 2012b. Special Report: European Private Equity. Available at https://www.preqin.
com/docs/reports/Preqin_Special_Report_European_Private_Equity.pdf.
PricewaterhouseCoopers, 2012. Private Equity Trend Report: Learning to Live with the New
Reaity. Available at http://www.pwc.de/de_DE/de/finanzinvestoren/assets/pwc-private-
equity-trend-report-2012.pdf.
Primack, Dan. 2015. 2014 Was A Huge Year for M & A and Private Equity. Fortune, January 5.
Available at http://fortune.com/2015/01/05/2014-was-a-huge-year-for-ma-and-private-equity/.
Pritchard, Becky. 2013. Secondary Buyouts Dont Mean Markets Healthy, Says Blackstones
Baratta. Wall Street Journal Private Equity Beat, October 22. Available at http://blogs.wsj.com/
privateequity/2013/10/22/secondary-buyouts-are-not-healthy-says-blackstones-baratta/.
Private Equity Growth Capital Council. 2013. Private Equity Trends. Available at http://www.
pegcc.org/wordpress/wp-content/uploads/2013-Q2-PEGCC-Private-Equity-Trends-Press-
Release-Attachment.pdf.
Roumeliotis, Greg, and Simon Meads. 2012. Private Equity? You Mean Alternative Asset Manager.
Reuters, February 28. Available at http://www.reuters.com/article/2012/02/28/private-equity-
superreturn-idUSL5E8DRAK520120228 http://www.reuters.com/article/2012/02/28/private-
equity-superreturn idUSL5E8DRAK520120228.
Shleifer, Andrei, and Robert W. Vishny. 1986. Large Shareholders and Corporate Control. Journal
of Political Economy 94:3, 461488.
Silbernagel, Corry, and Davis Vaitkunas. 2012 Mezzanine Finance. Bond Capital. Available at
http://www.salvador-montoro.com/uploads/3/2/0/7/3207272/mezzanine_finance_12.pdf.
Skeel, David. 2010. The New Financial Deal: Understanding Dodd-Frank Act and Its (Unintended)
Consequences. Hoboken, NJ: John Wiley and Sons, Inc.
Temple, Peter. 1999. Private Equity: Examining the New Conglomerates of European Business. Hobo-
ken, NJ: John Wiley and Sons, Inc.
Wright, Mike, Luc Renneboog, Tomas Simons, and Luis Scholes. 2006. Leveraged Buyouts in the
U.K. and Continental Europe: Retrospect and Prospect. Journal of Applied Corporate Finance
18:3, 3855.
Part Two

MAJOR TYPES OF PRIVATE EQUITY


4
Venture Capital in Europe
Myths and Reality
JEAN-MICHEL SAHUT
Professor, IPAG Business School

ERIC BRAUNE
Assistant Professor, INSEEC Business School

Introduction
During the past three decades, the U.S. venture capital (VC) industry has been the sub-
ject of substantial research. Many authors highlight the dynamics of innovation initiated
by massive funding of start-ups; others explain the financial performance of this indus-
try. Brown, Fazzari, and Petersen (2009) attribute 75 percent of the 1990s technology
boom to the massive growth of the offer of funding toward young, innovative compa-
nies during this period. According to Kortum and Lerner (2000), VC, which averaged
less than 3 percent of expenditures for research and development (R&D) firms between
1983 and 1992, was responsible for 10 percent of U.S. industrial innovations. Kaplan
and Schoar (2005) show that VC funds performance surpassed that of the S&P 500
index between 1980 and 2001.
The stream of innovation generated by U.S. VC financing as well as financial perfor-
mance recorded by this industry led to internationalization of the VC practices (Cressy
2006). Although the United States remains at the forefront of VC-backed innovation,
Europe is the second-largest venture hub for fund raising in the world. For their part,
Chinas and Indias VC industries continue their rapid growth as they capitalize on gross
domestic product (GDP) growth, growing domestic consumption, and a dynamic en-
trepreneurial ecosystem (Ernst & Young 2011).
However, the performance of this mode of funding outside the United States is still
questioned. In particular, the European VC industry performance has been widely criti-
cized (Hege, Palomino, and Schweinbacher 2009). As Bottazzi, Da Rin, and Hellmann
(2004) note, anecdotal evidence did not support this mode of financing outside the
United States and was prone to defend the VC industry as a part of the American culture
(Patricof 1989). Nevertheless, the European VC industry remains vigorous and Euro-
pean practices show a higher professionalism.

49
50 m a j o r t y p e s o f p r i vat e e q u i t y

The purpose of this chapter is to highlight the recent evolutions of practices and per-
formances of the European VC industry. A limitation of this chapter relates to the as-
sumption that Europe is a homogeneous economic area, when it is not. The remainder
of this chapter has the following organization. The first section provides an overview of
the European VC industry highlighting new aspects of the European VC practices. The
next section examines the performance of European VC financing and the convergence
of U.S. and European performances and practices. The final section offers a summary
and conclusions.

The Professionalization of European Venture


CapitalPractices
The European VC industry is still a young industry. Bottazzi et al. (2004) estimate 1998
as the median starting date of European VC firms. The following sections examine the
maturity and the professionalization of these firms as well as the evolving professional-
ism of practices used by European VC firms.

T H E E U R O P E A N V E N T U R E C A P I TA L I N D U S T R Y
The European Private Equity and Venture Capital Association (EVCA) (2012) reports
creating 102 VC funds in 2012 for 3.6 billion. Thus, 952 funds managed by 556 firms
have ensured VC financing of 2,923 companies in 2012. European VC investors are a
stable community dominated by European and national public players. Major European
industrial companies as well as insurance companies have also consolidated their posi-
tions in this industry. The stability of European investors shapes the VC industry in this
region. This stability has both positive and negative aspects.

Positive Aspects of the European Venture Capital Investors Stability


The stability of the European VC actors has several advantages. First, the continuity of
public action guarantees the regular intake of new funds. Government agencies pro-
vided 40 percent of funds raised in 2012 of which more than 9 percent came from the
European Investment Fund. The long-term mission entrusted to public funders makes
injections of additional funding insensitive to the previously recorded results. Thus,
funding from VC appears likely at least in the long run.
As Hochberg, Ljungqvist, and Lu (2007) note, the experience gained by the VC
investors community allows them to more effectively select the projects to be financed.
In fact, the EVCA reports the share of the funding devoted to seed or start-up projects
increased continuously since 1997. Since 1999, the European share of seed or start-up
projects has been greater than in the United States (Bottazzi and Da Rin 2002). In 2012,
seed or start-up investments amounted to more than 60 percent of VC investments in
Europe. Obviously, European investors are now able to identify projects earlier that are
likely to lead to marketable innovations and to invest in them.
Ghemawat and Khanna (1998) point out that suppliers permanently engaged in the
same market are more likely to recognize their dependence. This recognition is a prereq-
uisite for increasing the rate of VC investment syndication. Sorenson and Stuart (2001),
Ve n t u re C apit al in E u rope 51

Abell and Nisar (2007), and Hochberg et al. (2007) show that syndicating investments
positively affects fund performance. Therefore, creating a stable community of funders
should positively influence VC fund performance in Europe.

Negative Aspects of the European Venture Capital Investors Stability


The European VC industry lies at the conjunction of public policies and the interests of
major European industrial companies. Public policies are intended to orient innovation
toward some sectors whose development European governments deem strategic. Euro-
pean industrial firms look for innovation to improve the attractiveness of their products
(Chesbrough 2006). The industry structure for the funding projects industry shows little
change. European VC is oriented toward financing ventures related to the life sciences,
computers, consumer electronics, and communications. Thus, while Gompers, Kovner,
Lerner, and Scharfstein (2010) underscore the rapid changes of the destination of invest-
ments of VC in the United States to quickly capture the best investment opportunities,
the European venture capitalists seem tied to the public policy objectives that dominate
the investments performance. As such, investments allocated to the energy and environ-
ment sector decreased in Europe for the past four years while Ernst & Young report
increasing investments in this sector in other regions. Thus, the major place occupied
by public authorities in European VC industry may transform stability into rigidity that
impedes the investments in the more promising sectors and hampers fund performance.
Although the stable community of European VC investors appears unreceptive to
arriving new players, the share of investment from other parts of the world is changing
little. The United States is the first foreign investor in VC in Europe but accounts for
only 4.5 percent of the investments. Asia and the Australia represent a mere 0.3 per-
cent of the amount invested in European ventures. Schwienbacher (2002) shows the
financial instruments used in Europe are less sophisticated than in the rest of the world.
In particular, the European financiers use substantially less convertible securities than
their U.S. counterparts (60 percent vs. 80 percent). As Kaplan, Martel, and Strmberg
(2007) show, using convertible securities increases the ventures change of success.
According to Hege et al. (2009), the greater expertise and the more sophisticated
approaches to contracting of U.S. venture capitalists are not easily leveraged when these
one invest in Europe. U.S. Venture capitalists investing in Europe do not outperform
their home-grown counterparts, which may be because local investors are resisting the
adoption of more effective behaviors and thus impeding the evolution of VC in Europe.

SERIAL ENTREPRENEURS AND EXPERIENCED VENTURE


C A P I TA L I N V E S TO R S
The previous section highlighted the interests and limitations related to establishing a
stable community of VC investors in Europe. This section examines two aspects of the
professionalization practices in Europe. The first involves emerging professional entre-
preneurs in Europe and the second concerns professionalizing VC investors practices.

Serial Entrepreneurship in Europe


Developing the VC industry is directly related to the number of potential entrepre-
neurs (Hellmann 1998). Gompers and Lerner (2000) show that VC funds periodically
52 m a j o r t y p e s o f p r i vat e e q u i t y

compete for a limited number of attractive projects. As Kaplan and Schoar (2005) show,
the successive funds of the most efficient VC investors grow less quickly than those of
the other VC firms due to the low number of good projects to fund. In each of these
studies, VC funding is constrained on the demand side: the number of good projects
remains low and funds are always looking for promising companies to finance. In this
context, the emergence of a class of serial entrepreneurs likely to multiply the number of
successful ventures seems an important determinant of the growth of the VC industry
in the United States and Europe.
For both Gompers et al. (2010) and Rdis and Sahut (2013), the ventures man-
aged by serial entrepreneurs are more likely to succeed than those managed by other
entrepreneurs. Studying the innovative cluster of Cambridge in the United Kingdom,
Myint, Vyakarnam, and New (2005) suggest that social capital developed by serial
entrepreneurs contributes to the growth of this innovative cluster. Axelson and Marti-
novic (2013) show that success rates for serial entrepreneurs are higher both in Europe
and the United States. These authors also highlight the absence of difference between
Europe and the United States: the European serial entrepreneurs perform as well as
their American counterparts.
Disparities between the percentages of serial entrepreneurs in European countries
remain large. In the United Kingdom, serial entrepreneurs account for 19 to 25 percent
of entrepreneurs (Westhead, Ucbasaran, and Wright 2005). In Germany, 18 percent of
entrepreneurs are serial (Wagner 2003) and in Finland it is 30 percent (Hyytinen and
Ilmakunnas 2007). As the serial entrepreneurs account for only one-eighth of the entre-
preneurs in the United States (Headd 2003), a reasonable conclusion is that entrepre-
neurship is becoming more professional in Europe.

Professionalizing European Venture Capital Investors Practices


European venture capitalists have the reputation of being conservative and not inter-
fering in the management of companies they finance. However, Bottazzi et al. (2004)
point out that European investors are becoming increasingly involved in the compa-
nies management. For example, 68 percent of them sit on the board of directors and 69
percent of European VC investors monitor their companies monthly or weekly. These
authors also highlight the new profile of European venture capitalists. They are, on av-
erage, 42 years old and have seven years of work experience as venture capitalists. Most
have prior experience in the financial or in the corporate sector and 34 percent of Eu-
ropean venture capitalists have some work experience in the United States. The inter-
national profile of the European venture capitalists facilitates cross-border investments
that account for more than 25 percent of the investments total amount. This investment
requires overcoming the constraints of legal, administrative, and tax regulations that
differ depending on the country. Thus, the share of cross-border investments is also a
sign of professionalization of VC practices.
De Prijcker, Manigart, Wright, and De Maeseneire (2012) point out the importance
of experiential and inherited knowledge to overcome problems inherent to cross-border
investments. As Hege et al. (2009) show, European venture capitalists could overcome
these constraints that do not affect fund performance. These authors report that venture
investments based on European common law and European civil law have a comparable
value creation and the differences in the tax treatment or the legal environment among
Ve n t u re C apit al in E u rope 53

European countries have no effect on value creation. Finally, Hege et al. (2009) show
that U.S. venture funds investing in Europe do not outperform their European counter-
parts. In other words, when investing at home, European venture capitalists are as good
as their American peers.
Professionalizing practices can also be seen in the project selection process. Bottazzi
et al. (2004) report that while the average number of projects increased over time, the
number of those financed remains stable. Thus, European venture capitalists learned
to avoid bad projects over time and to select projects more efficiently. These reasons
explain the convergence of U.S. and European VC practices. As Krussl and Krause
(2013) report, the number of European companies financed by VC is similar to that of
the United States. These authors also note an increase in the successful exit rate by initial
public offerings (IPOs) in Europe is now at par with the U.S. rate.

Venture Capital Funding and Comparison: Europe


andthe United States
Understanding the basis of VC funding is necessary before comparing the performance
between European and U.S. VC. This section examines VC funds performance, the
J-curve effect, and the persistence of their performance.

V E N T U R E C A P I TA L F U N D I N G
Studies of PE funds performance can be classified into two categories. The first cate-
gory concentrates on studying the investment performance of individual assets held by
the fund. The second category examines how the funds as a whole perform. Use of this
second approach is more common because data are more easily accessible and accurate.

Performance of Investments
To evaluate investment performance, Woodward and Hall (2003) and Hwang, Quigley,
and Woodward (2005) build indexes that they use to calculate the correlation between
their index and a market index based on new funding rounds, IPOs, and acquisitions.
The common problem with these studies is that the authors consider only successful
exits. Moreover, mostly the observations are quarterly.
Cochrane (2005), who provides the leading study on the subject, uses an original
approach to correct for selection bias. He supposes the change in the logarithm of the
investment value follows a log normal distribution and the probability of observing a
new funding round follows a logic that depends on the firms value. He uses the maxi-
mum likelihood approach to calculate the alpha and beta of firms. Using the Venture-
One database, Cochrane values 7,765 U.S. companies between January 1987 and June
2000, for 16,613 funding rounds totaling $112 billion. He supplements this database
with other statistics on the financial results of IPOs and mergers and acquisitions show-
ing projects making successful exits. He calculates the return by measuring the value
created between a financing round and the VC exit, whether successful (e.g., IPO and
trade sales) or unsuccessful (e.g., gone out of business). Excluding the returns between
54 m a j o r t y p e s o f p r i vat e e q u i t y

intermediary rounds provides a more reliable measurement of value creation from VC.
However, this increased reliability is offset by the existence of a powerful selection bias
because the exit of the VC is overwhelmingly associated with success.
The return distribution of these 3,595 VC exits has an arithmetic mean return of 698
percent over the period with a high standard deviation. Based on a log normal distribu-
tion, the mean log return is 108 percent with a standard deviation of 135 percent. The
distribution of the non-annualized log returns depends little on the projects age, thus
testifying to the exit strategy used by the funds. A successful exit occurs when the mul-
tiple value creation exceeds a threshold.
Cochrane (2005) uses this multiple rule to correct for selection bias and thus to
estimate the distribution of (log) returns on all projects. After correcting for selec-
tion bias, the log returns are more reasonable. The mean annualized log return is 15
percent, which brings it more in line with the 15.9 percent of annualized log return
from the S&P 500 index. Idiosyncratic volatility among projects is high: the standard
deviation of log returns reaches 89 percent, far above that of 14.9 percent for the S&P
500 index. The high idiosyncratic volatility pushes the mean annualized arithmetic
return to an elevated level of 59 percent, far higher than the mean return of the S&P
500 index over the same period. The VC funded project asset is unlike average listed
assets as it has a slight chance of generating a huge return. Cochrane finds a beta of 1.7
and an alpha of 32 percent net of management fees. He concludes the rates of return
are highly volatile and that investments nearing exit have a lower volatility than those
at the early stage.

Performance of Funds
In this second category, researchers look for how the funds as a whole perform. A funds
performance depends on the strategy of its manager and on the proportion of the assets
invested in unlisted companies. The higher this proportion is, the more the fund will
experience a J curve effect in the first years.
This phenomenon describes funds investing in unlisted companies as the process of
managing such funds is broken down into two phases: (1) the investment phase (find-
ing and investing in the companies) and (2) the realization phase consisting of reselling
the portfolio of companies such as by industrial transfer or IPO. The first phrase can
last more than five years for funds that have a lifespan of 10 years. Fund performance is
generally negative for the first years and then grows exponentially once the capital gains
released by the portfolio cover the management costs. By comparison, a fund investing
the major part of its assets in listed companies (a fund that is 100 percent listed) does
not experience this J curve, but the fund is exposed to the variations of the market
throughout its lifespan. Assuming the market is bullish and the investment in the un-
listed companies achieves a return above that of the market, the performance of the 100
percent listed funds will be higher in the short-term, but lower in the long term than the
100 percent unlisted funds. Mixed funds with 60 percent unlisted companies (including
60 percent listed companies and 40 percent unlisted companies) have an intermediate
performance profile and are less influenced by the J curve effect.
Figure 4.1 shows the J-curve effect. This effect reflects the pattern that in the early
years of a fund, the book value of investments will show losses and decline, then in fol-
lowing years (usually three or more years), the funds value rises.
Ve n t u re C apit al in E u rope 55

Returns (%)

Smoothed returns

0%

Actual returns

1 2 3 4 5 6 7 8 9 10
Year

Figure 4.1 Fund Return and the J-Curve Effect A PE fund goes through many
stages during its life and its real return can be calculated only at the end of its life. For
example, imagine a life of 10 years and compare its return to the return of a risk free
investment with compound interest (called smoothed returns in the graphic). During
its first few years, the PE fund typically has negative returns due to management fees
and initial investment costs. During this early stage, the majority of the portfolio return
is unrealized. Over time, the portfolio returns become realized. As value is added,
improvements to portfolio investments lead to higher valuations. At a certain point in time
(here three years), the net cash flow position turns positive (value of investments minus
costs), and the funds value rises. After a certain period (five years in this case), the fund
reaches its break-even point (the actual value of the fund is equal to its initial value, and
the actual return is equal to zero). At the end of its life, the fund should have a positive
accumulated net cash position, and the actual return is above the smoothed return.
Source: Adapted from Meyer and Mathonet (2005).

Gompers and Lerner (1997) examine 78 PE funds. They adjust each funds perfor-
mance relative to the market and to each investment. Gompers and Lerner then regress
all portfolio values relative to a series of factors to calculate the funds performance. This
approach would get closer to a true risk measure, but only so far as the risk of the firms
being marked-to-market is similar to that of the indexes used to mark them.
Jones and Rhodes-Kropf (2003) introduce and test a model in which the principal
agent problem results in excess returns from funds that increase with systematic risk.
The authors find a positive alpha, which is not statistically significant at normal levels.
However, the alpha estimates are skewed because they are calculated using quarterly
data. Residual values are also determined at the discretion of the general partner (Blay-
don and Horvath 2003), referred to as GP1, and are mainly equal to the sums invested.
As Kaplan and Schoar (2005) show, the average net profitability of PE equity funds
in the United States is 5 percent higher than the average profitability of the S&P 500
index between 1980 and 2001. The authors calculate the profitability of these PE funds
after fund managers have been compensated (about 20 percent of carried interest
56 m a j o r t y p e s o f p r i vat e e q u i t y

and1.5 to 2.5 percent of the managed funds in management fees), which shows perfor-
mance well above that of funds invested in listed shares. Note that carried interest is the
part of profits the general partner (GP) receives as compensation, despite not contrib-
uting any initial funds. This method of compensation seeks to motivate the GP to work
toward improving the funds performance.
Ljungqvist and Richardson (2003) analyze the process of investment from the per-
spective of the GP by concentrating on the sums invested versus sums distributed.
Although they find that PE funds outperform the market, their sample is relatively
small. Moreover, the authors omit VC funds from their sample, which generally have
an average performance that is much lower than PE funds according to Kaplan and
Schoar (2005).
Kaplan and Schoar (2005) also try to assess the net return investors receive over the
funds lifespan. They use a broad sample of mature U.S. funds set up between 1980 and
1997. Their data from Venture Economics cover 746 funds operating in the VC and
buyout segments having an identified GP.
For each of these funds, Kaplan and Schoar (2005) have cash-flow records be-
tween limited partners (LPs) and general partners (GPs) until 2001, as well as the
residual value of the funds when the latter is inactive. For liquidated funds, they
calculate the return based on payments made during the investment horizon. For
inactive funds, the residual value is regarded as a cash flow from the last date. In-
stead of using an internal rate of return (IRR), Kaplan and Schoar measure the net
performance by a profitability or public market equivalent (PME) index. This index
compares the funds performance with that resulting from an investment, using a
timetable of equivalent cash flows, in an S&P 500 index-linked asset. The average
index (weighted by the funds committed capital) calculated on all the funds is 1.05,
which shows that PE outperforms the market. For funds with identical life spans,
investing 1 euro in a PE fund would, on average, be as profitable as investing 1.05
euro in an asset listed on the S&P 500 index. The average profitability from the VC
segment would be appreciably higher than that of the buyout segment with a PME
index of 1.21 compared with 0.93.
In annual terms, the gap between the average net return from PE and the return from
listed investments is positive but small. This result is surprising when considering the
specific features of the PE asset: risks linked to the agency relationship between the GP
and LPs, the nature of the projects funded, the level of debt leverage/equities of buyout
transactions, and the illiquidity of the investment. This small yield gap contradicts the
often more flattering level of returns announced by the media or the industry.
Artus (2008) analyzes the comparative returns of private and the public equity on
the U.S. and European markets over the periods 1995 to 2006 and 1996 to 2006, respec-
tively. Using a different method from Kaplan and Schoar (2005), Artus calculates the
aggregated returns from PE quarter after quarter considering the balance of cash flows
during the period and the differences in net asset value (NAV) of the funds between
the beginning and the end of the period. Evaluating the NAVs reported by the funds
is an estimated accounting procedure, which could be thought to smooth changes to
the true fund value. With this method, the annual net yield gap favoring PE over listed
assets reaches 6.99 percent in the United States and 8.29 percent in Europe. Consider-
ing the volatilities and correlation between the returns of the two categories of assets,
Ve n t u re C apit al in E u rope 57

Artus estimates that the proportion of PE held by investors is below the optimal level
resulting from a model of portfolio choice.
In summary, Kaplan and Schoar (2005) and Artus (2008) report conflicting results
about the aggregate performance of PE assets. Kaplan and Schoar (2005) concentrate
on the long-term returns and take into account the real cash flows distributed by funds.
In contrast, Artus computes a short-term return calculated period after starting from
accounting valuations (NAVs) of the fund assets.
Conversely, Artus and Teletche (2004) show that a smoothing bias resulting from
the methods used by the funds to value their net assets affects the accounting measure-
ment of the return, known as the time weighted return (TWR) based on the funds NAV
report used by the industry. The results from Kaserer and Diller (2004) and Kaplan and
Schoar (2005) using European data show that this bias affects not only the temporal
profile of a funds returns but also the pooled weighted return, which is calculated for
each period.
Short-term returns from PE make little sense when considering the assets illiquid-
ity. For an investor, the decision to add PE to a portfolio involves a commitment and is
therefore based on examining the funds long-term returns. Only an approach, such as
that of Kaplan and Schoar (2005), based on the records of actual cash flows offers solid
information on the returns. As an investment realization is a rare event, one can under-
stand both the difficulty for the analyst and the prudence of the investor.
As Gottschalg and Phalippou (2009) show, various biases affect measuring average
net return. Thus, their observation of a lower average performance (net of remunera-
tions) of PE than that obtained by an equivalent investment in listed shares is difficult to
contest. Gottschalg and Phalippou use data from Thomson Venture Economics (TVE)
resembling that of Kaplan and Schoar (2005). These data include 852 U.S. and non-U.S.
mature funds set up between 1980 and 1993, which cover 57 percent of the amounts in-
vested in the world and for which cash flow data are available until 2003. In this sample,
the average IRR (weighted by the size of the funds) given to investors is 15.2 percent
and the average profitability index (still weighted by committed capital) is 1.01. The
authors make a correction of aggregation by calculating weightings in terms of amounts
actually invested (discounted value of payments made by investors). This makes aggre-
gating the profitability indexes more transparent. Gottschalg and Phalippou report the
aggregate profitability index to be 0.99.
The data collected by TVE from the funds have a double defect. First, the sample con-
tains funds described as living dead, having exceeded the age of liquidation, not show-
ing any sign of activity, but which have nevertheless been given a residual positive net
value (29). When no longer considering this residual value as a final cash flow, the PME
index drops from 0.99 to 0.92. Second, by comparing the TVE data with the larger Ven-
tureXpert sample, Gottschalg and Phalippou (2009) notice the sample overrepresents
funds having experienced profitable investment exits (IPO or trade sales) because these
funds are also the best performers. By exploiting the relationship between performance
and the rate of profitable exits in the core sample, they extrapolate the performance using
a larger sample, which further lowers the PME by 0.04 to 0.88. After correcting for this
bias and adding it to an annual yield gap, the difference between private equity/public
equity would be around 3 percent against PE. This constitutes significant underperfor-
mance, which might be considered as the first ingredient of an enigma of PE returns.
58 m a j o r t y p e s o f p r i vat e e q u i t y

Phalippou and Zollo (2005) find an IRR of 16 percent and a profitability index of
1.05. The sample consists of 983 U.S. PE funds between 1980 and 1996. Excluding the
funds that have not yet been liquidated boosts the results. Moreover, the funds having a
weak performance will be tempted to artificially increase their IRR. Therefore, the deci-
sion to liquidate is endogenous and influenced by successful investments.
Leading market indexes published by TVE influence the results involving studies
on the performance of PE funds. The approach used to assess the performance of these
assets overestimates the funds performance. The method of assessment involves ag-
gregating the funds IRR that does not consider the funds variable life span. Funds with
a long life span have greater weight compared to other funds. The characteristics of the
database used for the statistics present a problem on two levels. First, the residual values
(i.e., investments that have not been realized but kept in the portfolio) are treated as
future cash flows and inflate performance. Second, the standards used for publishing
statistics overrepresent the best performing funds.
As Gottschalg and Phalippou (2009) note, the samples chosen as industry bench-
marks include assets with above-average performance. Using the approach employed by
TVE, the average performance of the 1,328 funds studied results in an IRR of 15.2 per-
cent. However, this rate only vaguely reflects the reality of the true return on investment.
The authors suggest using a more reliable assessment method, namely, the profitability
index, which is the current value of the cash flows received by investors divided by the
current value of the capital paid by the investors. After correcting for the bias relating to
the sample, the performance levels are on average 3 percentage points higher than those
of the stock markets. Moreover, the fees received by the managers sharply reduce inves-
tors profits. Thus, with an average rate of annual management fee at 6 percent, PE funds
offer a performance of 3 percentage points lower than stock markets.

The Persistence of Fund Performance


This section examines the main PE investment vehicles on the French market. It not
only explains the operating process and the characteristics of each fund type but also
examines mutual fund performance for investments in innovative enterprises to better
understand how these funds performed over time. Bursting the dot.com bubble in
March 2000 had a negative effect on such performance.
Kaplan and Schoar (2005), who report on the persistence of fund net returns, reveal
a potential lack of competition that gives an advantage to investors with longer experi-
ence. By having priority access to the most profitable investments, such investors create
barriers against new GPs entering the market. On this point, the empirical evidence pre-
sented by Kaplan and Schoar is ambiguous: the arrival of new funds effectively lowers
the performance of funds that are already present, especially if the latter funds are young.
More established funds appear less sensitive to the entry of competitors, especially in
the VC segment. Above all, such persistence relates to the funds net performance. Unhin-
dered competition among GPs in relation to investors should lead to the disappearance
of differences in anticipated net returns by considering the observation of track records.
The GPs remuneration should include systematic compensation due to talent and
decreasing returns. Persistence should affect only gross returns, not net returns. Due to
a lack of data on remunerations, Kaplan and Schoar (2005) cannot examine the division
of compensation between the GP and LPs. However, heterogeneity and persistence
Ve n t u re C apit al in E u rope 59

characterizing the distribution of net returns show that the relationship formed be-
tween the GP and LPs when constituting a fund concerns more a process of frictional
matching than a transaction in a perfect competition market.
The study by Lerner, Schoar, and Wongsunwai (2007) confirms this point. Using cross
data on returns from LPs and GPs, these authors show the investors net return depends
on the nature of the latter. Over the two last decades, universities and foundations (endow-
ments) earned an annual rate of return on their investments of 14 percent above that of the
average investor. Banks and investment advisers have the lowest performance among inves-
tors. The presence of a high quality investor in a fund thus increases its net performance. The
authors show the LPs market experience is a determining factor of performance. Lerner et al.
(2007) conclude that the behavior of LPs (i.e., their ability to use their previous experience
to select not only the funds but also the funds investment plans) is an essential part of perfor-
mance. They also note that when inexperienced LPs enter the industry during a boom, the
industrys cycle is accentuated. Thus, the match between LPs and the GP appears frictional,
which justifies a process of sharing of compensation between the two sides of the match.
Investing in PE may also occur for other reasons than realizing a direct return from
the operation. A bank can gain extra income by taking part in syndication and debt
management operations linked to the buyout. The nature of competition and the adjust-
ment between return and quantity are also subjects involving a vast amount of writing
on the cyclical character of the industry. How do the intrinsic characteristics of the in-
dustry and the competition contribute to accentuate the cycle?
Gompers and Lerner (2000) highlight the phenomenon of the money chasing
deal. They show that during boom times the surge of capital runs up against the re-
stricted number of investment opportunities increasing the value of these opportunities
and likely decreasing returns.
Kaplan and Schoar (2005) show that, with time, high net performance attracts
new GPs entering the market who raise large funds. These first funds, created after a
boom, do not perform well and are thus unlikely to be followed by a second fund from
the same GP. Remembering the best performing GPs limit the growth of their funds,
Kaplan and Schoar conclude the marginal dollar invested during a boom mainly goes
to the new GPs, who will be less able to create new funds. The growth of the industry is
accompanied by a decrease in average performances of the funds, which progressively
deflates the boom and propels the cycle.

C O M PA R I S O N O F V E N T U R E C A P I TA L P E R F O R M A N C E
I N E U R O P E A N D T H E U N I T E D S TAT E S
As previously discussed, the European VC industry has experienced improved per-
formance based on increased experience. This section compares VC performance in
Europe and the United States followed by questioning whether a convergence of prac-
tices has occurred on both sides of the Atlantic.

Comparative Performance of Venture Capital Funds in Europe


and the United States
Much research focuses on VC funds performance and shows that these funds
have achieved overall satisfactory levels of profitability. However, researchers lack a
60 m a j o r t y p e s o f p r i vat e e q u i t y

source of truly independent information to evaluate the profitability of these funds.


Hence, they rely either on measuring the degree of satisfaction of the fund managers
about the profitability achieved by their investments or refer to databases whose infor-
mation is based on voluntary reports of VC professionals. Kaplan and Schoar (2005)
follow this approach and few authors have sought to overcome this constraint.
Nonetheless, Medjad, Krieger, Gerasymenko, Grandsart, and Iselin (2011) explain
the importance of the gap between the IRR estimated by investors and the actual IRR
of VC funds. In particular, their study of 30 French VC investors shows that these
overestimate the average IRR of the ventures portfolios by at least 10 percent. Because
voluntary reports of VC investors appear biased, caution is necessary when analyz-
ing the IRR of funds. As a result, Hochberg et al. (2007) propose approximating the
fund performance by using the fraction of the funds portfolio companies sold or in-
troduced on the stock exchange. Using this fund performance proxy, despite being
noisy, allows investor cognitive biases to be overcome. Given that the number of
ventures financed by VC in Europe and the United States is similar, the comparison of
exit rates occurs on an analogous basis giving credence to this approach. Finally, this
proxy also enables fund performance to be distinguished in terms of IPOs and trade
sales. Therefore, it allows a detailed analysis of VC performance in Europe and the
United States.
Recent studies by Hege et al. (2009), Axelson and Martinovic (2013), and Krussl
and Krause (2013) show that IPO exit rates in the United States and Europe are com-
parable. These studies have complementary aspects that merit further examination.
These findings are particularly noteworthy because Gompers and Lerner (1999) and
Schwienbacher (2002) show that IPO exits provide higher returns when compared to
other exit strategies.
Axelson and Martinovic (2013) show that IPO exit rates dropped sharply in the
United States and Europe since the high-tech values bubble burst in 2001. Thus, the
fraction of deals for a given investment year that subsequently underwent an IPO de-
creased from 15 percent in 1997 to 4 percent in 2004. According to these authors, the
probability of an IPO exit within a certain time frame from the first round of VC financ-
ing has declined substantially regardless of the time interval considered. Thus, the prob-
ability of exit by an IPO in the 10 years after the date of the passage of the first funding
in the United States and Europe decreased from 12.5 percent (12.2 percent in Europe)
between 1995 and 1999 to 3.1 percent (4.4 percent in Europe) between 2000 and 2003.
Similarly, the probability of exit by an IPO in five years declined from 10.2 percent (8.3
percent in Europe) between 1995 and1999 to 1.4 percent (2.9 percent in Europe) be-
tween 2000 and 2003. Finally, the probability of an IPO exit in the two years after the
first round of financing declined from 4.8 percent (5.1 percent in Europe) to 0.2 per-
cent (1.2 percent in Europe) between the first and second reporting period. Reconcil-
ing IPO exit rates in the United States and Europe is therefore not an effect of portfolio
structure of the companies financed. As Axelson and Martinovic (2013) show, IPO exit
rates in the United States and Europe are close and fell sharply regardless of the duration
of the financing by VC or the relative maturity of the funded company.
Krussl and Krause (2013) explain the disparities of IPO exit rates by industry be-
tween Europe and the United States. During 2000 and 2009, European IPO exit rates
were higher than U.S. rates in three industries: computer hardware and software (5.95
Ve n t u re C apit al in E u rope 61

percent vs. 2.69 percent), semi-conductors and other electronics (7.07 percent vs. 4.82
percent), and communications and media (6.24 percent vs. 4.44 percent). However,
the United States outperformed Europe for the most promising industries: biotech-
nology (11.15 percent vs. 9.01 percent) and medical, health, and life sciences (9.57
percent vs. 8.48 percent). The low mobility of European VC capital among industries,
as documented previously, could impede European investors from moving quickly
toward more profitable opportunities. As Krussl and Krause (2013) point out, the
investment amount strongly affects the probability of IPO exists. Finally, Hege et al.
(2009) show the lack of active IPO markets for venture-backed companies in Europe
no longer impedes IPOexits.
In summary, recent studies show a reconciliation of European and U.S. IPO exit rates
because of the creation in Europe of financial markets dedicated to the venture-backed
companies. However, this reconciliation takes place in a context of an important re-
duction of IPO exit rates in Europe as in the United States. Moreover, these rates mask
the difference in industries structure of the European and U.S. IPOs (Sahut and Lantz
2009). Compared to its European counterparts, the U.S. VC industry can more quickly
take advantage of the opportunities offered by the most prosperous industries.
The trade sale is a second best exit route for VC-backed companies. Various research-
ers document that trade sales are associated with lower value creation than IPOs. Trade
sales in both Europe and the United States present substantially different figures. As
Krussl and Krause (2013) document, trade sales exit rates exhibit a decrease for the
period starting in 2000 but this reduction is not as sharp as for IPOs. U.S. VC firms
largely outperform their European counterparts. For the period starting in 2000, nearly
22 percent of U.S. VC-backed companies have been acquired, this ratio falls to 11 per-
cent for European VC-backed companies. Moreover, Axelson and Martinovic (2013)
report that this gap depends neither on the industry considered nor on a firms maturity.
According to Krussl and Krause, the higher U.S. average investment explains the better
performance of the U.S. VC firms in trade sales. Overall, the origins of the trade sales
exit rates difference between Europe and the United States remain largely unclear and
require further study.

Toward a Convergence of Venture Capital Practices in Europe


and the United States
The previous sections suggest increasing professionalism of the European VC industry
and greater comparability of IPO performance between European and American VC
firms. Schwienbacher (2002) and Bottazzi et al. (2004) show that European venture
capitalists are committed as much as their American colleagues are to the management
teams of the companies they finance. According to Hege et al. (2009), U.S. VC firms
investing in Europe do not outperform their European counterparts. Have European
VC investors adopted the practices of their U.S. peers? This is probably not the case
as Schwienbacher (2002) shows the practices of most experienced European venture
capitalists have the best performances when they differ from common U.S. practices.
In fact, the European VC industry follows a trajectory based on its own experience
that partially overlaps experience of U.S. VC investors. As De Prijcker et al. (2012) ob-
serve, only the experience forged in house or the knowledge of the management team
may lead to an increase in performance. Their results suggest that a VC firm cannot
62 m a j o r t y p e s o f p r i vat e e q u i t y

easily obtain knowledge outside of the firm. Thus, mimicking U.S. practices cannot ex-
plain the performance registered by the European venture capitalists .
Studies on European venture capitalists confirm that they resort less often to com-
plex instruments of financing than their U.S. counterparts. As Kaplan and Strmberg
(2001) note, U.S. Venture capitalists use convertible preferred stocks in 80 percent of
financing rounds. By contrast, Schwienbacher (2002) shows that European venture
capitalists use convertible securities three times less often than their U.S. counterparts,
which leads him to develop several financial and fiscal arguments to explain the reasons
for this difference.
Evidence suggests that European venture capitalists replace the entrepreneurs at
the head of the company less often than U.S. venture capitalists. For instance, Schwien-
bacher (2002) indicates that, on average, European venture capitalists replaced only 22
percent of former entrepreneurs before the venture capitalists exit compared to 34 per-
cent in the United States. Thus, another line of argument is needed to justify this differ-
ence. According to Schweinbacher, the labor market is less liquid in Europe than in the
United States. Thus, European venture capitalists should have greater difficulty replac-
ing the management of companies they finance than U.S. venture capitalists.
Finally, Schwienbacher (2002) reports the percentage of syndicated investments re-
mains less important in Europe than for the United States (60 percent vs. 90 percent),
which implies, on average, fewer partners (three vs. four). Thus, another justification is
needed to explain the difference.
In summary, finding a common denominator that could explain the origin of the sus-
tainable differences between VC industries on each side of the Atlantic is difficult. One
area that calls for future research involves the consequences of the predominance of
public funding in the European VC industry. Indeed, a likely explanation is that public
funding largely orientates the VC investments choices in Europe. These entities are
mainly for start-ups in industries deemed strategic by the public authorities. As major
players, the public authorities can also weigh in on the common practices of the Euro-
pean VC industry. Thus, further research is needed to ascertain the impact of public
funding predominance on the practices of others European VC investors.

Summary and Conclusions


This chapter discusses how the European VC industry has evolved, benefited from ex-
perience, and achieved higher returns as a result of investor stability. Thus, the perfor-
mance gap in IPOs with the United States has almost disappeared. Although the chapter
offers insights about the practices and performance of European venture capitalists, the
picture would be more complete if information about VC was more available. Despite
its efforts, the information produced by the EVCA is less detailed than the National
Venture Capital Association, its American counterpart. Additionally, the databases of
Thomson Reuters or Dow Jones favor information from U.S. firms (Kaplan, Sensoy, and
Strmberg 2002).
Although European venture capitalists have succeeded in overcoming some fiscal,
legal, and administrative constraints, they still exist and are likely to hinder the invest-
ment attractiveness of a country compared to the others. Although young, mobile
Ve n t u re C apit al in E u rope 63

entrepreneurs flourish in all European countries, whether professionalization within


the VC industry will continue at the same rate in all European countries is unclear.

Discussion Questions
1. Identify and discuss the main features of the VC industry in Europe.
2. Discuss whether the American and European VC practices converged.
3. Indicate two approaches for analyzing the performance of PE funds.
4. Explain why returns of PE funds have a J-curve effect.

References
Abell, Peter, and Tahir M. Nisar. 2007. Performance Effects of Venture Capital Firms Networks.
Management Decision 45:5, 923936.
Artus, Patrick. 2008. Private Equity: Un succs transitoire d lenvironnement ou un succs du-
rable. Rapport Private Equity et Capitalisme Franais, Conseil Danalyse Economique, 75.
Artus, Patrick, and Jrme Teletche. 2004. Asset Allocation and European Private Equity: A First
Approach Using Aggregated Data. Performance Measurement and Asset Allocation for Euro-
pean Private Equity Funds, EVCA Research Paper.
Axelson, Ulf, and Milan Martinovic. 2013. European Venture Capital: Myths and Facts. Research
Reports, British Venture Capital Association, January. Available at http://www.bvca.co.uk/
Portals/0/library/Files/News/2013/European_MandF_Report_21Jan13.pdf
Blaydon, Colin C., and Michael Horvath. 2003. LPs Need to Trust General Partners in Setting
Valuations. Venture Capital Journal 43:3, 4849.
Bottazzi, Laura, and Marco Da Rin. 2002. Venture Capital in Europe and the Financing of Innova-
tive Companies. Economic Policy 17:34, 229270.
Bottazzi, Laura, Marco Da Rin, and Thomas Hellmann. 2004. The Changing Face of the European
Venture Capital Industry: Facts and Analysis. Journal of Private Equity 7:1, 2653.
Brown, James R., Steven M. Fazzari, and Bruce C. Petersen. 2009. Financing Innovation and Growth:
Cash Flow, External Equity, and the 1990s R&D Boom. Journal of Finance 64:1, 151185.
Chesbrough, Henry. 2006. Open Innovation: The New Imperative for Creating and Profiting from Tech-
nology. Boston, MA: Harvard Business School Press.
Cochrane, John H. 2005. The Risk and Return of Venture Capital. Journal of Financial Economics
75:1, 352.
Cressy, Robert. 2006. Venture Capital. In Mark Casson, Bernard Yeung, Auradha Basu, and Nigel
Wadeson, eds., The Oxford Handbook of Entrepreneurship, 353386. Oxford: Oxford University
Press.
De Prijcker, Sofie, Sophie Manigart, Mike Wright, and Wouter De Maeseneire. 2012. The Influence
of Experiential, Inherited and External Knowledge on the Internationalization of Venture Cap-
ital Firms. International Business Review 21:5, 929940.
Ernst & Young. 2011. Globalizing Venture Capital, Global Venture Capital Insights and Trends
Report. Available at http://www.ey.com/Publication/vwLUAssets/Globalizing_venture_
capital_VC_insights_and_trends_report_CY0227/$FILE/Globalizing%20venture%20
capital_VC%20insights%20and%20trends%20report_CY0227.pdf.
European Private Equity and Venture Capital Association. 2012. 2012 Pan-European Private
Equity and Venture Capital Activity Data on Fundraising, Investments and Divestments.
Available at http://www.evca.eu/media/12067/2012_Pan-European_PEVC_Activity.pdf.
Ghemawat, Pankaj, and Tarun Khanna. 1998. The Nature of Diversified Business Groups: A Re-
search Design and Two Case Studies. Journal of Industrial Economics 46:1, 3561.
64 m a j o r t y p e s o f p r i vat e e q u i t y

Gompers, Paul A., Anna Kovner, Josh Lerner, and David S. Scharfstein. 2010. Performance Persist-
ence in Entrepreneurship. Journal of Financial Economics 96:1, 1832.
Gompers, Paul A., and Josh Lerner. 1997. Risk and Reward in Private Equity Investments: The
Challenge of Performance Assessment. Journal of Private Equity 1:2, 512.
Gompers, Paul A., and Josh Lerner. 1999. The Venture Capital Cycle. Cambridge, MA: MIT Press.
Gompers, Paul A., and Josh Lerner. 2000. Money Chasing Deals? The Impact of Fund Inflows on
Private Equity Valuations. Journal of Financial Economics 55:1, 281325.
Gottschalg, Oliver, and Ludovic Phalippou. 2009. The Performance of Private Equity Funds.
Review of Financial Studies 22:4, 17471776.
Headd, Brian. 2003. Redefining Business Success: Distinguishing between Closure and Failure.
Small Business Economics 21:1, 5161.
Hege, Ulrich, Frederic Palomino, and Armin Schweinbacher. 2009. Venture Capital Performance:
The Disparity between Europe and the United States. Finance 30:1, 750.
Hellmann, Thomas. 1998. The Allocation of Control Rights in Venture Capital Contracts. Journal
of Economics 29:1, 5776.
Hochberg, Yael V., Alexander Ljungqvist, and Yang Lu. 2007. Whom You Know Matters: Venture
Capital Networks and Investment Performance. Journal of Finance 62:1, 251301.
Hwang, Min, John M. Quigley, and Susan E. Woodward. 2005. An Index for Venture Capital,
19872003. Contributions to Economic Analysis and Policy 4:1, 145.
Hyytinen, Ari, and Pekka Ilmakunnas. 2007. What Distinguishes a Serial Entrepreneur? Industrial
and Corporate Change 16:5, 793821.
Jones, Charles M., and Matthew Rhodes-Kropf. 2003. The Price of Diversifiable Risk in Ven-
ture Capital and Private Equity. Working Paper, Columbia University Graduate School of
Business.
Kaplan, Steven N., Frederic Martel, and Per Strmberg. 2007. How Do Legal Differences and Ex-
perience Affect Financial Contracts? Journal of Financial Intermediation 16: 3, 273311.
Kaplan, Steven N., and Antoinette Schoar. 2005. Private Equity Performance: Returns, Persistence
and Capital Flows. Journal of Finance 60:4, 17911823.
Kaplan Steven N., Berk A. Sensoy, and Per Strmberg. 2002. How Well Do Venture Capital Data-
bases Reflect Actual Investments? Working Paper, University of Chicago. Available at http://
ssrn.com/abstract=939073.
Kaplan, Steven N., and Per Strmberg. 2001. Venture Capital as Principals: Contracting, Screening
and Monitoring. American Economic Review 91:2, 426430.
Kaserer, Christoph, and Christian Diller. 2004. European Private Equity Funds: A Cash Flow
Based Performance Analysis. CEFS Working Paper.
Kortum, Samuel, and Josh Lerner. 2000. Assessing the Contribution of Venture Capital to Innova-
tion. Journal of Economics 31:4, 674692.
Krussl, Roman, and Stefan Krause. 2013. Has Europe Been Catching Up? An Industry Level Anal-
ysis of Venture Capital Success over 19852009. LSF Working Paper Series 1316, Luxem-
bourg School of Finance, University of Luxembourg.
Lerner, Josh, Antoinette Schoar, and Wan Wongsunwai. 2007. Smart Institutions, Foolish Choices?
The Limited Partner Performance Puzzle. Journal of Finance 62:2, 731764.
Ljungqvist, Alexander, and Matthew P. Richardson. 2003. The Cash Flow, Return and Risk Char-
acteristics of Private Equity. NBER Working Paper 9454.
Medjad, Karim, Etienne Krieger, Violetta Gerasymenko, Romain Grandsart, and Frederic Iselin.
2011. You Said Successful? Actual and Perceived Performance of Venture Capital in France.
International Journal of Business 16:4, 353366.
Meyer, Thomas, and Pierre-Yves Mathonet. 2005. Beyond the J Curve: Managing a Portfolio of Venture
Capital and Private Equity Funds. New York: John Wiley & Sons.
Myint, Yin M., Shailendra Vyakarnam, and Mary J. New. 2005. The Effect of Social Capital in New
Venture Creation: The Cambridge High-Technology Cluster. Strategic Change 14:3, 165177.
Patricof, Alan. 1989. The Internationalization of Venture Capital. Journal of Business Venturing 4:4,
227230.
Ve n t u re C apit al in E u rope 65

Phalippou, Ludovic, and Maurizio Zollo. 2005. The Performance of Private Equity Funds. Work-
ing Paper, INSEAD-Wharton Alliance.
Rdis, Jean, and Jean-Michel Sahut. 2013. Entrepreneuriat rpt, capital organisationnel et accs
au financement par capital-risque. Gestion 2000 31:4, 85108.
Sahut, Jean-Michel, and Jean Sebastien Lantz. 2009. Active Financial Intermediation and Market
Efficiency: The Case of Fast-Growing Firms Financed by Venture Capitalists. International
Journal of Business 14:4, 321339.
Schwienbacher, Armin. 2002. An Empirical Analysis of Venture Capital Exits in Europe and the
United States. European Finance Association-Berlin Meetings Discussion.
Sorenson, Olav, and Toby E. Stuart. 2001. Syndication Networks and the Spatial Distribution of
Venture Capital Investments. American Journal of Sociology 106:6, 15461588.
Wagner, Joachim. 2003. Testing Lazears Jack-of-All-Trades View of Entrepreneurship with German
Micro Data. Applied Economics Letters 10:11, 687689.
Westhead, Paul, Deniz Ucbasaran, and Mike Wright. 2005. Experience and Cognition. Interna-
tional Small Business Journal 23:1, 7298.
Woodward, Susan E., and Robert E. Hall. 2003. Benchmarking the Returns to Venture Capital.
NBER Working Paper 10202.
5
Leveraged Buyouts
CHRISTIAN RAUCH
Assistant Professor of Finance, Goethe University Frankfurt, Germany,
and SAFE Center of Excellence of Goethe University Frankfurt

M A R C P. U M B E R
Assistant Professor of Finance, Frankfurt School of Finance and Management, Germany

Introduction
Buyout funds have established themselves as an important factor on the market for
corporate control, as well as an important asset class to institutional investors and high
net worth individuals. Thus, buyout funds form the largest fraction of private equity
(PE) funds with total investment volume. In 2013, the total volume of capital raised for
leveraged buyouts (LBOs) totaled $169 billion while the second largest PE fund type
only raised $76 billion. Not surprisingly, LBOs are also among the largest investments
across all types of PE investments. A leveraged buyout is the acquisition of a private or
public company by a PE investor. To a great extent, an LBO is financed by debt capital
while the fraction of equity originates from an investment fund, usually referred to as
the buyout fund. The fund is managed by a PE firm and typically set up as a limited part-
nership in which the PE firm acts as the general partner (GP). The PE firm raises money
from institutional investors or high net worth individuals (HNWIs) who become a lim-
ited partner (LP) by committing capital to the fund. After fundraising, the GP invests
the funds capital along with external debt to conduct LBOs. In a typical LBO, a target
companys debt and equity is entirely purchased and replaced by the capital structure
used in the acquisition.
The sole purpose of an LBO is to generate returns to buyout fund investors. This
goal can be accomplished in two ways: (1) recapitalizing the target company using a
highly levered financing structure and (2) restructuring the business of the company
using various value-enhancing strategies. The latter of these return sources seems intui-
tive. By restructuring a company, any existing inefficiencies in the operating business are
eliminated and the company can later be sold at a premium yielding a positive return to
buyout fund investors.
A more subtle rationale is behind the value creation through financial leverage.
Buyout funds use the leverage effect to increase equity returns. Debt has two primary

66
L e v e rag e d Bu y ou t s 67

effects on equity returns. First, interest expenses on debt are tax-deductible, therefore
shielding some of the firms cash flows from being paid as taxes. This tax shield adds to
the firms value through a higher overall free cash flow to the firm. However, interest
obligations shift the risk profile of a firm making profits more lucrative to equity inves-
tors but also making losses more severe. The second effect is commonly known as the
mortgage effect. The underlying idea is that buyout funds use only a small fraction of
the funds equity and a large fraction of external debt to purchase all outstanding equity
and debt securities of the target company. At the time of the LBO, the target company
therefore endures a recapitalization. One important aspect is the debt that is used to
finance the acquisition of the target company becomes part of the capital structure of
this company. Thus, the company is responsible for repaying the debt from its free cash
flows as quickly as possible. By paying down the debt, the equity stake of the buyout
fund becomes more valuable over time, analogous to a mortgage. This mortgage effect
further adds value to the company, mainly to its equity holders.
Due to the complex and often opaque intricacies of an LBO transaction, buyout
funds are often misunderstood. Based on the economic relevance and size of their deals,
a solid understanding of LBOs is paramount. This chapter examines the LBO process,
how the investment funds are organized, and their institutional characteristics. The re-
mainder of the chapter is organized as follows. The first part explains the life cycle of
buyout funds, followed by a detailed explanation of debt capital markets and the im-
portance they have for LBOs in the second part. In the third part, the chapter discusses
the mechanics of LBO transactions from a theoretical perspective. Part four then pres-
ents the real-life case of the Warner Music Group LBO. Part five summarizes and draws
conclusions.

The Life Cycle of Buyout Funds


According to Preqin (2014), the financial crisis of 20072008 dramatically affected the
global fundraising for buyout funds. From a pre-crisis level of $462 billion, total fund-
raising volume lost more than 80 percent to $77 billion in 2011, as shown in Figure5.1.
Buyout fundraising has recovered. Although buyout funds still represent the largest
volume among different types of PE funds, the financial crisis affected them the most. In
absolute numbers worldwide, buyout funds ranked third in 2013 with 145 funds behind
venture capital (199 funds) and real estate (162 funds).
The average GP starts a new fund every three to five years. To raise the desired funds,
the GP conducts a road show to meet with potential investors. With an average time
on-the-road of 16.8 months, buyout funds took slightly less time than the average PE
fund with 18.6 months in 2013. According to a survey among LPs, the GPs experi-
ence and past performance are the most dominant factors in fundraising (Preqin 2014).
The findings by Kaplan and Schoar (2005) and Robinson and Sensoy (2011), who find
a strong persistence in fund performance of buyout funds, support this conventional
wisdom. To document their superior skills and experience, GPs commonly reflect their
past funds performance against carefully selected benchmarks. The most commonly
reported performance measures are the internal rate of return (IRR) and the cash mul-
tiple (CM) of past funds. When possible, GPs are proud to report that their past funds
68 m a j o r t y p e s o f p r i vat e e q u i t y

250

200

150

100

50

0
ut

re

es

ne

th

ies

ds

er
ita
tat

sse

ce
tu

th
rc
yo

ow

un
ni

ar
ap

lan
Es

ou
uc

O
atu stre

za
Bu

nd

f-F
Gr
eC
str

ez

Ba
es
al

co

-o
i
Re

lR

M
ur
fra

Se

nd
nt
ra
In

Fu
Ve
N

Number of Funds Capital Raised ($bn)

500
450
400
350
300
250
200
150
100
50
0
2008 2009 2010 2011 2012 2013

Total Buyout Fundraising (bn $)

Figure 5.1 Fundraising Levels in 2013 by Fund Type The upper graph shows the
number and the total volume in billions of U.S. dollars of fundraising capital in 2013. The
lower graph depicts the global buyout fundraising volume over time in billion U.S. dollars.
Source: Preqin (2014).

performance was among the top quartile of all funds during the same period, a
classification commonly referred to in PE. Remarkably, given the variety of data sources
and vintage year definitions, Harris, Jenkinson, and Stucke (2012b) show that more
than 50 percent of all funds could claim top-quartile performance.
The strong focus on past performance has two implications for the economics of
buyout funds. First, past over-performance allows a GP to raise more capital in sub-
sequent funds. Second, better past performance and higher fundraising also affects a
GPs compensation. For buyout funds, Metrick and Yasuda (2010) show that by in-
creasing the size of following funds, GPs can increase the fraction of fixed compensation
dramatically over time. They find that successful GPs can increase their fixed-revenue
component to roughly two-thirds of their overall compensation. GPs are usually com-
pensated based on both a fixed component through a fund management fee, which is
L e v e rag e d Bu y ou t s 69

Table 5.1 Compensation Structures in Leveraged Buyouts, 1990 to 2013

Compensation % of Funds with Component Mean (%) Median (%)


Management fee 100.0 1.9 2.0
Carried interest 100.0 20.3 20.0
Hurdle rate 83.5 7.3 8.0

Note: This table shows summary statistics of fund-level compensation components in LBO funds
in the United States between 1990 and 2013. The table displays the average number of funds having
one of the three components included as part of their overall compensation package, as well as the
mean and median numbers for the volume of the three compensation components.
Source: Preqin (2014).

linked to the fund size, and a variable component called carried interest, linked to the
funds profits.
As Table 5.1 shows, average management fees are set around 2 percent of the funds
total funding volume, and the performance-linked compensation component carried
interest is set at 20 percent of the funds profits. To create added performance incentives,
the compensation schemes often contain a so-called hurdle rate or preferred return, which
requires the fund to reach a certain initial fund return before the fund manager earns the
carried interest. Table 5.1 shows that the average required rate is around 8 percent but
16.5 percent of all LBO funds do not have a hurdle rate in place. Despite the elaborate
structure of performance-based compensation in buyout funds, the fixed compensation
through fund management fees plays an important role in GP compensation.
The persistence of top-quartile fund performance seems to have vanished. In a
recent study, Harris, Jenkinson, Kaplan, and Stucke (2012a) show that the analysis un-
derlying the time period largely drives fund persistence. For fund performance before
2000, the authors replicate the findings of other studies showing a strong persistence in
fund returns. Yet, in the period after 2000, this persistence seems to have disappeared.
When sorting GPs according to their past fund performance, which is common indus-
try practice when benchmarking, little predictive power remains in past performance on
how their subsequent funds will perform. Still, Harris et al. find a positive correlation
between past and current fund performance in a multivariate setting, which is consist-
ent with a certain degree of GP skills behind buyout fund performance.
Since 2008 the number of funds raising capital has been increasing steadily while
the total amount of target capital has not changed substantially, resulting in a much
fiercer competition among GPs in recent years. This is also reflected in a decrease of
investors willingness to commit capital early on in fundraising. In 2006, one-third of
funds could raise more than 75 percent of their target capital by the time of their first
investment (i.e., the first close). This number decreased to a mere 12 percent of funds
in 2012. Investors are more reluctant to commit capital before the GP starts to invest,
which makes the first close another important factor in effective fundraising. According
to Preqin (2014) data, more than one-third of all investors are unwilling to commit cap-
ital to a fund before a first close. To some LPs, the first investment acts as a reassurance
as theycan observe other LPs who committed capital to the fund, and they see how
70 m a j o r t y p e s o f p r i vat e e q u i t y

investors money is spent. To soften this reluctance, GPs offer more favorable terms and
conditions to investors who commit capital before the first close, and they try to have a
first close early-on in a funds lifetime. In 2012, although 26 percent of all buyout funds
had their first close within the first three months, 9 percent of all buyout funds made
their LPs wait for more than one year before conducting the first LBO.
A gradual transition occurs from the fundraising to the investing phase. Once the GP
succeeds in raising the target capital, the typical buyout fund is closed for other investors
and the LPs are committed to contribute the agreed-upon capital when needed over the
lifetime of the fund. Upon closing, the buyout fund enters its investment period and the
GP focuses on structuring and conducting LBOs. According to the average partnership
agreement, this period is around five years (Metrick and Yasuda 2010; Arcot, Fluck,
Gaspar, and Hege 2014).
Once the first buyout target is identified, the GP makes the first call on the capital
commitments and the LPs make their first contributions to the fund. This first capital
draw-down is commonly used to define the vintage year of a buyout fund. However,
some discretion exists regarding how GPs define the vintage year of their fund such as
the first investment, first capital call, or final close of fund (Harris et al. 2012b). The time
between closing of the fund and the first contribution can vary between a few months
and several years.
Over the investment period an increasing amount of capital becomes invested in
buyouts and the fraction of uncalled capital commitments, also known as dry powder,
is reduced. For example, the total amount of dry powder in global buyout funds was
roughly $400 billion at the end of 2013, which puts a total buyout fundraising of $169
billion in a new perspective. Due to reputational concerns, GPs try to avoid having too
much dry powder as their funds move toward the end of their investment period. Future
fundraising strongly depends on current fund performance. If a GP cannot invest the
committed capital of its current fund, this action could be viewed as a negative signal to
potential future investors about the abilities of the GP.
Once a portfolio of buyout companies has been established, the fund enters its
harvesting period and the GP eventually generates cash flows to the LPs, also called
distributions. These distributions can be in form of either dividends paid by portfolio
companies or divestments of shares of portfolio companies eventually leading to the
exit of an investment (once all shares of the buyout company have been sold). The har-
vesting period ends with the expected lifetime of the buyout fund, which is commonly
10 to 12 years. Similar to having too much dry powder toward the end of the investment
period, having few exits toward the end of the funds lifetime puts GPs under pressure.
To analyze the effect of a GPs investment and exit pressure, Arcot et al. (2014)
create buy- and sell-pressure indices for funds based on how close the funds are toward
the end of their investment period and lifetime, respectively. They show that funds
under buy pressure pay higher multiples and use less leverage relative to other buy-
outs as they are keener to spend the funds committed capital. Buy-pressure funds also
engage more in secondary buyouts (SBO) in which a funds portfolio company is sold
to another buyout fund. Conversely, funds under sell pressure are also more likely to
engage in SBOs from the sell side and they make substantially lower exit multiples. As
Degeorge, Martin, and Phalippou (2013) show, these multiples actually lead to lower
returns for fund investors.
L e v e rag e d Bu y ou t s 71

Ljungqvist and Richardson (2003) show that funding inflows from investors can in-
fluence the exit behavior of buyout funds during the harvesting period. They show that
an increase in funding inflows shortens the investment time and increases overall exits.
Having to set up new funds and investing newly committed capital lead to a higher fluc-
tuation in a funds portfolio companies.

Buyout Funds and Debt Capital Markets


As external debt plays an important role in LBOs, the supply of debt capital is an impor-
tant reason in buyout financing. In peak times, ample supply allowed for leverage ratios
of even more than 90 percent debt to capital in the late 1980s (Axelson, Jenkinson,
Strmberg, and Weisbach 2014). Over time, leverage ratios of LBOs decreased steadily
to pre-crisis levels of above 60 percent as is shown in the upper chart of Figure 5.2. Al-
though the financial crisis of 20072008 drastically reduced the supply of debt capital
leading to lower debt ratios for LBOs in later years, the debt ratios returned to a median
of 65 percent in 2013.
Due to their pronounced demand for debt capital, LBOs to some extent helped
create the syndicated loan market during the mid-1980s. For buyout funds, syndicated
loan markets provide large amounts of debt well in excess of what individual institutions
might be able or willing to provide. For debt capital providers, syndicated loans are usu-
ally more cost-efficient than maintaining bilateral agreements. During the financial
crisis, however, buyout funding suffered severely from dried-up debt capital markets.
Figure 5.3 shows the total volume and the average tranche size of syndicated loans
used in LBO transactions. Since the trough in 2009, leveraged loans show a weak sign of
recovery reaching a total volume of $114.4 billion in 2013, which is far below pre-crisis
levels. Still, buyouts regained attention because interest rate levels were low and debt
capital markets were again willing to fund large transactions.
After years of absence, the PE industry has anxiously watched the large-scale buyouts
of Dell Inc. and H. J. Heinz Company in early 2014 financed with debt of $5.5 billion
and $9.5 billion, respectively, fueling hope for a resurrection of the market. Although
the debt ratios in both buyouts were far from pre-crisis levels, they set a new record with
their pricing. With an interest rate of 4.25 percent and a maturity of more than seven
years, the Heinz LBO paid the lowest interest rate in record of buyout debt according to
S&P Capital IQ (2014) Leveraged Commentary and Data. In fact, debt capital market
conditions are a decisive factor in LBOs.
To some extent, GPs can improve their conditions through relationship banking.
Ivashina and Kovner (2011) document the economic importance of the relationship
between a bank and the GP in LBO financing. The stronger the ties between a bank and
the GP, the lower is the credit spread and the weaker are the financial covenants because
of lower information asymmetries and a higher potential for cross-selling opportunities.
For the broader market environment, Axelson et al. (2014) show the leverage struc-
ture in LBOs is actually unrelated to what classic capital structure theory would predict.
Rather, the supply and pricing in leveraged loan markets mainly explain the debt capital
in LBOs. More favorable debt conditions lead to higher leverage ratios in LBOs. The
authors contend that one possible explanation is that PE might be able to arbitrage debt
72 m a j o r t y p e s o f p r i vat e e q u i t y

66%
64%
62%
60%
58%
56%
54%
52%
50%
48%
2005 2006 2007 2008 2009 2010 2011 2012 2013
Median Debt Ratio
12

10

08

06

04

02

00
2005 2006 2007 2008 2009 2010 2011 2012 2013
Debt/EBITDA Equity/EBITDA

Figure 5.2 Median Leveraged Buyout Debt Levels and EBITDA Multiples
over Time in the United States The upper exhibit shows the median debt-to-total
capital ratio of U.S. buyouts. The lower exhibit shows EBITDA multiples over time. The
overall size of each bar reflects the enterprise value (EV) to EBITDA multiple. Each bar is
split into a light and a dark grey area. The light grey depicts the fraction of equity and the
dark grey shows the fraction of debt-to-EBITDA multiple. Source: The authors created these
illustrations based on data provided by PitchBook (2014).

markets against equity markets thereby reaping benefits for their fund investors. Unfor-
tunately, they find that more favorable debt conditions lead to higher entry multiples
and to lower returns for fund investors. Axelson et al. (2014, p. 2226) conclude that PE
funds tend to overpay for deals at times when leverage is cheap.
Given the recovery of syndicated loan markets since the financial crisis of 2007
2008 and the current amount of dry powder and funds raised, the main obstacle that
hampers the current LBO climate resides in high valuations on equity markets. Ample
supply of capital after the financial crisis has led to high entry multiples resulting in a
difficult market environment for most types of PE.
For example, in their sample of large-scale LBOs between 1995 and 2005, Acha-
rya, Gottschalg, Hahn, and Kehoe (2013) document a median entry earnings before
L e v e rag e d Bu y ou t s 73

700

600

500

400

300

200

100

0
2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
Total Loan Volume (bn USD) Average Tranche Amount (mil USD)

Figure 5.3 Total Volume and Average Tranche Size of Syndicated Loans used
in Leveraged Buyouts The exhibit shows the total syndicated loan volume (in billions
of U.S. dollars) and the average tranche amount of syndicated loans (in millions of U.S.
dollars) based on Thomson Reuters LPC DealScan data. Source: The authors created this
illustration based on data provided by Thomson Reuters LPC DealScan (2014).

interest, taxes, depreciation, and amortization (EBITDA)-multiple (i.e., the ratio of en-
terprise value to EBITDA of a LBO target company at acquisition time) of 6.5 while
the median exit multiple was 7.9. According to PitchBook data shown in Figure 5.2, the
median U.S. multiple at the peak before the meltdown in 2008 was 9.6 times EBITDA.
In 2013, the multiple was even higher at 10.0 times EBITDA. Although high multiples
are favorable for the exit of buyout investments, they are strongly impeding the engage-
ment into new buyouts. The question of entry and exit multiples and deal valuation is
especially important when addressing the deal level of buyout funds. After looking at
various aspects of the environment of buyout funds, the next section provides a more
detailed discussion of the mechanics of LBOs on an individual deal level.

The Mechanics of Leveraged Buyouts


As mentioned in the introduction, LBOs apply two main value drivers to increase
equity returns: (1) recapitalizing the target company using a highly levered financing
and (2) restructuring the company using various value-enhancing strategies. To imple-
ment these value drivers successfully, an LBO is conducted in a three-step process. In
the pre-investment phase, the LBO fund manager identifies a suitable target company
and structures the financing with debt and equity for the deal. In the investment phase,
the acquired company is restructured to increase economic value and to repay the LBO
transaction debt. In a final exit phase, the target company is sold, some of the remaining
debt is paid off, and the proceeds from selling the company are distributed to the fund
74 m a j o r t y p e s o f p r i vat e e q u i t y

investors. The details of how fund managers successfully execute an LBO to generate
equity returns in this three-step process are now explained.

PRE-INVESTMENT PHASE
The most important part of the pre-investment phase is choosing a suitable target com-
pany. Since it typically takes five to six years to complete an LBO and the transaction
often requires billions of dollars of invested capital, the choice of target is critical to the
buyout fund and its investors. Choosing the wrong target can be damaging to the suc-
cess of the fund and the investors equity, which is why fund managers sometimes take
years to find eligible companies and negotiate their acquisition terms.
What factors are most important in choosing a target? Usually, companies should
meet three main criteria to be considered as an LBO target.

Economic value creation. The company should have a sufficient potential for eco-
nomic value creation. Often, targets are large, inefficiently run conglomerates. Fund
managers are fond of such companies because they can swiftly create value through
management replacements, tighter governance structures, and the divestiture of un-
derperforming business divisions.
Free cash flows. The free cash flows of the company (or at least its potential for cash
flow creation) must be sufficiently high to allow for a successful implementation of
the leverage plan. Some companies generate ample free cash flow that they could use
to pay down the debt taken on during the LBO.
Low price. The company should be obtainable at a low price (i.e., the current market
value of the company should be reasonably low and its takeover must be relatively
cheap). Although this last point seems trivial, it can be the main deterrent for many
potential LBO targets. In fact, many large conglomerates use antitakeover provisions
allowing the incumbent management to fend off hostile takeover attempts or make
them costly. After all, the potential benefits from value creation and debt capacity
trade-off against the costs of making the LBO happen.

INVESTMENT PHASE
In the investment phase, the target company is restructured to reap the benefits of eco-
nomic value creation and to generate free cash flow to pay down its debt. There are four
distinct categories of restructuring efforts: (1) financial engineering, (2) operational
engineering, (3) governance intervention, and (4) management monitoring (Kaplan
and Strmberg 2008).
Financial engineering describes restructuring efforts connected to the financing
structure and the financial accounting of the target company. Restructuring the financ-
ing is both directed at the leverage effect and at the choice of the funding instruments.
As previously explained, acquiring the target company is financed with only little equity
but large amounts of debt. The goal is to acquire both equity and debt instruments of the
portfolio company to replace its old capital structure with the new acquisition financing
structure. The portfolio company is then responsible for repaying the principal and the
L e v e rag e d Bu y ou t s 75

interest payments of debt. This capital structure is chosen for two reasons. First, debt
interest payments are tax-deductible. The company therefore generates higher overall
cash flows to the firm, effectively creating a higher value. Second, by paying down the
debt, the equity stake becomes more valuable over time, similar to the impact of making
mortgage payments over the life of a loan for real estate property. Figure 5.4 displays
the effect.
The LBO fund has to obtain financing instruments that allow the successful imple-
mentation of the leverage plan. The funding costs have to be sufficiently low to maintain
free cash flow available for debt pay-down after the interest payments have been made.
The repayment conditions of the debt instruments need to allow for a timely repay-
ment, given the usually short time horizon of an LBO of typically five to six years. Also,
the contractual covenants need to allow for all other restructuring activities such as asset
divestitures.
As Ivashina and Kovner (2011) show, large LBO firms that often put together fund-
ing packages for their transactions maintain excellent relationships with banks and insti-
tutional debt investors. Thus, they can negotiate the desired terms of low interest rates
and few covenants. Demiroglu and James (2010) show that this is especially the case for
reputable buyout funds.

Year 1 Year 2 Year 3 Year 4


Entry Exit

Remaining
Debt 1 (Debt paydown through FCF) Debt

Debt 2 Debt 2

Debt 3 Debt 3 Debt 3

Equity
Debt 4 Debt 4 Debt 4 Debt 4

Equity Equity Equity


Equity
Entry financing:

-25% Equity
-75% Debt (Equity value increase due to value creation)
FCF can repay
All preexisting large amounts of
debt is repaid to debt plus increase
fully refinance the in equity value
target Target IRR of
25 to 30%

Figure 5.4 The Value Creation Principle in Leveraged Buyouts The figure
shows a typical LBO leverage structure over time.
76 m a j o r t y p e s o f p r i vat e e q u i t y

The second part of financial engineering is optimizing the financial accounting of


the target company. LBO firms often use the full potential of discretion in valuing cer-
tain assets to lower the target companies taxable income through higher depreciations.
Firms also engage in earnings management, which can be used to tweak the financial
statements of a target company shortly before the exit to achieve a higher sales price.
Chou, Gombola, and Liu (2006) document that LBOs show significantly stronger dis-
cretionary accruals around their initial public offerings (IPOs) than non-LBO firms that
go public, which is a clear sign of earnings management.
Operational engineering, also known as operational excellence, describes restructuring
the operating business of the target company. The goal is to either increase revenues or
lower the costs to create more free cash flow from the operations. As Muscarella and
Vetsuypens (1990) contend, the general goal is to make the portfolio companies more
efficient in their operations. Frequent strategies involve reducing corporate overhead
through layoffs, a shutdown of unprofitable production lines or divestiture of unprofita-
ble divisions, and strengthening and expanding profitable products or divisions.
Asset sales are also a vital part of operational engineering. Spinning off unprofita-
ble or cash-weak divisions or product lines into separate entities and selling them can
accomplish two things: a singular cash inflow from the asset sale and a higher overall
profitability of the company. Holthausen and Larcker (1996) as well as Murray, Niu,
and Harris (2006) document that buyout funds successfully boost the operating per-
formance of their portfolio companies using the previously mentioned strategies. In the
years leading up to the buyout funds exit, the portfolio companys operating perfor-
mance steadily increases to above industry-level averages. This superior performance is
maintained until well after the funds exit.
Governance intervention and management monitoring are two mechanisms to
lessen agency costs in target companies. Governance intervention describes the active
restructuring of the governance structures of the target company to lower agency costs.
As Jensen (1989) describes in his early formalized note on LBOs, buyout funds often
turn to target companies in which value creation can be accomplished easily by reduc-
ing excessively high agency costs because of bad corporate governance. To accomplish
this, the LBO funds select a new management team that replaces the incumbent man-
agement of the target company on the acquisition. To align the interests of equity inves-
tors and managers, the target companys management is awarded with ownership stakes
in their company. Bruton, Keels, and Scifres (2002) document that operational perfor-
mance of portfolio companies increases for firms with stronger managerial ownership
during the LBO investment. To further reduce agency costs and to make sure the man-
agers act in the best interests of the equity investors, the LBO fund managers hold board
seats to closely oversee the managers and their actions. Taken together, these measures
accomplish two things. First, the target company becomes more valuable due to a re-
duction of agency costs. Second, the LBO firm can make sure that all desired restruc-
turing mechanisms as previously described are fully implemented by the management.
Besides restructuring and leverage, some buyout funds conduct so-called dividend re-
capitalizations as a third strategy to create value for their investors, especially in large and
profitable portfolio companies. In this special recapitalization, the portfolio company
issues new debt and immediately pays out the proceeds as a special dividend during a
later stage of the LBO. Buyout funds use these transactions to extract debt-funded cash
L e v e rag e d Bu y ou t s 77

flows from the portfolio companies, which are subsequently distributed back to the
LPs of the fund. To accomplish this feat, the portfolio company gets more debt fund-
ing through short-term notes or bank bridge loans. The proceeds from this debt are
directly paid out to the buyout fund in the form of a dividend. Often, portfolio compa-
nies execute this transaction shortly before their IPOs to use the IPO proceeds for debt
repayment.
A dividend recapitalization has four advantages. First, it allows the buyout fund to
distribute cash to its investors before the actual exit of the company. Second, the IRR
of the deal and the fund will benefit from this transaction because cash flows are paid
out much earlier. Third, the buyout fund can lock in some proceeds from the deal
without having to bear market risk from post-IPO share sales. And fourth, it allows
GPs to time certain cash flows such as when they are still in need of clearing the hurdle
for the preferred return of their management contract. However, dividend recapitaliza-
tions are also subject to criticism. The additional debt burden is costly for the portfolio
company, and, if all IPO proceeds are used for debt pay-down, they cannot be used
for investments in positive NPV projects. Since these transactions are not part of the
typical restructuring of an LBO, they should be seen as an alternative measure of value
creation.
Table 5.2 shows selected restructuring strategies in 224 LBOs that had an IPO exit in
the United States between 1998 and 2012. The table provides four important insights.
First, oversight and control by the GP over the portfolio company are pivotal parts in
an LBO structure. Almost all buyout funds are strong majority owners and hold board
seats in their portfolio companies. Second, the portfolio companies management re-
ceives an average of 10 percent of the shares of their company to align the interests of
owners and managers. Third, merger and acquisition (M&A) deals are a vital restruc-
turing tool to create economic value. Fourth, financial engineering is mainly exercised
through increases in leverage and frequent earnings management. Finally, every second
portfolio company pays out dividends to their equity owners to create early-on pro-
ceeds. Dividends in the form of dividend recapitalizations are used in 8.5 percent of all
deals.

EXIT PHASE
After a successful restructuring, the target company is exited. The exit is comprised of
two successive steps: (1) selling the target company and (2) distributing the sale pro-
ceeds to the LPs. Generally, a buyout fund can use four exit channels: a trade sale to a
strategic investor, a trade sale to another financial investor (a so-called secondary), an
IPO, or a combination of different exit channels, known as a break-up-and-sale. Each exit
choice has its advantages and disadvantages. A trade sale to a strategic investor can usu-
ally generate a higher sale price because a strategic investor might be willing to pay an
additional premium to get the operating business of the company for its own strategic
purposes such as synergies. A sale to a secondary generates lower prices but is therefore
usually conducted faster because the seller and the buyer are serial M&A transaction
parties. An IPO is the most expensive process and at 6 to 12 months preparation time
takes longer to complete than other exit options, but it usually generates the highest
proceeds for buyout funds.
78 m a j o r t y p e s o f p r i vat e e q u i t y

Table 5.2 Restructuring Strategies in Leveraged Buyouts

Restructuring Tool Number of Companies (of 224) Mean


Board seat 216 96.4%
Chairman is GP 49 21.9%
Ownership held by buyout fund All 71.0%
Management ownership 210 10.3%
Management change at LBO 105 46.9%
M&A deals 78 34.8%
Leverage pre LBO All 0.71
Leverage at LBO All 1.49
Leverage at IPO All 1.44
Earnings management 47 21.0%
Firms with dividends 115 51.3%

Note: The table shows summary statistics of restructuring activities in LBOs. The numbers are
based on a sample of 224 LBOs exited through IPOs on U.S. stock exchanges between 1998 and 2008.
The restructuring information was collected using S-1 stock offering prospectuses of the Securities and
Exchange Commission (SEC). The table displays the number of companies that underwent a certain
restructuring activity and their percentage of the overall sample. For restructuring activities used in
every deal (ownership by buyout firms and leverage), the mean is the average percentage of shares held
by the buyout funds in the portfolio companies and the average leverage, calculated as net debt over
equity.
Source: SEC S-1 stock offering prospectuses serve as the basis for all information except for M&A
deals (Thomson ONE) and earnings management (contains information from the Government Account
Office, on financial restatements).

The empirical literature on the subject suggests that GPs approach the exit as a stra-
tegic part of the overall LBO process. This finding is true both for the timing of the exit
and the choice of exit channel. Schmidt, Steffen, and Szab (2010) show the financial
success of the target company mainly drives the choice for a given exit channel. The
most profitable LBO targets are taken public, whereas the least successful are sold via
trade sale to strategic investors. Ljungqvist and Richardson (2003) show the investment
period depends on the investment pressure. More funding inflows from LPs demand
a higher investment activity of LBO firms, which in turn leads to a higher turnover of
single deals. Thus, investment periods are shorter when the funding availability is larger.
Cao (2011) also shows that equity market conditions can play a crucial role both for
the type of exit and its timing. IPO exits are increasingly faster if the market conditions
allow for a high relative equity valuation.
Figure 5.5 shows the frequency of each exit option in an international sample of
LBOs between 1990 and 2013. The trade sale is by far the most popular option, fol-
lowed by the secondary. IPOs are only chosen in 5 percent of all cases. The incumbent
management of the portfolio companies purchases the buyout funds equity stake in
L e v e rag e d Bu y ou t s 79

Sale to
Management
IPO
3%
5%

Trade Sale Secondary


58% 34%

Figure 5.5 Choice of Exit Channels by Leveraged Buyouts, 1990 to 2013This


figure shows the relative frequencies of exit channels in all LBOs in the United States
between 1990 and 2013. Source: Preqin (2014).

only 3 percent of all LBO exits. But even though the IPO is not the most frequent form
of exit, it still is the most profitable route from the perspective of the buyout fund.
As Figure 5.6 shows, the relationship of total sale proceeds to the equity investment
(known as a multiple, a popular return measure in LBOs) is by far the highest for LBOs.
Trade sales are more profitable than secondaries, a finding that supports the assumption
that strategic investors are willing to pay more for portfolio companies than LBO firms.
A new phenomenon in LBO exits is the sale of a portfolio company to so-called spe-
cial purpose acquisition corporations (SPACs). SPACs are empty shell corporations that

4.92
4.39
3.71

2.19

Sale to Secondary Trade Sale IPO


Management

Figure 5.6 Cash Multiples of Leveraged Buyouts by Exit Channel, 1990 to


2013 This figure shows the average (mean) cash multiples of LBOs with different exit
channels. The cash multiples are calculated as the ratio of total proceeds from the exit sales
to total equity investments in the portfolio companies. Source: Preqin (2014).
80 m a j o r t y p e s o f p r i vat e e q u i t y

raise capital through an IPO. Their goal is to use the IPO capital for acquiring a privately
held company, to merge the target company into the publicly listed shell corporation,
and therefore to turn the SPAC investors ownership rights into ownership rights of
an operating company. Since these vehicles provide mostly institutional investors with
the possibility to invest in privately held companies through highly liquid investment
funds, SPACs are also known as one-time liquid private equity funds. SPACs have estab-
lished themselves as exit vehicles for LBO portfolio companies. Since SPACs are usually
run by former PE managers and closely resemble PE funds, selling portfolio companies
to SPACs has become an alternative to secondary sales for buyout funds. One promi-
nent example is Burger King, which 3G, a PE firm, sold to a SPAC called Justice Hold-
ings in June 2012. The SPAC had been set up by notable financiers William A. Ackman
and Nicolas Berggruen and offered 3G a quick and profitable exit opportunity for the
company.
Once the exit is fully conducted, the generated proceeds have to be distributed to
the buyout fund investors. The way the proceeds are distributed in the so-called distri-
bution waterfall is regulated by the partnership agreement between investors and fund
managers. Based on the institutional features of the fund and its fee structure, the fund
managers receive their share of fees and profits, with the remaining proceeds being dis-
tributed to the fund investors.

A Leveraged Buyout Case Study: Warner Music Group


On March 1, 2004, Warner Music Group was spun off from Time Warner and sold to
a conglomerate of financial investors in an LBO transaction for a price of $2.59 bil-
lion. The buyers were Thomas H. Lee Funds (purchasing 48.6 percent), Bain Capital
(20.8 percent), Music Capital Partners (12.3 percent), and Providence Equity Partners
(11.1 percent). Before the LBO, Warner Music was the music production and record
company of Time Warner. Founded in 1958 with the Warner Brothers Movie produc-
tion company, it originally scored soundtracks for Warner Brothers films. Between
1970 and 2000, Warner Music established itself as one of the leading record companies
worldwide, producing and selling some of the most famed artists of this period. Warner
Musics demise came with a difficult transition into the digital age of freely download-
able music, and facing the problems brought on the company by the various restruc-
turings of its corporate parents. Warner Music was at the losing end of a battle against
newer and more successful record companies in the early 2000s. Thus, buyout funds
stepped in to buy and restructure the company.
A combination of debt and equity funded the purchase price of the company: $1.04
billion was investor capital (equity), $1.2 billion was raised in form of a syndicated bank
term loan, and $650 million stemmed from issuing senior subordinated notes, owned
by the four invested buyout funds. As a consequence of the LBO, Warner Music un-
derwent a drastic recapitalization. Before the deal, it was a wholly owned subsidiary
of Time Warner with $130 million in long-term debt and annual interest payments of
$2 million. After the LBO, it had $1.84 billion with $80 million in annual interest pay-
ments. The companys leverage increased from a debt-to-equity ratio of 0.07 to 6.57.
After the acquisition, the buyout funds exercised a rigorous restructuring plan to save
L e v e rag e d Bu y ou t s 81

costs and further expand the business into profitable areas. As stated in official SEC
filings, the restructuring plan saved $250 million in annual costs, and only cost the com-
pany $225 million to set up and implement. Also, the company founded a joint ven-
ture with P. Diddys Bad Boy Records music production company to market, promote,
and distribute Warner Musics artists through P. Diddys name and brand. In December
2004, 10 months after the LBO and six months before its planned IPO, Warner Music
paid $472 million to its equity investors in the form of a dividend recapitalization. Fi-
nancing for the recapitalization mostly occurred by issuing senior notes. The company
also used $209 million of the funding to repurchase its own preferred stock from the
equity investors. The total distribution to the LBO funds as part of the transaction was
therefore $681 million, representing 65 percent of the initial (equity) purchase price.
Warner Musics IPO in May 2005 was then used to repay the recap debt. The IPO gener-
ated $554 million in proceeds, all of which went to the holders of the senior notes issued
as part of the recap transaction. Interestingly, the IPO did not mark the exit of the LBO
funds. Instead of opting for post-IPO share sales into the market, the funds remained
actively invested in the company until August 2011, when they sold their equity stakes
in a post-IPO M&A transaction to a secondary buyer.
Financially, the deal was a success for the investors. The deals IRR stood at 16.03 percent
at the exit of the funds and the cash multiple at 1.8. The deal performance measures are
strongly affected by the value contribution to equity investors by the dividend recapitaliza-
tion. When excluding this single cash flow generated by the dividend recap payment, this
would lower the IRR to 3.69 percent and the cash multiple to 0.96. Had the buyout funds
not used the debt-funded dividend, the deal would have been less successful.

Summary and Conclusions


This chapter provides an overview of the mechanics and institutional features of LBOs.
Since their start in the early 1980s, LBOs have established themselves as an important
factor on the market for corporate control. Although some LBOs with excessive degrees
of leverage or disproportionately large layoffs did a disservice to the reputation of PE,
two factors also need to be mentioned. First, LBOs impose a strong disciplining effect
on the market for corporate control for publicly listed companies. Second, PE has estab-
lished itself as being one of the largest asset classes in alternative investments, which is
an important step toward a complete capital market. LBOs are a well-established device
in capital markets. Given the current dry powder and the recent amount of capital raised
by new funds, many LBOs might soon follow once valuation multiples go back to more
reasonable levels.

Discussion Questions
1. Discuss the role of leverage in an LBO.
2. Identify typical restructuring mechanisms of LBO funds in their portfolio compa-
nies and discuss why some restructuring mechanisms may be detrimental to the
portfolio companies.
82 m a j o r t y p e s o f p r i vat e e q u i t y

3. Define dry powder and discuss how it affects a funds investment decisions?
4. Discuss the importance of maintaining a relationship with banks for a buyout fund.

References
Acharya, Viral V., Oliver Gottschalg, Moritz Hahn, and Conor Kehoe. 2013. Corporate Govern-
ance and Value Creation: Evidence from Private Equity. Review of Financial Studies 26:2,
368402.
Arcot, Sridhar, Zsuzsanna Fluck, Jos-Miguel Gaspar, and Ulrich Hege. 2014. Fund Managers
under Pressure: Rationale and Determinants of Secondary Buyouts. Journal of Financial Eco-
nomics, forthcoming.
Axelson, Ulf, Tim Jenkinson, Per Strmberg, and Michael S. Weisbach. 2014. Borrow Cheap, Buy
High? The Determinants of Leverage and Pricing in Buyouts. Journal of Finance 68:6, 2224
2267.
Bruton, Garry D., J. Kay Keels, and Elton L. Scifres. 2002. Corporate Restructuring and Perfor-
mance: An Agency Perspective on the Complete Buyout Cycle. Journal of Business Research
55:9, 704724.
Cao, Jerry X. 2011. IPO Timing, Buyout Sponsors Exit Strategies, and Firm Performance of
RLBOs. Journal of Financial and Quantitative Analysis 46:4, 10011024.
Chou, De-Wai, Michael Gombola, and Feng-Ying Liu. 2006. Earnings Management and Stock Per-
formance of Reverse Leveraged Buyouts. Journal of Financial and Quantitative Analysis 41:2,
407438.
Degeorge, Franois, Jens Martin, and Ludovic Phalippou. 2013. The Performance of Secondary
Buyouts. Working Paper, European Corporate Governance Institute (ECGI) and Swiss Fi-
nance Institute.
Demiroglu, Cem, and Christopher M. James. 2010. The Information Content of Bank Loan Cove-
nants. Review of Financial Studies 23:10, 37003737.
Harris, Robert S., Tim Jenkinson, Steven N. Kaplan, and Rdiger Stucke. 2012a. Has Persistence
Persisted in Private Equity? Evidence from Buyout and Venture Capital Funds. Working
Paper, Darden Business School, No. 2304808.
Harris, Robert S., Tim Jenkinson, and Rdiger Stucke. 2012b. Are Too Many Private Equity Funds
Top Quartile? Journal of Applied Corporate Finance 24:4, 7789.
Holthausen, Robert W., and David F. Larcker. 1996. The Financial Performance of Reverse Lever-
age Buyouts. Journal of Financial Economics 42:3, 293332.
Ivashina, Victoria, and Anna Kovner. 2011. The Private Equity Advantage: Leveraged Buyout
Firms and Relationship Banking. Review of Financial Studies 24:7, 24622498.
Jensen, Michael C. 1989. Eclipse of the Public Corporation. Harvard Business Review September
October, 129.
Kaplan, Steven N., and Antoinette Schoar. 2005. Private Equity Performance: Returns, Persistence,
and Capital Flows. Journal of Finance 60:4, 17911823.
Kaplan, Steven N., and Per Strmberg. 2008. Leveraged Buyouts and Private Equity. Working
Paper, Booth School of Business, University of Chicago.
Ljungqvist, Alexander, and Matthew P. Richardson. 2003. The Cash Flow, Return, and Risk Char-
acteristics of Private Equity. Working Paper, Stern School of Business, New York University.
Metrick, Andrew, and Ayako Yasuda, 2010. The Economics of Private Equity Funds. Review of Fi-
nancial Studies 23:6, 23032341.
Murray, Gordon, Dongmei Niu, and Richard D. F. Harris. 2006. The Operating Performance of
Buyout IPOs in the UK and the Influence of Private Equity Financing. Working Paper, Uni-
versity of Exeter.
Muscarella, Chris J., and Michael R. Vetsuypens. 1990. Efficiency and Organizational Structure: A
Study of Reverse LBOs. Journal of Finance 45:5, 13981413.
L e v e rag e d Bu y ou t s 83

PitchBook. 2014. 2Q 2014 U.S. Private Equity Breakdown Report. PitchBook Data, Inc.
Preqin. 2014. 2014 Preqin Global Private Equity Report. London: Preqin Ltd.
Robinson, David T., and Berk A. Sensoy. 2011. Private Equity in the 21st Century: Cash Flows,
Performance, and Contract Terms from 19842010. Working Paper, Fuqua School of Busi-
ness, Duke University.
S&P Capital IQ. 2014. Leveraged Commentary and Data. New York: Standard & Poors. Available
at https://www.lcdcomps.com.
Schmidt, Daniel, Sascha Steffen, and Franziska Szab. 2010. Exit Strategies of Buyout Investments:
An Empirical Analysis. Journal of Alternative Investments 12:4, 5884.
Thomson Reuters LPC. 2014. DealScan database. New York: Thomson Reuters. Available at
https://www.loanpricing.com.
6
Mezzanine Capital and Commercial
Real Estate
J. DEAN HELLER
Senior Real Estate Partner, Seyfarth Shaw LLP

Introduction
In the geography of private equity (PE), mezzanine capital inhabits the narrow, often
rugged terrain dividing the usually sunny uplands where secured debt resides from the
marshy delta populated by common equity. More succinctly, in commercial real estate
(CRE) financing, mezzanine capital lies between senior debt and common sweat.
The chapter first discusses the fundamentals of CRE mezzanine financing: the hallmark
of mezzanine investment, its funding sources and consumers, and its primary categories.
Next, it provides a brief history of mezzanine financing in CRE from the 1980s until the pres-
ent. Then, the chapter offers a more detailed analysis of the providers and users of CRE mez-
zanine capital and the different modes of mezzanine investing. Finally, the chapter addresses
the market and legal factors distinguishing such modes in the eyes of capital providers.

Fundamentals of Commercial Real Estate Mezzanine


Financing
Mezzanine investments take various forms from pure debt with fixed interest rates and
maturity dates to pure equity with neither rate nor maturity fixed (Robins, Wallace,
and Franke 2012). In debt forms, mezzanine investment may be secured or unsecured,
while as equity it may be certificatedevidenced by a certificate-like instrument
or simply embodied by provisions in a partnership, limited liability company, or other
entity agreement. The investor may hold a direct interest in the investment property, or
in the propertys owner, or its interest may be several times removed from the asset on
which investment success depends.

HALLMARK OF MEZZANINE INVESTMENTS


Whatever form they may take, all mezzanine investments have one unchanging char-
acteristic: the right to recover the invested capital from the underlying property is

84
Mezz anin e C apit al an d R e al E s t at e 85

always superior to common equity and always inferior to senior debt. In more complex
financings, mezzanine investment may be multi-storied with several layers of the capital
stack above or below it, but at least one higher and one lower level must exist for the in-
vestment to be mezzanine. Mezzanine capital occupies a position akin to the second
person in a four-person bobsled team: right behind the driver, but not in control of the
sled or able to exit quickly in the event of an impending crash. Debt forms of mezzanine
investment are often secured in the legal sense by a pledge of the entity that owns the
true real property collateral. Mezzanine equity investments usually have legal rights to
control that entity if the investment encounters distress. Nevertheless, such rights may
be frail bulwarks against the superior legal rights of a senior lender and the common eq-
uitys actual possession of the property.

S O U R C E S O F M E Z Z A N I N E C A P I TA L
The peculiarly exposed position of mezzanine capital not only drives the risk/reward
profile of mezzanine capital but also determines what kind of investor chooses, or is best
suited for, mezzanine investments.
Heightening the risk to mezzanine investors of a total loss is the relatively small
amount of the mezzanine capital usually present in most CRE financings relative to
senior debt. If the underlying property fails at some point to generate enough revenue to
cover the senior debt payments, a mezzanine investors only alternative to abandoning
its capital may well be to buy out the senior loan. Even if it has sufficient funds to pur-
chase the senior loan, the investor would have to increase its investment in a distressed
property by a multiple and at a lower rate of return.
Given these risks, the most suitable providers of mezzanine capital should combine
substantial liquidity with a healthy appetite for risk as well as reward, and considera-
ble sophistication and resources to oversee their investments expertly. Predictably, over
time most mezzanine investors match this profile. Yet, as typical of investment cycles,
high returns garnered by the early entrants in mezzanine financing have lured later in-
vestors who were less capable of assessing the risks and less adept at managing them. By
2006, foreign banks and other traditionally more conservative institutional investors
were taking large bets on mezzanine debt at relatively small spreads above CRE mort-
gage rates.

CONSUMERS OF MEZZANINE FINANCING


Outside of CRE, the most frequent use of mezzanine financing has been in private ac-
quisitions of corporate businesses (Babson Capital 2010). With CRE, mezzanine cap-
ital began in the early 1990s to fill the gap left by retreating loan-to-value (LTV) ratios
in mortgage financing. It also became particularly important in the structured financing
of real estate portfolio acquisitions, high value trophy properties, and large commer-
cial and resort developments (Fastov and Foley 2002). In the middle of the last decade,
the supply of mezzanine capital rapidly expanded. As a result, even relatively modest
property acquisitions and developments could obtain mezzanine as well as mortgage
financing.
86 m a j o r t y p e s o f p r i vat e e q u i t y

P R I M A R Y C AT E G O R I E S O F M E Z Z A N I N E F I N A N C I N G
The major division in mezzanine investment falls between CRE and othermostly
corporate acquisitionfinancing. Though both share the hallmark of lying between
senior debt and common equity, subordination of corporate mezzanine debt is deeper,
usually requiring the mezzanine lender to stand still during the first several months
of a senior debt default (Robinson, Fert, and Brod 2011). Thus, the only sure remedy
available to the corporate mezzanine lender is to pay off the senior loan. While this may
often be the practical consequence of a senior debt maturity default in CRE financing,
the standard CRE intercreditor agreement gives the mezzanine lender a brief period to
cure delinquent payments during which the senior lender must stand still (Forti and
Stafford 2002).
The other major distinction between real estate and private corporate mezzanine fi-
nancing is that preferred equity plays little part in the latter except for special cases, such
as venture capital start-up financing and convertible debt in corporate workouts, both of
which could fit the broad definition of mezzanine financing if senior debt is also present.
CRE financing, in contrast, has developed more diverse forms of mezzanine investment.

A N OT E O N A U T H O R I T I E S
Any discussion of mezzanine financing necessarily rests heavily on anecdotal evidence,
here including the authors 20-plus years of experience in many real estate mezzanine
investment transactions and perusal of trade periodicals such as Commercial Mortgage
Alert, published by Harrison Scott. While various articles discuss legal structures and
issues in mezzanine financing, scant literature beyond promotional materials and primer-
like general summaries deals with other aspects of such investments. Also lacking are
systematic collections of data on volume, rates, yields, and maturities that, for example,
the CRE Finance Council routinely collects for mortgage loan transactions. Mezzanine
capital, perhaps, is the most private of private equity, flowing almost outside the direct
view of government agencies, rating agencies, and other overseers of capital transactions.

A Brief History of Mezzanine Commercial Real Estate


Financing
Mezzanine CRE financing is at least as old as the second mortgage and second
mortgages are as old as the involvement of seventeenth-century courts of equity in
enforcing mortgages (Plucknett 1956). Yet, contemporary forms of mezzanine in-
vestment in U.S. CRE financing evolved only since the early 1990s, largely with the
rise of commercial mortgage-backed securities (CMBS) as a capital source for senior
debt financing.
Before 1990, relatively little mezzanine investment existed in CRE. Banks, sav-
ings and loans (S&Ls), and life insurance companies, the major sources of mortgage
financing, discouraged subordinate mortgage debt, so the typical financing structure
was a mortgage loan plus equity, with no middle layer. A developer or purchaser of
Mezz anin e C apit al an d R e al E s t at e 87

commercial property might entice equity co-investors in a project, but as a rule the pas-
sive equity capital received no priority in either return on or return of invested capital.
The situation changed dramatically after the 1980s crisis in the S&L industry. In
1982, Congress enacted the Depository Insurance Flexibility Act substantially dereg-
ulating S&Ls so thrifts could compete more effectively for deposits against non-bank
money market funds (Financial Crisis Inquiry Commission 2011). The Act, which
was another law of unintended consequences, led quickly to the S&L crisis, as the liber-
ated thrifts soon ventured far beyond making only 30-year fully amortizing home loans.
As a result, S&Ls became an important source of CRE financing, which events proved
them ill-equipped to underwrite.
As the dominos began to fall in 1983, the CRE debt finance market experienced
one of its periodic, constrictive spasms. The failure of many S&Ls and the eventual ex-
tinction of that industry removed a now important source of CRE financing, especially
for development and construction. Liquidation of the assets of failed thrifts, first by
the Federal Savings and Loan Insurance Corporation (FSLIC) and later by the newly
formed Resolution Trust Corporation (RTC) soaked up capital that might have other-
wise found its way into new loans and projects. Congress then re-regulated mortgage
lending by federally insured institutions, further restricting debt financing from tradi-
tional CRE lenders (Financial Crisis Inquiry Commission 2011). By 1990, few of these
lenders were willing to consider levered loans at the higher ratios of loan amount to
property value (LTV) that real estate entrepreneurs craved.
Wall Street, never too fastidious where large profits could be made with other peo-
ples money, then entered the scene. Debt securitization (i.e., pooling individual loans to
create a more risk-diversified and liquid investment) had begun in earnest in 1970, after
Congress authorized the government-sponsored residential mortgage lenders Fannie
Mae and Freddie Mac to bundle loans in their portfolios and sell participations in the
bundles, sweetened by Fannie or Freddie guarantees.
These securitizations were relatively simple structures, with each participant get-
ting a percentage interest in the entire mortgage pool. However, when large investment
banks took up mortgage securitization, they introduced tranching, in which bundled
loans were sliced both horizontally and vertically with the vertical tranches assigned se-
quential priorities (Benmelech and Dlugosz 2010). The RTC sponsored the first large,
tranched mortgage securitizations, engaging underwriters from the private sector to
create pools from loan assets of failed S&Ls. The success of the RTC offerings inspired
investment banks to enter the commercial loan securitization market. By the late 1990s,
sale of CMBS had become a vital capital source for CRE financing (Financial Crisis
Inquiry Commission 2011).
The CMBS securitization market received a further boost when the corporate bond
rating agencies started rating the senior participations in CMBS pools. Investment
grade ratings made CMBS participations attractive to institutions whose investments,
by law or policy, needed such a third-party seal of approval (Robins et al. 2012). Rating
agencies guidelines, however, required that pooled mortgage loans have conservative
LTV ratios, especially relative to the exuberant over-leveraging of the 1980s, and not
share collateral with subordinate mortgages (Standard & Poors 2003). For commercial
mortgage borrowers, this de-leveraging of their properties largely negated the benefit of
the lower CMBS interest rates.
88 m a j o r t y p e s o f p r i vat e e q u i t y

Mezzanine debt, occupying the space in the capital stack between 75 and 90 percent
of LTV, turned out to be the answer to this dilemma (Nijs 2014). Although the cost
of such mezzanine financing could be two to three times that of senior mortgage debt,
when the rates were blended this structured financing provided attractive leverage
to real estate developers and equity investors. The ability to offer a mezzanine piece
to borrowers improved the ability of CMBS loan originators and CMBS emulators to
compete, especially for larger property and portfolio acquisition financing.
Mezzanine loans soon became a significant part of CRE financing but determining
how significant is difficult because no regulatory agency or trade association systemat-
ically gathered the requisite data. Moodys, though, reported that in 2006 alone more
than $3.2 billion of mezzanine debt went into collateralized debt obligation (CDO)
pools, a figure presumably based on Moodys rating of those CDOs (Rubock 2007).
Four years earlier, two Goldman Sachs executives had estimated that the overall po-
tential for mezzanine CRE financing exceeded $100 billion (Fastov and Foley 2002).
The crash in late 2008 ended a golden era of real estate mezzanine financing. The Fi-
nancial Crisis Inquiry Commission (2011) stated that by 2010 over half of CRE mort-
gages were under-secured, making any associated mezzanine loan worthless. Whether
a like percentage of all existing commercial mezzanine loans evaporated is speculation,
but the many losses that occurred were usually total.
CMBS financing has rallied since 2011 and, while not reaching the volumes of the
pre-crash years, is again substantial. For 2014, CMBS originations are estimated to ap-
proach if not exceed $100 billion (Egan 2014). Harder to detect is whether the rising tide
in CMBS lending has also re-floated mezzanine lending. Yet, a strong demand for mezza-
nine capital in CRE financing persists, bolstered by the large amount of CMBS debt ma-
turing in the next few years. Though some former sources of mezzanine capital may have
left the market for good, others appear primed to meet that demand (Robins et al. 2012).

Providers and Users of Mezzanine Capital


If lack of systematically compiled data on mezzanine capital transactions makes esti-
mating the overall volume difficult, then estimating the amounts by source or use is
almost impossible. Making precise statements about the who, why, and how much of
consumers of mezzanine capital is also difficult. Yet, some reasonable generalizations
are possible based on reports of individual transactions and other anecdotal evidence.

SOURCES
Between 1998 and 2008, when structured financing boomed, a fair guess would be that
CMBS conduit lenders originated most real estate mezzanine loans, but these lenders
intended only to be intermediate sources of mezzanine capital. As soon as practicable,
they would sell their mezzanine loans to eventual investors. Only by miscalculation did
they become an ultimate source of funds for mezzanine loans.
The real sources of mezzanine funds in CRE financing were large pension funds es-
pecially governmental funds, some foreign banks, high net worth individuals, and hedge
funds, the primary investors in which were also pension funds, endowments, and high net
Mezz anin e C apit al an d R e al E s t at e 89

worth individuals (Silbernagel and Vaitkunas 2012). These investors obviously shared a
desire for higher yields than other debt obligations could produce. For governmental and
other large pension funds, conventional debt investments could not generate the yields re-
quired to meet defined pension benefit obligations. Mezzanine loans offered a high-return
alternative, similar to the corporate junk bonds, which had turned out well for earlier inves-
tors. Other mezzanine investors, such as hedge funds, discovered that, in an environment
in which capitalization (cap) rates were falling to historic (if not hysteric) lows, mezza-
nine loans and debt-like preferred equity could produce considerably higher returns than a
common equity investment and with less risk. (A cap rate defines market prices for a class
of real estate in terms of the acceptable, annual return that an equity investment in such real
estate will generate by its expected yearly net operating cash flow.) Predictably, early suc-
cess led to late-cycle excess. As less savvy investors began to overfund the real estate mezza-
nine debt market, the spread between mezzanine interest rates and those on the associated
mortgage debt narrowed substantially, turning risk-reward ratios topsy-turvy.
Currently, hedge funds dominate the sources of mezzanine capital. Perhaps as a
result, preferred equity investments, which were relatively rare during the decade pre-
ceding the Great Recession, have become more common (Nijs 2014). Unlike pension
funds and foreign banks, hedge funds may attach less value to the label debt and more
value to involvement in creating the investment, thus negating the perceived advantages
of mezzanine debt over preferred equity as an investment.

USERS
The attributes of mezzanine financing naturally lead to self-selection among its consumers.
First, the expenses related to structured financing discourage its use in smaller deals, partic-
ularly since mezzanine capital usually fills in only 10 to 20 percent of the capital stack. Mort-
gage-mezzanine financing for a $25 million property would mean a mezzanine loan of $5
million or less, which is hardly worth the additional costs of complicating the transaction.
Second, mezzanine capital is expensive relative to mortgage loan funds. Stabilized, rent-
generating properties with long-term credit-worthy tenants can borrow mortgage money at
higher LTVs and have less need to leverage with mezzanine financing. Conversely, proper-
ties with more volatile cash flows, including hotels and properties in need of development,
renovation, or market re-positioning are prime candidates for mezzanine financing.
Finally, mezzanine investors rely heavily on experience and reputation of the capital
consumer. Strong sponsorship is a phrase often found in investment and loan recom-
mendations for a particular mezzanine transaction; reasonably so, given that the secu-
rity for such investments has a certain spectral quality. Users of mezzanine capital are
typically well established in their sector of the real estate industry, with proven past suc-
cesses and a visible profile in the real estate capital market (Fastov and Foley 2002).

Modes of Mezzanine Investments


Real estate mezzanine investments take many forms, but the basic dichotomy lies be-
tween equity and debt. Legally, the defining characteristic of debt is that its holder, if not
paid when due, can sue the borrower for what is owed. If the loan is secured, the lender
90 m a j o r t y p e s o f p r i vat e e q u i t y

can also sue, or proceed through statutorily provided non-judicial procedures, to force
a sale of its collateral and either use its debt to purchase the collateral or receive the pro-
ceeds from a sale of the collateral to a third party. The holder of an equity investment can
neither sue for its investment nor foreclose on the property that underlies its value,
not even on an investment with a fixed payment and a mandatory redemption period.
Yet in practice, this black line distinction often becomes smudged.

D E BT O B L I G AT I O N S
The classic form of mezzanine debt occupies the middle both in the chain of property
ownership and the capital stack. Mezzanine borrowers are not owners of real property;
rather, they own the property owners and nothing else. This situation makes a mezza-
nine loan structurally subordinate to a mortgage loan, which has not only a senior
lien on the property but also a different and, in proximity to the underlying collateral, a
clearly superior borrower. In very large transactions, such as portfolio financings, mul-
tiple mezzanine loans and borrowers can exist with each borrower resting on top of the
other in an ascending chain of ever more subordinate links.
The rating agencies mandated structural subordination of mezzanine debt, requir-
ing that CMBS loans have bankruptcy remote borrowers to earn an investment grade
rating (Standard & Poors 2003). Bankruptcy remote does not mean that the mortgage
borrower cannot go bankrupt, but it does strictly limit the borrowers freedom to incur
other debt. The mortgage borrower must be a special purpose entity (SPE), owning
only the mortgaged property and, except for trade debt, owing only the mortgage
lender. The second mortgage loan, once truly the classic form of real estate mezzanine
debt, became virtually extinct as a means of investment in institutional quality CRE.
Before the CMBS era, second mortgage loans, although not popular with conventional
mortgage lenders, had been tolerated if the junior mortgagee agreed to subordinate
completely its lien on the mortgagees shared collateral (Forte 2002). For CMBS loans,
the rating agencies anathematized all junior mortgage debt. They found unacceptable
the risk that the holder of a defaulted second mortgage could precipitate a bankruptcy
of the mortgage borrower, where a reorganization plan might stretch out payments on
the loan for years (Standard & Poors 2003).
The CMBS-blessed form of mezzanine loan mimics a second mortgage loan by re-
quiring collateral: namely, a pledge of the ownership interests in the property owner or,
for a junior mezzanine loan in a multi-tiered financing, in a more senior mezzanine bor-
rower entity. While the pledge has first priority, in that any subordinate pledge of the
equity collateral is strictly forbidden, the mezzanine lender is practically subordinated
to any obligations of the property owner, since by enforcing the pledge the lender only
steps into the shoes of an equity investor in the property owner (Robins et al. 2012).
Structural subordination means the mezzanine lender comes behind every creditor of the
property owner whether secured or unsecured.
CMBS rating agencies and mortgage lenders are not content, however, to rely solely
upon structural subordination of mezzanine debt. They also require that mezzanine
lenders enter a lengthy, and largely standardized intercreditor agreements contractu-
ally subordinating their payment and other rights to mortgage lenders. Whenever the
Mezz anin e C apit al an d R e al E s t at e 91

senior loan is in default beyond applicable cure periods, a mezzanine lender must agree
to forego payments on its loan including payments from guarantors or other third par-
ties, and to limit its rights to foreclose on its collateral. The CMBS standard intercredi-
tor agreement does confer some benefits on the mezzanine lender. One benefit is the
right to foreclose and become the owner of the mortgage borrower without triggering
acceleration of the mortgage debt, provided that the mezzanine lender can satisfy cer-
tain conditions including replacing any borrower guaranties that lapse by reason of such
foreclosure. Another benefit is a grant of limited rights to cure defaults on the mortgage
loan before they mature into events of default that permit the mortgage lender to fore-
close on the property thus wiping out the mezzanine capital. A third benefit is the right
to purchase the defaulted mortgage loan at par. A recent New York court ruling, though,
forbade the mezzanine lender from foreclosing unless it first cured all of the senior loan
defaults, which in this instance meant paying off the senior loan. If followed elsewhere,
this decision would collapse the mezzanine lenders rights into but onethe right to
buy the mortgage loanwhenever the mortgage loan was in uncured payment default
(Prendergast 2012).

EQUITY
Preferred equity is the other common form of real estate mezzanine investment. Be-
tween the two forms, preferred equity seems to have been under-employed. Preferred
equity is the real estate equivalent of preferred stock: among equity interests it has a
prior legal right to a return on its investment and, upon liquidation, a prior right to
the return of that investment. When coupled with a mandatory redemption right, as is
often the case with such a real estate investment, preferred equity bears strong resem-
blance to a bond.
The distinction between mezzanine debt and preferred equity is more legal form
than economic substance. Preferred equity investments are often designed to provide
functionally the same rights and remedies as mezzanine loans, while mezzanine loans
rest entirely on an equity interest. As Table 6.1 illustrates, preferred equity investments
can replicate almost every important characteristic of mezzanine debt.
The comparison in Table 6.1 shows that a mezzanine loan secured by a pledge of
the equity in the property owner is no more secure than a direct, preferred interest
in that selfsame equity. An even more fundamental identity exists between real estate
mezzanine debt and preferred equity investments: the nature of the economic risk as-
sumed. True asset-based lending such as a mortgage relies on the liquidation value of
the loan collateral, avoiding risks associated with the borrowers conduct of its business
and, except in the worst circumstances, risks of market volatility in asset prices. Equity
investments, conversely, are underwritten based largely on predicted future operating
cash flows and asset appreciation, both of which often depend on the skill and experi-
ence of management of the asset. Exactly where an investment in commercial real prop-
erty crosses the line between debt-like risk and equity-like risk varies with the nature of
the property, but as a generality once the ratio of investment amount to property value
exceeds two-thirds the risk profile begins to change. In CMBS financings, 67 percent
LTV just about demarcates the limit for senior rated mortgage debt while mezzanine
finance usually occupies areas of the capital stack below this depth.
Table 6.1 Comparison of a Mezzanine Loan and Preferred Equity
Right or Remedy Mezzanine Loan Preferred Equity
Repayment Mandatory repayment at Mandatory redemption right
specified maturity date at specified date
Return Fixed or floating interest Fixed or floating preferred
rate (may also have an return rate (sometimes with
equity kicker, but usually additional equity kicker)
does not)
Priority Structurally subordinate Subordinate to all property
to mortgage and all other owner debt, before
property owner debt, common equity as to return
prior to common equity on (and often return of)
in property capital
Control of major actions Covenants in loan Approval rights embedded in
documents; action entity governing documents,
without mezzanine lender legally disabling entity
consent from taking action without
is an event of default preferred equity holder
consent
Protection against Bankruptcy filing by Requirement for preferred
bankruptcy property owner triggers equity holder approval is
full recourse for loan under embedded in the property
a non-recourse carve-out owners governing
guaranty from a document, disabling it from
creditworthy affiliate of being able to file a voluntary
property owner bankruptcy petition
Recourse Borrower is SPE whose Recourse against other
only asset is 100 percent members in borrower entity
of equity in property- (also likely, though, to be
owning entity already SPEs). Recourse against
pledged to lender, so an affiliate under a put
personal recourse against and indemnity agreement
borrower is meaningless. (similar to a guaranty) with
Recourse against affiliated, right to put the investment to
creditworthy guarantor the indemnitor triggered by
for bad acts the same kinds of bad acts
Security Pledge of equity interest Direct holding of equity
in property-owning entity interest in property-owning
entity
Remedy for breach Foreclose on equity Take over management of
interest in property owner property owner and force a
sale of property

Source: Adapted from Heller (2012, pp. 4243) with permission from the Stanford Journal of Law,
Business and Finance.
Mezz anin e C apit al an d R e al E s t at e 93

HYBRID FORMS OF MEZZANINE INVESTMENT


As the label suggests, hybrid mezzanine investments combine debt characteristicsa
specific rate of return and priority over all distributions to the common equitywith
a participation in any ultimate gain the underlying property earns. A typical example
would be a mezzanine loan or preferred equity investment with a fixed rate of say 12 per-
cent coupled with a 20 percent interest in any residual cash received from the operation
and ultimate sale of the property. Myriad variations exist on this theme: the mezzanine
investors profit participation may kick in only after the common equity has received its
investment plus a specified return, or the mezzanine investors participation percentage
may decrease as the cumulative distributions to it surmount successive return hurdles,
which are often defined in terms of internal rate of return (IRR) targets. The partici-
pation part in these forms of mezzanine investment resembles the traditional money
partner interest in real estate ventures, where, after both partners recovered their cap-
ital and received a specified return, they would share residual profits in ratios that favor
the operating partner.
Other hybrid forms of mezzanine investment found in corporate finance are rare in
real estate finance. An example is a loan that can be converted into stock shares or that
is coupled with an option to buy into the common equity at a pre-set price. These forms
have obvious advantages for investors in public or going-public companies because they
give mezzanine investors an opportunity to take a ride on rising stock prices as well as
to convert debt generating ordinary income into equity producing capital gains. Except
in large portfolio deals, real estate projects rarely tap into public equity markets. Further,
many mezzanine level real estate investors, such as tax-exempt pension funds, are indif-
ferent to the different tax rates. As a likely result, corporate forms of hybrid mezzanine
investments remain exotics among the species of CRE finance.

A-B LOANS
Although not often labeled as such, subordinated participation in senior mortgage debt
is another form of real estate mezzanine investment. In CMBS parlance, this is the B-
piece in an A-B loan (Fastov and Foley 2002). The A-piece is the securitized and rated
part of the loan. The B-piece is either retained by the originator or sold to a third party,
which is often the special servicer for the related mortgage pool (Levidy 2000). Large
financings may also contain C, D, and more alphabetized pieces. As long as the mortgage
loan remains current, the B-piece receives its proportionate share of interest and prin-
cipal payments. When the mortgage goes into payment default, all payments and other
recoveries including from any sale of the property go first to pay off the A-piece.
The B-piece resembles the traditional second mortgage loan in that it is directly se-
cured by the underlying property. Unlike a second mortgagee, the B-piece investor lacks
direct access to either his security or debt obligation. Some B-pieces are evidenced by
a separate instrument, but rating agencies prefer that they be no more than contractual
rights in a participation agreement with the holder of the A-piece, which doubles as
the legal owner of the entire loan (Levidy 2000). Regardless of form of the interest, the
B-piece holder by agreement surrenders all powers of a mortgage lender to the senior
lender and its servicing agent. Only the senior can foreclose or otherwise enforce the
94 m a j o r t y p e s o f p r i vat e e q u i t y

lenders rights on the loan, notice or waive defaults, grant or withhold forbearances, or
agree to modifications (Fastov and Foley 2002).
Contractual limits exist on the seniors authority if not its power to do these things.
Most important, as long as the mortgage property supports some fraction (usually 25
percent) of the B-pieces par value, the holder has the right to name the special ser-
vicer that will take over administration and enforcement of the loan upon a mone-
tary default. Working through the special servicer the B-piece can often direct enforcing
the defaulted loan, though the special servicer is obligated to disregard directions that
would disfavor the A-piece to benefit the B-piece ( Jones 2008).
In the Great Recession of 2008 and aftermath, B-piece holders fared better than mez-
zanine lenders, as would be expected from their more secure position in the debt stack.
Their rights to appoint and direct the special servicers helped in preserving B-piece
values, but these rights sometimes engendered sharp disputes between the pieces. In
the face of an endemic collapse in CRE values, complex CMBS structures proved less
well-fortified against attacks from the more junior classes of the structured debt than
their designers had intended. Nonetheless, because of the usefulness to CMBS origina-
tors in satisfying rating agency and other regulatory requirements, creation and sales
B-pieces are likely to remain important in CMBS financings.

Factors Distinguishing the Forms of Mezzanine


Financing
Leaving B-piece investments aside, the fundamental similarity between mezzanine debt
and preferred equity invites the question: How do investors choose between them? Var-
ious factorssome compelling, others more subjectiveinfluence that choice for a
particular investor. The following section discusses the more common factors.

TA X C O N S I D E R AT I O N S
For certain mezzanine investors, including real estate investment trusts (REITs), foreign
investors, and tax-exempt entities such as pension funds, preferred equity can create tax
problems where the underlying property generates operating income as opposed to
rents. Hotels are a prime example of such properties, but nursing homes, parking garages,
theaters, and even amenity-laden apartment projects can also produce bad income for
REITs or unrelated business income for tax-exempts. Although ways are available to
structure around these tax problems, they require separating the ownership of the prop-
erty from its operation through an operating lease structure that may contort the underly-
ing economics of the investment. For foreign investors, a preferred equity investment may
mean withholding tax and, for some, the loss of the portfolio interest exemption from U.S.
income tax. Mezzanine loans avoid all these unwanted tax results (Heller 2012).
Yet, mezzanine debt poses its own tax problems. The Internal Revenue Code and
its regulations allow REITs to hold mortgage loans, but not other forms of debt invest-
ment. In Revenue Procedure 200365, the Internal Revenue Service states that it will
treat a mezzanine loan as a mortgage for REIT purposes as long as it meets eight criteria.
Mezz anin e C apit al an d R e al E s t at e 95

Since the usual CMBS form of mezzanine loan never satisfies all eight, REITs holding
mezzanine debt remain in income tax limbo.
Further, under various state and local real estate transfer tax laws with New Yorks
among them, a mezzanine loan foreclosure can result in a hefty transfer tax to the fore-
closing lender that will have to be paid again when the lender sells the underlying prop-
erty. A preferred equity investor forcing a sale of the underlying property only bears this
tax once. Finally, for high net worth individuals and other ordinarily taxable investors,
mezzanine debt generates income taxable at ordinary rates, while much of the return on
a successful preferred equity investment will be taxable as a capital gain (Heller 2012).
Conversely, from a tax perspective, most issuers of mezzanine investments are indif-
ferent to whether the investment takes the form of debt or equity. Unlike corporate is-
suers, for which interest payments are tax-deductible but dividends are not, mezzanine
investment vehicles are almost invariably pass-through entities, such as limited liability
companies, and their payments to mezzanine investors, whether interest or dividends,
are excluded from the net income allocated to the common equity owners.

ENFORCEABILITY
Some analysts suggest that streamlined enforcement of legal remedies may be one
reason to prefer mezzanine debt (Rubock 2007). Enforcement of mezzanine debt se-
cured by a pledge of equity collateral is governed by Article 9 of the Uniform Com-
mercial Code (UCC), adopted with little variation in all 50 states, Washington, DC, and
Puerto Rico (Sebert 2011). Upon material default on a mezzanine loan, the lender can
force a sale of its collateral. If the sale satisfies the UCC definition of a public sale, the
lender can bid in up to the full amount owed to it by and thus acquire 100 percent of
the property owner and, indirectly, the property itself. Under UCC provisions for sale
of collateral, the whole process might take as few as 10 days, although a longer period
(such as 30 days) is more prudent (Compton and Fisch 2008).
Preferred equity investments usually bake the investors remedies into the property-
owning entitys governing documents. Commonly, these remedies consist of the right
to (1) take over management of the property owner, (2) direct the entity to sell the
property, and (3) trigger a buy-sell in which the common equity holder in the prop-
erty owner must either buy out the preferred investor or sell out to him, in either case at
a price derived from whatever value for the underlying property the preferred investor
specifies. Events that trigger the preferred equitys remedies always include a contri-
bution default or other material breach of the property owners partnership or limited
liability company agreement by the common equity holder or its surrogate manager.
Other typical triggers are the property owning entitys inability to make required pe-
riodic distributions to the preferred equity or to redeem the entire investment by the
mandatory redemption date or, for to-be-improved properties, the failure to meet speci-
fied development, construction, or lease-up benchmarks.
Preferred equity enforcement rights are, on paper, self-help remedies and so appear
even more streamlined than a UCC sale (Robins et al. 2012). If, however, the common
equity holder proves uncooperative (e.g., refusing to acknowledge the investors take-
over of the property owners management or contesting the right to cause a sale of
the property in statements to prospective buyers or title insurers), the investors only
96 m a j o r t y p e s o f p r i vat e e q u i t y

recourse is litigation, which is often expensive and protracted. Active opposition may
not even be necessary: just by declining to acknowledge the preferred equity holders
authority the common equity may prevent a sale.
Such criticism of preferred equity remedies does not miss the mark, but a contention
that mezzanine lender remedies are clearly superior does. Assuming the borrower does
not attempt to block the UCC sale, the foreclosing mezzanine lender ends more or less
where the preferred equity investor begins, namely, holding a controlling equity interest in
the property owner. If the borrower chooses to contest the sale, the mezzanine lender can
also wind up in protracted litigation and unable to exploit its control effectively. The UCC
provides a resisting mezzanine borrower with entirely arguable grounds for challenging
the sale: the requirement that any sale under the UCC be commercially reasonable. Both
the UCC text and the official comments on it are not particularly helpful in defining
commercially reasonable. Yet, an important test for a public sale (the only kind where the
lender itself can purchase the collateral) is that the sale be advertised and otherwise con-
ducted so as to ensure that the sale is well attended by legitimate bidders (Heller 2012).
The catch is that the appearance of any third-party bidder at an UCC auction of the
equity interest in a private entity is rare. Applying the well-attended test would make
a commercially reasonable public sale of such an interest commercially impossible. A
mezzanine lender might try to avoid objection to an unattended sale by obtaining a
credible appraisal of the underlying property to show that the amount of its debt was no
less than the net equity in the property, meaning the lender could with its debt outbid
any other rational bidder anyway. Given that a typical mezzanine loan represents no
more than 10 to 15 percent of the underlying propertys gross value, though, whether
the debt exceeds net equity value may be well within the margin for valuation error,
permitting the borrower to engage in combat by appraisal.
In terms of practical enforcement, the presumed advantage of mezzanine debt over
preferred equity has never been tested extensively in practice and in theory appears
mostly unexamined conjecture (Rubock 2007). Because preferred equity remedies do
not deprive the common equity of all interest in the underlying property or carry the
stigma of foreclosure, one can even argue that such remedies are less likely to encounter
vigorous opposition.

FIDUCIARY DUTY
Another legal point often made in favor of mezzanine debt is that fiduciary duties to the
common equity may hinder enforcement of a preferred equity investment. This is partly
true because a foreclosing mezzanine lender does not owe its borrower any duty of care
and loyalty, of the kind that directors and officers owe to corporations. Also true is that,
under general legal principles, partners in a partnership owe these duties to each other
and a manager of a limited liability company owes them to the non-manager members.
By extension, a preferred equity holder who takes over management of an entity could
be held to owe fiduciary duties to the common equity, thus inhibiting its freedom of
action and exposing it to potential liability (Robins et al. 2012). Since about 2000,
however, limited liability company and limited partnership statutes have substantially
increased the freedom of contracting parties to disclaim fiduciary duties of partners,
members, and managers. This is particularly true in states such as Delaware and Nevada
Mezz anin e C apit al an d R e al E s t at e 97

seeking to attract the formation of business entities to their jurisdictions. When a de-
cision of the Delaware Supreme Court restricted the statutory language to allow only
moderation and not elimination of such duty, the Delaware legislature quickly amended
the statute, leaving no doubt as to the broad grant of freedom of contract it intended.
A preferred equity investor must take care that the investment receptacle is formed in
one of the lenient jurisdictions (Delaware still being best because the statute and the
case law are now aligned) and that the limited liability company or limited partnership
agreement contains the requisite disclaimers of fiduciary duties. If these precautions are
observed, though, avoiding fiduciary liability is no longer a sound reason to choose a
mezzanine debt investment form over preferred equity (Heller 2012).

S TA N D A R D I Z AT I O N
The centripetal forces that led to a high degree of document standardization in CMBS
transactions, including the concentration of most of the initial CMBS players within
a few square miles in Manhattan, also resulted in standardization of mezzanine loan
documentation. In theory, documentation that remains relatively consistent from deal-
to-deal saves considerable time and money otherwise consumed in the negotiation of
documents. Standardization also facilitates secondary market transactions including ex-
pediting rating agency review of the transaction. This explains why the most standard-
ized of all mezzanine debt documents is the intercreditor agreement with the senior,
rated debt. Understandably, standardized documents made mezzanine debt a favorite
form of mezzanine real estate investment for CMBS sponsors (Heller 2012).
Standardized documentation may also explain some of the appeal that mezzanine
debt has had for investors seeking an efficient means of investment. Preferred equity
documentation, though often fundamentally similar, is diverse in form and language.
Most importantly, no generally recognized form of intercreditor-type agreement be-
tween preferred equity holders and mortgage lenders exists and mortgage lenders have
not adopted a standard response to requests by preferred equity investors for such a
recognition agreement. Unquestionably, the lack of standardization complicates doc-
umentation of preferred equity transactions, but less certain is whether this eventually
results in a more efficient transaction. A thorough negotiation of documents can iden-
tify and resolve issues long before they can ripen into a source of seriously disappointed
expectations and litigation. Standardized documents all too often paper over potential
issues until they occur, and flaws in the original document forms replicate themselves in
every new transaction (Heller 2012).

Summary and Conclusions


In the last two decades, mezzanine capital has become an important, even critical, com-
ponent in CRE financing, particularly in property portfolio acquisitions and financing
of larger properties that require development, rehabilitation or re-purposing, or lack a
current, stabilized operating cash flow. With the retreat of traditional balance sheet
mortgage lenders such as banks, S&Ls, and insurance companies that began in the mid-
1980s, real estate entrepreneurs faced a sizable gap between the first loss equity they
98 m a j o r t y p e s o f p r i vat e e q u i t y

could muster for their projects and the debt financing available. Hedge funds and other
PE investors, sensing an opportunity to bend the risk-reward curve in their direction,
offered new sources of mezzanine CRE financing. CMBS lenders, needing someone to
plug the hole in the capital stack created by strict rating agency LTV requirements, cre-
ated an attractive vehicle for high-volume mezzanine real estate financing in the form of
the modern real estate mezzanine loan.
By 2007, outstanding CRE debt was in the hundreds of billions of dollars. Five years
later, losses on such debt also reached well into the billions. In one highly publicized
deal alone, specifically the 2007 purchase financing of the huge Peter Stuyvesant Town/
Peter Cooper Village residential project in lower Manhattan, the mezzanine lenders lost
$1.4 billion (Prendergast 2012).
Losses that large quickly took the bloom off the rose and, despite the early successes
of mezzanine lenders and the resurgence of CMBS financing after 2012, mezzanine lend-
ing has not regained its earlier luster. Preferred equity investments, once rare in CMBS
financings, have become more common but remain far from the levels mezzanine debt
attained before 2008. Whatever the form, mezzanine investment is likely again to play
an important role CRE financing, as long as mortgage lenders remain conservative in
their valuations and LTV ratios, mortgage interest rates remain low, and large investors
require uniform rates of return that other debt-like investments cannot yield.

Discussion Questions
1. Identify the hallmark of mezzanine investment and discuss some other important
characteristics.
2. List the factors leading to the upsurge in mezzanine investment in real estate during
the 20-year period preceding 2008 and identify which still apply today.
3. Compared to mortgage debt, identify the primary risks inherent in mezzanine in-
vestments in real estate and discuss how they differ between preferred equity or
mezzanine loans.
4. Standardization of documentation appeared to be one feature of mezzanine loans
attracting investors such as foreign banks. Discuss whether such packaging of invest-
ment products should weigh heavily in the choice of mezzanine investment forms.
5. Discuss how the rating agencies shaped CRE mezzanine investment.
6. Given that rating agency guidelines favor preferred equity over mezzanine loans in
rating CMBS debt, explain why CMBS lenders so strongly favored mezzanine debt
as the mezzanine level in their structured financings.

References
Babson Capital. 2010. Middle Market Mezzanine Debt. White Paper, Babson Capital Manage-
ment, LLC. Available at https://www.babsoncapital.com/BabsonCapital/http/bcstaticfiles/
Research/file/Babson%20Capital%20Mezz%20Middle%20Market%20WP.pdf.
Benmelech, Efraim, and Jennifer Dlugosz. 2010. The Credit Rating Crisis. NBER Working Paper
Series 15045. Available at http://www.nber.org/chapters/c11794.pdf
Mezz anin e C apit al an d R e al E s t at e 99

Compton, S. H. Spencer, and Peter E. Fisch. 2008. Foreclosing on a Mezzanine Loan under UCC
Article 9. First American Title. Available at http://www.firstamny.com/detail.aspx?id=15.
Egan, Matt. 2014. Spotlight: CMBS Grew in 2013, Another Climb Expected this Year. Available at
http://commercialobserver.com/2014/01/spotlight-cmbs-grew-in-2013-another-climb-
expected-this-year/.
Fastov, Jeffrey, and Robert Foley. 2002. Subordinate Financing of Commercial Real Estate. CMBS
World 4:1, 1216. Available at http://crefc.org/assetlibrary/00C7066F-C722-4408-B235-
A94685C9CC02/ae3dc2ec685d4a319e5d45a30f87f82d2.pdf.
Financial Crisis Inquiry Commission. 2011. The Financial Crisis Inquiry Report. Washington,
DC: U.S. Government Printing Office. Available at http://www.gpo.gov/fdsys/pkg/GPO-
FCIC/pdf/GPO-FCIC.pdf.
Forte, Joseph P. 2002. Mezzanine Finance: A Legal Background. CMBS World 4:1, 2025. Availa-
ble at http://www.crefc.org/crefinanceworld/crefinanceworld_toc.aspx?folderid=1654.
Forti, David W., and Timothy A. Stafford. 2002. Mezzanine Debt: Suggested Standard Form of
Intercreditor Agreement. CMBS World 4:1, 2627. Available at http://www.crefc.org/
crefinanceworld/crefinanceworld_toc.aspx?folderid=1654.
Heller, J. Dean. 2012. Whats in a Name: Mezzanine Debt Versus Preferred Equity. Stanford Journal
of Law, Business and Finance 18:1, 4071.
Jones, Richard D. 2008. Working Out Mortgage Loans and B Notes. Working Paper presented to
the American College of Real Estate Lawyers, October 2008. Available at http://www.acrel.
org/Documents/Seminars/Jones-Working%20Out%20Mortgage%20Loans%20and%20
B%20Notes%20(Outline%20I).pdf.
Levidy, Nicholas J. 2000. CMBS: Moodys Approach to A-B Notes and Other Forms of Subordi-
nate Debt. Moodys Investors Service.
Nijs, Luc. 2014. Mezzanine Financing. Chichester, U.K.: John Wiley & Sons.
Plucknett, Theodore F. T. 1956. A Concise History of the Common Law. Boston, MA: Little, Brown
and Company.
Prendergast, James D. 2012. Foreclosure Becomes More Difficult for Mezzanine Lenders. Practical
Real Estate Lawyer 28:2, 5157.
Robins, Jon S., David E. Wallace, and Mark Franke. 2012. Mezzanine Finance and Preferred Equity
Investment in Commercial Real Estate: Security, Collateral & Control. Michigan Journal of
Private Equity & Venture Capital Law 1:1, 93162.
Robinson, Arthur D., Igor Fert, and Mark A. Brod. 2011. Mezzanine Finance: Overview. Practice
Note, Practical Law Company.
Rubock, Daniel B. 2007. US CMBS and CRE CDO: Moodys Approach to Rating Commercial
Real Estate Mezzanine Loans. Moodys Investors Service. Available at http://www.firstam.
com/assets/ucc/articles/moodys-3-2007.pdf.
Sebert, John A. 2011. Report of the Permanent Editorial Board for the Uniform Commercial Code:
Application of the Uniform Commercial Code to Selected Issues Relating to Mortgage Notes.
The American Law Institute and the National Conference of Commissioners on Uniform State
Laws. Available at http://www.uniformlaws.org/Shared/Committees_Materials/PEBUCC/
PEB_Report_111411.pdf.
Silbernagel, Corry, and David Vaitkunas. 2012. Mezzanine Finance. Available at http://www.
salvador-montoro.com/uploads/3/2/0/7/3207272/mezzanine_finance_12.pdf.
Standard & Poors. 2003. U.S. CMBS Legal and Structured Finance Criteria. Standard & Poors.
7
Distressed Debt Investments
STEPHEN G. MOYER
President, Distressed Debt Alpha and Adjunct Professor, University of Southern California

JOHN D. MARTIN
Carr P. Collins Chair of Finance, Baylor University

Introduction
The involvement of private equity (PE) firms in distressed debt investment has grown in
the last decade in tandem with the expansion of the distressed debt market. Distressed
debt investing arguably began as a consequence of the adoption of the Bankruptcy Act
of 1978, which introduced a rehabilitation ethic into the corporate reorganization
process in the United States. While this development is discussed in more detail later in
the chapter, the new legal scheme essentially fostered a restructuring framework dedi-
cated to preserving the going concern value of bankrupt businesses. It also improved
the rights of creditors and simplified the process by which creditors claims could be
restructured, often into a controlling share of the distressed companys equity.
The rapid development of the high-yield or junk bond market followed this legal
development, but this does not necessarily suggest a causal link. The growth of the junk
bond market facilitated two important preconditions for what is now a vibrant dis-
tressed investment asset class. First, it created large issues of tradable bonds and later
loans that were accessible investments for institutional investors. Second, it facilitated
higher use of financial leverage in corporations that inevitably increased the likelihood
that they would experience financial distress (Baribeau 1989; Bernanke, Campbell, and
Whited 1990).
As corporate default rates increased, shrewd investors began to realize that inefficien-
cies in this nascent market created investment opportunities. Specialized investment
funds were often structured similar to PE funds, in terms of the long-term investment
commitment required of investors, and began to post attractive returns. This develop-
ment prompted institutional investors to view distressed debt as a separate alternative
asset class. PE funds, particularly those that specialized in leveraged buyouts (LBOs)
and thus had considerable experience with complex debt capital structures, recognized
distressed debt represented an expansion and diversification opportunity. Many estab-
lished specialized distressed funds in addition to their core buyout funds. Distressed

100
Di s t re s s e d De bt I n v e s t m e n t s 101

debt was also extremely profitable for limited partner (LP) investors. Based on the Pri-
vate Equity Quarterly Index (PrEQIn), distressed debt was the top performing PE strat-
egy between 2001 and 2011 (Preqin 2012).
This chapter surveys the role of PE investing in distressed debt markets. Next, the
historical evolution of the market for distressed debt is discussed. This is followed by a
short primer on distressed investing since this type of investing is relatively new and less
studied than other active investing strategies. Next the primary investment strategies
distressed for control, loan to own, special situations, and turnaround investing
employed by distressed-asset PE firms are reviewed. Finally, the current state of
distressed investing is analyzed including the growing efficiency of the U.S. market, the
expansion into less efficient markets such as Europe, and the evolution of the buyout PE
sponsor as a potentially important player in the distressed market place.

Historical Evolution of the Market for Distressed Debt


Distressed debt investing would not have evolved into a dynamic asset class without
an accommodative legal framework that facilitated the efficient restructuring of finan-
cially distressed corporations. The enactment of the Bankruptcy Act of 1978 (the 1978
Act), the first comprehensive overhaul of restructuring legislation in the United States
in roughly 40 years, provided such a construct.
Chapter 11 of the 1978 Act was crafted explicitly for rehabilitating financially dis-
tressed corporations with a goal of preserving businesses and employment while still
protecting the rights of creditors and other claimants (Miller and Waisman 2005).
Among the manifold changes, Miner (1979) identifies five of the more important
changes:

The 1978 Act eliminated a prior requirement that the debtor must demonstrate it
was insolvent (e.g., miss an interest payment) before it was eligible for bankruptcy
protection. This change allowed firms to voluntarily file for bankruptcy much earlier
in their financial distress allowing them to start the rehabilitation while they were
merely wounded, as opposed to on their deathbed.
It established that leaving management in place, rather than an appointed trustee,
better-preserved value by allowing those with the most knowledge of the business to
remain in control.
The 1978 Act expanded protection against secured creditors seizing their collateral
and thereby effectively forcing a liquidationconsider what would happen to an
airline if all the aircraft could be repossessed by a secured creditor.
It provided that essentially all debts and claims against the debtor could be compro-
mised or eliminated, allowing the debtor to have a genuine fresh start.
It exempted securities distributed to creditors from certain provisions of the securi-
ties laws that made reselling the securities difficult in many cases, which substantially
increased risk and reduced returns.

Procedurally, the 1978 Act established a more efficient framework for negotiating
and approving the plan of reorganization (generally referred to simply as plan). The
102 m a j o r t y p e s o f p r i vat e e q u i t y

streamlined Chapter 11 process envisioned that the company and an official committee
of unsecured creditors would negotiate a proposed plan and then creditors would vote
to accept or reject their treatment under the plan. This change was a sharp contrast to
the process that was previously followed in many cases where a trustee developed a
plan, the creditors were limited to making suggestions, and the Securities and Exchange
Commission (SEC) had an important role in the approval process (Miner 1979). Taken
as a whole, the 1978 Act represented a dramatic advancement in the U.S. bankruptcy
process.
The next step in the evolution of the distressed market was the expansion of junk
bond financing by Michael Milken and his firm, Drexel Burnham Lambert. Before
Milken, the only below investment grade debt available for distressed investors was
from fallen angels (i.e., investment grade companies that had fallen on hard times and
been downgraded to below BBB/Baa), which were fairly uncommon. In the mid-1980s,
Milken developed the market for high yield bonds and issuance grew dramatically. As
Figure 7.1 illustrates, between 1980 and 1990 the market grew tenfold from less than
$20 billion to more than $200 billion.
Since junk bond issues are riskier by definition, corporate default rates started to in-
crease. Altman and Hotchkiss (2006) document that bond defaults increased from $1.4
billion in the decade ending in 1980 to $43.3 billion for the decade ending in 1990. As
default rates increased, specialized investors identified opportunities to purchase the
bonds of companies experiencing financial distress. Often the original investors, who
initially were insurance companies and pension plans, had little experience with the cor-
porate reorganization process. They were often fearful that if the issuer of the bonds filed
for bankruptcy, their investment might be totally lost. As a result, when the original
buyer learned that they could mitigate risk by selling bonds, albeit at steep discounts
to the original principal amount, they often hit the bid. Initially, the buyers of this
distressed debt were relatively small hedge funds. Their basic strategy was to identify
undervalued bonds and then hope some catalyst would occur to cause the value of the
bond to increase and they would exit. Their skill was identifying the misvaluation and
understanding the workout process. This allowed them to predict events that would

300

250

200

150

100

50

0
1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990
HY Bonds Outstanding ($MM)

Figure 7.1 Size of the Early High Yield Market This figure shows the amount of
high yield (HY) bonds between 1980 and 1990 based on data from Credit Suisse.
Di s t re s s e d De bt I n v e s t m e n t s 103

result in the security appreciating in value. Generally buyers did not make the invest-
ment with the goal of owning or controlling the distressed company and they did not
have the expertise or resources to create value through fundamentally improving the
companys economic performance.
Over time some notable exceptions to this generalization emerged. The biggest was
Oaktree Capital Management, founded by Howard Marks. Markss value strategy was
similar to his smaller predecessors, but Oaktree was among the first that could raise
longer term, locked-up PE-like capital. This allowed the firm to take large, illiquid posi-
tions in distressed firms, often with the goal of influencing the course of the restructur-
ing process. Capitalizing on the 1990 recession, Oaktrees 1988, 1990, and 1991 vintage
distressed funds posted an average net return of 30 percent (Oaktree Capital Group,
LLC 2011). However, Preqin (2011) data indicate in Figure 7.2 that returns for dis-
tressed investments for the mid-1990s were unexceptional largely because the 1990s
were generally a period of economic expansion with relatively low corporate default
rates. This changed with the turbulence that occurred from 1998 to 2002. Many funds
that focused on distressed investing during that period capitalized on rising default rates
and earned returns in excess of 20 percent.
Another observation from Figure 7.2 is that fund returns vary dramatically by vintage
and returns are typically best during economic recessions. The basic reason for these ob-
servations is that during a recession valuations decline and investors become cautious,
tending to sell investments that they view as risky. This selling allows distressed-asset
investors to acquire assets at attractive valuations. As Warren Buffet famously said, Be
fearful when others are greedy and be greedy when others are fearful (Buffett 2008).
What tends to set successful distressed asset managers apart is their contrarian courage,
that is, they are willing to make investments during bad market environments precisely

30%

25%

20%

15%

10%

5%

0%
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
Median Distressed Fund Return

Figure 7.2 Median Distressed Fund Return This figure shows the median
cumulative IRR return for distressed funds by vintage between 1997 and 2008. The median
cumulative return for a fund launched in 2002 was about 27 percent compared to a fund
launched in 2006, when it was less than 10 percent. Source: Preqin (2011).
104 m a j o r t y p e s o f p r i vat e e q u i t y

because they realize that misvaluations have occurred and that they can take advantage
of them.
The persistence of attractive returns began to draw the attention of institutional in-
vestors who started to recognize distressed investing as a specific asset class or strat-
egy. This asset class also had the added benefit of being somewhat countercyclical in
that distressed returns were often best during periods of economic contraction when
equity markets typically fell in value (Stus 2011). Buyout style PE firms such as Carlyle
and Apollo took notice, recognizing that developing distressed debt focused funds pre-
sented a potential growth opportunity. This ushered in the modern era of substantial PE
participation in the distressed debt arena.

A Brief Primer on Distressed Investing


Since distressed investing is a relatively new investment strategy and not as often dis-
cussed as, for example, LBO or activist investing, briefly explaining the process used by
distressed investors and how they create investment value could be helpful.
First, funds that have evolved to dominate distressed investing are structured similar
to the well-known buyout PE funds. Managers solicit investments from large institu-
tional investors or high net worth individuals in a limited partnership. The manager acts
as general partner (GP) and has an investment management contract with the limited
partnership. These contracts typically provide that the manager is entitled to an annual
management fee of 1.5 to 2.0 percent on committed capital and a profit participation
(commonly referred to as the carry or carried interest) of 15 to 20 percent on invest-
ment profits after a minimum investment return (hurdle rate, typically 8 percent) are
reached (DePonte 2010). The capital commitment made by the limited partnership is
called pursuant to draws made by the manager at intervals dictated by the pace of invest-
ment. The funds usually have a 10-year life with a 5-year investment period.
Several practical differences from buyout funds also exist. First, while the classic
buyout fund invests in the investees (target) equityhence the characterization as
PEdistressed-asset funds are investing in the targets debt securities. Before 2004
this was typically a targets unsecured junk bonds. However, with the growth of the
leveraged loan market over the last decade, increasingly the investment is made in the
targets syndicated leveraged loan, which is usually secured by the targets assets. Invest-
ing in debt securities at a discount tends to reduce downside risk because recoveries
on debt investments are not typically zero, whereas they certainly can be with equity
investments in LBOs. A second consequence of investing in debt instruments is that
funds earn interest income for part or all of the investment period. Further, since the
instruments are purchased at a discount to face value (hereinafter discount), the cur-
rent yield on the investments can be substantial. For example, the current yield on a 10
percent coupon bond purchased at 60 percent of face value is 16.6 percent. Although
since the target is in financial distress, a substantial risk exists that a default will occur
and interest payments will cease at some point.
While several common investment strategies are outlined in the following section,
the basic process is to identify a target whose valuation has declined; resulting in its
capital structure concurrently declining in value. To be more realistic about the process,
Di s t re s s e d De bt I n v e s t m e n t s 105

the manager observes that the targets bank loans and/or bonds (sometimes collectively
referred to as debt securities) trade at large discounts, and analyzes the potential re-
turns from alternative scenarios, including a restructuring. Large companies with com-
plicated, multi-layer capital structures (e.g., secured bank loans, senior unsecured notes,
and subordinated unsecured notes) may present the distressed-asset investor with mul-
tiple investment choices.

I L L U S T R AT I N G T H E D I S T R E S S E D D E BT I N V E S T I N G P R O C E S S
A simple hypothetical example illustrates the investment process and strategies.
Assume that two years ago an LBO of TargetCo took place. At the time of the deal, Tar-
getCos last 12 months earnings before interest, taxes, depreciation, and amortization
(EBITDA) was $200 and comparable companies were trading at 7.0x EBITDA so the
LBO price was $1,400. The acquisition was financed with $400 in equity and $1,000
in debt as detailed in Table 7.1. Now the economy is in recession and TargetCo has
additional firm-specific challenges. Its EBITDA falls to $150 with a negative trend. Fur-
ther, market comparables now trade at 6.0x to 6.5x EBITDA. Reflecting these changed
circumstances, assume TargetCos debt securities are trading at the discounts listed. The
rationale for why any of these prices may or may not be appropriate is beyond the scope
of this overview. Securities generally tend to reflect their collateral and contractual
rights. Moyer, Martin, and Martin (2012) provide a more complete discussion. Here
bank loans may be adequately secured so holders are unwilling to sell at too big a dis-
count. Conversely, given that the current value of TargetCo may be $975 (6.5 x $150)
at best, its subordinated note holders might be concerned that TargetCo will go into
bankruptcy, and they will not recover their investment. In fact, if EBITDA were to fur-
ther decline to $130, using the more conservative valuation metric of 6.0x EBITDA the
implied value of TargetCo will only be $780 (6.0 x $130) and the subordinated notes
might not receive any recovery considering the priority status of the aggregate $800 in
bank loan and senior notes.
A distressed-asset investor analyzing TargetCo might see several opportunities. An
investor with a very negative outlook might want to be risk-averse and invest in the bank
loan. If the bank loan only had a three-year remaining term, the expected return would

Table 7.1 TargetCo Debt Capital Structure

EBITDA 150
Coupon Security Amount Price Coupon Leveraged Leveraged
($ millions) Yield (%) Face Market
L + 3% Bank loan 500 80 6.3 3.3x 2.7x
7.00% Senior notes 300 50 14.0 5.3x 3.7x
9.00% Subordinate notes 200 20 45.0 6.7x 3.9x
Total debt 1,000

Source: Authors.
106 m a j o r t y p e s o f p r i vat e e q u i t y

be around 13 percent. Yet, a more optimistic investor who perceived ways of improving
TargetCos performance might conclude attractive investment returns are achievable by
purchasing control of TargetCo at its current valuation. In simple terms, one strategy
to accomplish this would be to aggressively purchase the senior and/or subordinated
notes and then devise a way to essentially force TargetCo into a bankruptcy or other
restructuring. During this restructuring process, the investor would seek to convert or
exchange the distressed funds senior and/or subordinated notes for a controlling share
of the equity, thereby reducing TargetCos debt load to perhaps only the bank loan and
improving its financial viability. If the reorganization is done within a bankruptcy con-
text, the new equity is referred to as the post reorganization equity. After attaining
control, the investor will then apply performance improvement and financial engineer-
ing skills to enhance TargetCos value and then exit, perhaps via an initial public offering
(IPO) or sale of the company. This scenario is essentially the classic buyout PE strategy,
except instead of gaining control directly by purchasing TargetCos equity, the distressed
investor does it indirectly by first investing in the targets debt securities and then using
the reorganization process to convert the debt into equity.
Regardless of strategy, distressed-asset investors need to be flexible in their approach
because many aspects of the investment process are often outside of their control. For
example, since relatively small issues of debt securities are often fairly illiquid, the man-
ager might start purchasing notes but quickly push up that market price such that the
risk-adjusted investment return is no longer attractive. As was increasingly the case in
the 2009 distressed cycle, the original PE firm that sponsored the LBO might get ac-
tively involved, financially and otherwise, and effectively thwart the distressed funds
efforts to force a restructuring. While the returns can be very attractive, distressed in-
vesting is a complex, volatile, and risky process.

Strategies Used by Private Equity Firms for Investing


in Distressed Situations
PE firms use several strategies in the distressed-asset arena. Sometimes funds specialize
only in one approach that the manager believes may be a special aptitude for the fund.
In other instances, funds engage in multiple approaches. Still other times, one strategy
may be initially attempted, but later combined with other approaches as the situation
evolves. The four main strategies explored here are (1) distressed for control, (2) loan-
to-own, (3) special situations, and (4) turnaround investing.

DISTRESSED FOR CONTROL


Distressed for control is a strategy in which a fund purchases a targets existing debt with
the intent of controlling the target at the conclusion of the investment/workout proc-
ess. This strategy is analogous to a hostile takeover except that instead of purchasing
the targets stock and then perhaps launching a tender offer to acquire the majority of
the shares, the PE firm purchases the targets debt securities with the aim of converting
the debt securities into a large or controlling interest in the targets post-reorganization
Di s t re s s e d De bt I n v e s t m e n t s 107

equity. An important advantage distressed-asset funds have in their quest for control
versus an equity hostile takeover is that they do not need to disclose their accumulation
of debt securities to the market (Harner 2011). After gaining control, the PE firm then
applies its normal strategic, financial engineering, and process improvement skills to at-
tempt to improve the valuation of the target before exiting. This strategy was illustrated
in the previous section.
Distressed for control has had more investment dollars allocated to it than any other
distressed strategy and tends to involve the largest funds (Preqin 2011). This is partly
driven by the funding requirements of the strategy. To achieve control over a target,
particularly a large target, often necessitates purchasing more than $100 million in face
amount of securities which requires a substantial amount of investment capital. If the
fund intends to remain reasonably diversified, it usually must be larger than $1 billion in
size. However, smaller distressed for control funds can exist. These funds usually target
smaller, middle market companies that do not require as much capital in order to apply
the strategy.
As alluded to previously, distressed for control managers face challenges and risks.
The first is accumulating a sufficient amount of the appropriate debt securities to gain
control. This can be challenging both because most bond issues are inherently less
liquid than stocks and because more than one fund may be pursuing the same distressed
target. This often leads to a competitive situation in which neither manager can accu-
mulate as much as they would like without driving up prices. Often, if the styles and
objectives of both managers are compatible, they may informally partner and agree on
a common strategy. If they are successful in forcing a reorganization, the managers may
be jointly involved in the restructuring negotiations and subsequent rehabilitation of
the target. The partnering, or club, approach has the added benefit of allowing managers
to mitigate risk through reducing the level of financial exposure in a single investment.
Adding to the challenge of accumulating enough securities is the problem of deter-
mining which securities will receive the post-reorganization equity. Referring to the Tar-
getCo example, if the manager chooses to accumulate the cheaper subordinated notes
assuming potentially more upside opportunity, but instead valuations and EBITDA de-
cline, in a subsequent reorganization the subordinated notes may be deemed worthless
(sometimes referred to as out-of-the-money). Even if the result is not that draconian,
the equity may be allocated between the senior and subordinated notes such that the in-
vestor only has a minority equity position. In fact, junior securities are sometimes given
non-equity forms of consideration as their recovery expressly so that a control-minded
creditor in a more senior class does not have to dilute its position (Wachtel 2013).
Finally, some catalyst is needed to effect the reorganization. Management and equity
owners of the distressed firm often have incentives to delay a restructuring as long as
possible. The equity holders, often a buyout PE fund, realize that if the restructuring
is done at a low point in the companys operating performancewhether due to weak
economic conditions or firm-specific issuesthe equity may be out-of-the money
and the restructuring would likely result in a transfer of ownership to the distressed
investors. Of course, the distressed investors would view this as the optimal time for
the restructuring to occur. For distressed-asset investors to force a restructuring proc-
ess, they need some type of negotiating leverage, such as a default in the bond or loan
that will give them the right to accelerate the debt. Therefore, an important strategic
108 m a j o r t y p e s o f p r i vat e e q u i t y

consideration for the distressed investor is when and how the investor will be able to
influence the targets behavior (Moyer 2005).

L O A N - TO - O W N
Loan-to-own is a recent distressed investment strategy in which the distressed investor
intends to own the target. However, instead of purchasing the existing debt securities,
the distressed fund makes a new loan, typically with very expensive and onerous terms,
to the target. For the target to be enticed into this deal, it needs to be in severe distress,
with few alternatives.
Returning to TargetCo, assume that operations continued to deteriorate and that
TargetCo is running out of cash to continue operating. TargetCo will likely have
breached covenants in its bank loan making the lenders nervous and unwilling to
extend any more credit. While normally waivers of a covenant violation are negoti-
ated during the early stages of such a decline, the lenders effectively retain the power
to declare a default, which would cause a bankruptcy. The PE sponsor is now well
out-of-the-money and might deem any more investment as throwing good money
after bad. The distressed-asset fund manager in the example, which for current pur-
poses is assumed not to have purchased any notes, might approach TargetCo with the
following proposal. It offers TargetCo a new 15 percent secured loan in the amount
of $550. Of this total, $500 will be used to pay off the existing bank loan (this is es-
sential in order for the new loan to have first-lien collateral rights) and the incremen-
tal $50 can be used to save the business. This type of funding is sometimes called a
rescue loan.
TargetCo and its PE sponsor accept the loan in a last ditch effort to save the business.
If management can turn things around, they will refinance the rescue loan with some-
thing more reasonably priced at the earliest opportunity. In that event, the distressed-
asset manager fails to gain control but earns an attractive 15 percent or higher return
for relatively little work. However, if the business does not sufficiently recover, then the
terms (e.g., an interest coverage or leverage covenant) of the rescue loan will likely be
breached and the distressed-asset manager will have the power to force a restructuring.
Except, in this scenario, the distressed-asset manager will be arguing that TargetCos
value is much lower and that both the senior and subordinated notes are out-of-the-
money and together with the prior equity should be extinguished. Even if some recov-
ery needs to be given to some other creditor constituencies, the distressed manager is
likely to end up with control and start the process of value enhancement in an effort to
exit later with a substantial gain.

S P E C I A L S I T UAT I O N S
The term special situations is used in many contexts other than distressed debt such as in
equity and real estate contexts. Even within the distressed arena, funds that label them-
selves as special situation funds engage in a broad range of investments. For example,
a fund rarely, if ever, labels itself as a loan-to-own fund. However, many, if not most,
special situation funds would include rescue type loans as special situations they
identify for their investors.
Di s t re s s e d De bt I n v e s t m e n t s 109

For purposes of this discussion, the term special situations is analogous to what in
the equity context, especially merger arbitrage, is usually labeled event driven strate-
gies. The essence of this strategy is to identify securities that are misvaluedfor reasons
other than a misperception about the underlying value of the issuerwith a theory
for what event might occur in the near-future that would cause the security to be prop-
erly valued. This subsequent event is sometimes referred to as the catalyst. It is dis-
tinguished from distressed for control and loan-to-own by the fact that the distressed
manager is not trying to obtain the post-reorganization equity. Instead, the strategy is
to profit from buying the misvalued security, waiting for or causing the catalyst to occur
and then exiting the investment with a profit.
Usually the situations are highly technicalotherwise the average market partic-
ipant would recognize the misvaluation and drive the price to fair value. Returning
to TargetCo, perhaps the distressed-asset investor during due diligence, identifies an
error in the legal documentation relating to the collateral pledge such that the manager
concludes the bank loan is effectively unsecured. Then the investor might invest in the
senior notes on the theory that when the restructuring event (the catalyst) subsequently
occurs, the unsecured creditors will challenge the validity of the liens and the value of
the senior note will increase once the bank lenders are shown as not being entitled to a
priority recovery. In other words, the bank loan and senior notes are pari passu or on an
equal footing. A challenge similar to this occurred in the 2012 restructuring of Hawker
Beachcraft.1
The special situation investor in credit-related contexts often needs to resort to litiga-
tion, either in or outside of bankruptcy court, because the benefit that has been identified
is hotly disputed by the other party. In this example, the bank lenders would certainly
assert the validity of their liens on various theories. Several recent, fairly high-profile ex-
amples of litigation-driven special situation investments have occurred. For example, in
the Lehman Brothers bankruptcy, John Paulson and other investors purchased certain
bonds within the Lehman Brothers capital structure and argued that a relatively ob-
scure legal doctrine called substantive consolidation should apply. In simple terms, the
thrust of this claim was that various subsidiaries within Lehman should be consolidated
and all creditor claims paid from a common pool (Ryan and Freed 2011). Under this
approach, Paulsons bonds, which related to weaker subsidiaries, would benefit from
sharing the assets of stronger subsidiaries. Naturally, the creditors of the stronger sub-
sidiaries argued separate treatment was more appropriate. Paulsons group ultimately
prevailed and their bonds, which they reportedly purchased for as low as 7.5 percent of
face value received 24 percent of face value under the final plan of reorganization (Wirtz
and Spector 2011).

TURNAROUND INVESTING
Turnaround investing is another label that has been applied to a broad class of investment
strategies. Some use it to describe the simplistic strategy of identifying a cheap stock of a

1
Declaration of Robert S. Miller, In Re Hawker Beachcraft, No.1211873, Docket No. 20 (Bankr.
D. Del. 2012).
110 m a j o r t y p e s o f p r i vat e e q u i t y

struggling company and purchasing it in anticipation that operations will turn around
and its price will rise. As used in the distressed context, the process is more analogous
to classic PE investing in that the turnaround investor identifies a distressed target and
offers to make a direct investment in new equity capital to shore up finances and give the
company a second chance.
Of course, this investment offer typically incorporates a very low valuation for the
company so the equity infusion substantially dilutes existing equity holders and leaves
the turnaround investor with substantial influence if not control. Funds that do these
types of investments tend to view strategic and operational improvement as their core
competence and rely on this rather than financial engineering to make the investment
successful. The typical target tends to be a small or middle-market private business in
which operations can be easily revamped by the funds professionals.
TargetCo would be an unlikely candidate for a turnaround investor because of its
high leverage. A new equity investment in TargetCo could easily be lost if operations
did not improve sufficiently to make the debt serviceable. Also, the turnaround investor
would view most of the value-added of improved operations as benefiting the credi-
tors rather than the funds equity investment.
While most television and movie depictions of the investment process are unreal-
istic, The Profit, a reality television series that chronicles turnaround investments by
Marcus Lemonis, generally depicts how institutional turnaround investors function. In
the show, Lemonis identifies real businesses (generally fairly small companies such as
pet grooming stores or wine bars) that are struggling, often because of the owners mis-
management. After spending time analyzing the business, Lemonis makes the owner
an offer to purchase a substantial equity stake so long as the capital is invested in the
business as opposed to being taken by the owner. Lemonis gets to take charge of the
operating turnaround.
In the institutional turnaround investing context, the companies and correspond-
ing investments are larger and the formalities surrounding the investment much more
legally rigorous, but the general process is similar. Since the companies are larger with
more complex operating problems, up-front due diligence is done to thoroughly under-
stand the relevant operational, marketing, and strategic challenges. After conducting its
due diligence a firm will evaluate how much time and capital is required to determine if
it will pursue the investment. This process differs substantially from the previously dis-
cussed strategies because management fully cooperates with the turnaround investors
in order to obtain capital. A negotiation will occur to determine the businesss value
and what percentage of equity the turnaround investor will receive. The investment
agreement also includes provisions designed to allow the fund to influence an exit proc-
ess so that it can monetize its investment and return capital to its LPs.

Current State of the Distressed-Asset Private


Equity Market
Distressed-asset PE funds have burgeoned since the tech bubble burst and institutional
investors began to recognize it as an investable, distinguishable, and counter-cyclical
asset class. As Figure 7.3 illustrates, during 2007 and 2008 (the lead up to the 2009
Di s t re s s e d De bt I n v e s t m e n t s 111

50
45
40
35
30
25
20
15
10
5
0
2004 2005 2006 2007 2008 2009 2010 2011
Distressed Funds Raised ($B)

Figure 7.3 Distressed Private Equity Funds Raised in North AmericaThis


figure shows the aggregate amount ($ billions) in capital raised by distressed-asset PE
funds from 2004 to 2011. Source: Preqin (2011).

financial collapse), distressed funds raised $86 billion. This amount was more than what
was collectively raised in the prior decade (Preqin 2012).
This fundraising success reflected a growing conviction on the part of institutional
investors, based in part on the well-crafted pitches of distressed-asset PE fund manag-
ers, that distressed investment opportunities would be plentiful. The perception of an
attractive potential investment opportunity set was, at its core, grounded on the argu-
ment that credit markets go through credit underwriting cycles. The basic thesis is that
low default rates coupled with an expansion of capital allocated to credit markets, leads
to declining underwriting standards. This results in an increase in overall credit risk.
When such cycles end, a sharp uptick in defaults usually occurs and leads to a widen-
ing in credit risk spreads and a decline in credit availability. This perfect storm tends to
create fertile distressed investment opportunities. OKeefe (2010) provides a detailed
analysis of this phenomenon.
Figure 7.4 shows the aggregate merger and acquisition (M&A) transaction value
($ billions) involving North American based firms in each year from 1995 to 2009.
Over this period, two cycles are apparent. First, the M&A transaction value increased
from 1995 to 2000 and then reached low in 2003. Next, values started to rise in 2004,
peaked in 2007, and then started to decline.
In 2000 and 2007, all the preconditions for a buoyant distressed market were in place.
As Figure 7.5 illustrates, a large increase in LBO activity occurred that was fi nanced by
expanding issuance of B-rated or lower debt securities.
An increase in overall credit risk in the market accompanied this expansion of credit
market activity. This relationship is illustrated by the growth in issuance of bonds and
loans with lower credit ratings at issuance as Figure 7.5 shows, and the higher debt
multiples used in LBO transactions illustrated in Figure 7.6. In simple terms, buyout
managerswere paying higher prices for companies and financing more of their pur-
chases with risky debt securities.
Such trends cannot last indefinitely. Figure 7.7 documents the sharp increase in de-
fault rates starting in 2000 and 2009, respectively. Note that the more extended default
112 m a j o r t y p e s o f p r i vat e e q u i t y

3000

2500

2000

1500

1000

500

0
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
M&A Transaction Value ($B)

Figure 7.4 Merger and Acquisition Transaction Value, 1995 to 2009 This figure
shows the aggregate M&A transaction value ($ billions) involving North American based
firms in each year from 1995 to 2009 based on data from Bloomberg.

cycle that occurred between 2000 and 2002 reflected a prolonged recessionary environ-
ment that peaked with the bursting of the tech bubble in 2001. In contrast, the sharp
uptick in defaults in 2009 can be attributed to the shock to the capital markets caused by
the sudden collapse of Lehman Brothers.
These high-default periods created an excellent distressed-asset investment en-
vironment that allowed funds to deploy significant capital and realize attractive returns
as Figure 7.2 illustrates. However, the expansion in capital to distressed-asset funds
between 2007 and 2008, as Figure 7.3 shows, was so large that even the robust invest-
ment environment that followed the 2009 financial collapse could not absorb all this
supply. As of 2013 when default rates had fallen to below average levels, $34 billion in

60

50

40

30

20

10

0
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
B or Lower rated issues ($B)

Figure 7.5 Low Rated High Yield Issuance, 1995 to 2009 This figure shows the
aggregate issuance, excluding refinancings, of bonds rated B or lower in each year between
1995 and 2009. Source: J. P. Morgan (2013).
Di s t re s s e d De bt I n v e s t m e n t s 113

7.0
6.0
5.0
4.0
3.0
2.0
1.0
0.0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Senior Debt Sub Debt

Figure 7.6 Acquisition Debt Multiples This figure shows the average debt multiple
used in LBO transactions between 2000 and 2010. The bars show the composition of the
debt between senior and subordinated (sub) debt. Source: Bain (2012).

14.0%

12.0%

10.0%

8.0%

6.0%

4.0%

2.0%

0.0%
1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
HY Bond Default % Lew Loan Default %

Figure 7.7 Leveraged Finance Default Rates This figure shows the percentage
(as a percentage of the aggregate principle amount outstanding) of high yield (HY)
bonds and leveraged (Lev) loans that defaulted each year between 1998 and 2012.
Source: J. P. Morgan (2014).

committed capital, or dry powder, still remained in distressed-asset PE funds that were
searching for investments in a shrinking pool of opportunitiesat least in the United
States (Preqin 2014).

G E O G R A P H I C E X PA N S I O N O F D I S T R E S S E D I N V E S T I N G
As the high return potential for investing in distressed-assets began attracting increased
competition in the United States, investors began to look to other markets for distressed
investment opportunities. Besides nearby Canada, managers quickly identified Europe
as another large potential market. Various countries in Europe have well-developed
114 m a j o r t y p e s o f p r i vat e e q u i t y

200
180
160
140
120
100
80
60
40
20
0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
Total Value of Buyout (B)

Figure 7.8 European Leveraged Buyout Volume This figure shows the aggregate
transaction value ( billions) of LBOs completed in Europe for each year between 1997
and 2008. Source: Deva (2010).

capital markets. As Deva (2010) chronicles, the European market for LBOs developed
robustly in the 2000s as Figure 7.8 shows.
As discussed previously, successful distressed investing requires the proper legal en-
vironment. In this regard, some considered Europe a minefield for distressed investors
until the early 2010s (Khosla 2013). One major reason is that Europe had no recog-
nized common forum in which to complete restructurings, particularly if the target had
operations in several sovereign jurisdictions. In the latter case, an investor faced the risk
of having to manage multiple legal actions in different forums under different legal re-
gimes (DePonte 2010). This situation alone was a severe deterrent. Even if a target op-
erated primarily in one country, laws relating to creditor rights could wildly vary from
creditor friendly to debtor protective. Assuming creditor rights were strongly respected,
this did not necessarily lead to a constructive workout process.
For example, in Germany if a companys board of directors concludes that the com-
pany was insolvent, the board has the legal obligation to immediately file for bankruptcy
or the board members themselves can be personally liable. This framework tends to
constrain the negotiations that a company might have with creditors before an insol-
vency process. Once the filing occurs, secured creditors are often allowed to immedi-
ately enforce their lien rights, which typically leads to a quick liquidation. While this
result may be acceptable for secured lenders, it typically provides little recovery to un-
secured financial creditors because the going-concern value of the company is essen-
tially lost (Rkollp 2013).
Finally, many European cultures have a much different ethos on the role of govern-
ment in protecting local companies and local employment than in the United States.
Thus, the distressed-asset investor has to weigh the risk that even if it manages to gain
control of a going-concern business, it might be unable to restructure a targets opera-
tions in a way that would maximize its potential value (Khosla 2013).
Structural issues also made investment more difficult. In Europe, banks play a more
central role in capital markets. They provide a much greater percentage of financing
Di s t re s s e d De bt I n v e s t m e n t s 115

relative to the role of bond capital markets. Until recently, European banks retained a
much larger portion of the loans they underwrote on their balance sheet, limiting the
secondary market for these securities. In contrast, banks in the United States typically
syndicate the loans they underwrite providing a more liquid market in which distressed
investors can accumulate positions. Compounding the problem of purchasing European
loan interests, European banks often delay making aggressive loss reserves for under or
non-performing loans. Thus, a sale at a material discount would effectively result in a
substantial charge to capital, which crimps the liquidity of European distressed debt
(Ernst & Young 2013). Depending on the country, regulatory constraints also limit
the type of entity that can purchase a distressed claim thus often barring a traditionally
structured distressed-asset PE fund (Rkollp 2013).
Since the mid-2000s, several major developments have improved the climate for dis-
tressed investing in Europe. One important legal development involved the European
Commissions adoption of a rule to determine the appropriate forum for a restructuring
proceeding depending on a companys center of main interests (COMI) (Weil 2013).
The legislation provided market participants with guidance on where restructurings
should take place. The effect, however, was broader. With some planning and organiza-
tional changes, companies could shift the COMI from one jurisdiction to another. This
effectively allowed forum shopping, which means in some cases the debtor could move
to the most favorable jurisdiction for a restructuring. The jurisdiction of choice quickly
became England, which had several reorganization friendly legal regimes (Weil 2013;
DePonte 2010). On the structural side, the syndication of European loans has broad-
ened with the growth of a local collateralized loan obligation (CLO) market and banks
have aggressively improved their capital positions such that loan sales are more feasible
at appropriate discounts.
Taken together, along with numerous other evolutions, the result has been a consider-
able pick-up in corporate distressed-asset investing in Europe. As Figure 7.9 shows, dis-
tressed funds raised substantial capital targeting Europe. While European managers lead

0
2004 2005 2006 2007 2008 2009 2010
Amount Raised ($B)

Figure 7.9 Fundraising for European Distressed-Asset Private EquityThis


figure shows the aggregate amount of capital raised by distressed-asset PE for each year
between 2004 and 2010. Source: Preqin (2011).
116 m a j o r t y p e s o f p r i vat e e q u i t y

several of these funds, the majority are controlled by U.S.-based distressed-asset funds.
These U.S. managers have aggressively moved into the market in an effort to d iversify
away the dependence on U.S. credit cycles.

B U Y O U T P R I VAT E E Q U I T Y F I R M S G O O N T H E O F F E N S E
A description of PE involvement in distressed investing would be incomplete without
discussing the evolution of the LBO sponsor as a potentially important player in the dis-
tressed marketplace. In the TargetCo example, the impression may be that distressed PE
funds hover over the mortally wounded company like vultures deciding how to dissect
a helpless cadaver. The vulture analogy might be apt, but increasingly, distressed targets
are hardly helpless.
Distressed investing is not a win-win game for all participants but involves winners
and losers. On one side, a distressed-asset firm is attempting to own the target, and, on
the other side, a buyout PE firm is trying to retain control and not lose its original in-
vestment in the target. Particularly in the 2009 distressed cycle, many buyout PE firms
were aggressive in countering different approaches used by distressed-asset investors to
gain control of targets. Sometimes these techniques involved redesigning the terms of
the financing instrumentssuch as the development of payment in kind (PIK) toggle
notes (Brittenham and Selinger 2014) or equity cure provisions (Mincemoyer 2011)
to limit or delay the distressed investors opportunity to threaten a default and thus add
risk and/or lower prospective returns.
Other strategies involved financial engineering, such as the coercive exchange offer. In
an exchange offer, the company offers to exchange one bond for another bond or a pack-
age of securities that have an expected value greater than the bond being exchanged.
This attempts to entice participants to willingly choose to exchange. In a coercive ex-
change offer, the deal is structured such that the original holder is potentially worse-off
if it does not participateessentially coercing participation (Moyer 2005). The typical
mechanism involves the distressed company creating a new second lien bond and offer-
ing to exchange these secured bonds, often at substantial reductions in face value com-
pared to the existing unsecured notes, to existing holders. For example, in 2009 Harrahs
Entertainment faced a series of near-term bond maturities aggregating over $1 billion.
It offered holders of its senior notes maturing in 2013 the opportunity to exchange into
new second lien notes but at a dramatic discount of only 50 percent of face value (Cae-
sars 2009). Why would the holder accept such an offer? The answer lies in the fact
that Harrahs bankruptcy was a significant risk and the second lien bond was senior in
recovery rights to the existing unsecured senior notes. The implicit threat was that if the
holder held out and did not exchange and Harrahs subsequently filed for bankruptcy,
the unsecured notes recovery might be de minimus after the second lien notes received
a full recovery. The coercive exchange offer was a particularly powerful tool for spon-
sors because it allowed them to extract considerable reductions in debt principal claims,
which inherently increased the value of the sponsors equity position.
Lastly, the LBO sponsor, or in some cases its affiliate, would compete with the dis-
tressed investor and start purchasing the distressed companys bonds at steep discounts
for its own account. Multiple rationales underlie these purchases. Most directly, if the
sponsor subsequently contributed the notes to the company and they were cancelled,
Di s t re s s e d De bt I n v e s t m e n t s 117

the value of the sponsors equity would increase. Such a cancellation also effectively low-
ered leverage, which might avoid the breach of a covenant that would give distressed-
asset investors negotiating power. Second, the sponsors purchases inherently prevent
a distressed investor from acquiring the same bonds, which increases the difficulty for
the distressed-asset investor to amass a potentially dominant or controlling position.
Finally, in the worst-case scenario from the sponsors perspective, if the target has to
restructure and the sponsors equity position is wiped out, its distressed bond position
may give it means to retain an equity share and attempt to recoup some of its earlier loss.
A primary benefit attributed to LBOs is that they reduce agency costs. When early
distressed-asset investors targeted publicly held fallen angels, they preyed upon an or-
ganization where the equity class often could not organize to protect its interests. Now
the dynamics of the game have fundamentally changed with sophisticated and deep-
pocketed buyout PE firms challenging distressed-asset PE firms in their efforts to create
and control a restructuring.

The Outlook for Distressed Private Equity Investing


As markets mature, pricing tends to become increasingly efficient. This increases the
difficulty of finding positive risk-adjusted returns. Distressed debt investing by PEs or
others is arguably entering this stage of evolution, particularly in the United States. In
the 1990 cycle, distressed investing was essentially a new discovery. In the 2000 cycle,
people remembered the 1990 era but the market remained relatively inefficient due
to few experienced investors and relatively little dedicated capital. Since the 2009 re-
cession, distressed investing has matured. Currently, much more dry powder exists in
committed capital than opportunities. Scores of dedicated funds with hundreds of ex-
perienced analysts hold this capital. These analysts pore over the outlook for any bond
or loan that trades at even a modest discount. This level of attention and scrutiny leads
to more efficient pricing and makes earning surplus returns more challenging, a condi-
tion which could persist even if an increase in the default rate improves the supply of
distressed opportunities.
Beyond the near-term supply/demand technicals, the intermediate-term environ-
ment also seems challenging. During the prolonged period of historically low interest
rates that has persisted since the Federal Reserves quantitative easing policy begin-
ning in 2008, more corporations have been able to refinance their debt capital struc-
tures at comparatively low interest rates. For example, according to Bloomberg, the
average coupon on a triple C rated bonds issued in the first quarter of 2014 was ap-
proximately 8.50 percent, almost 50 percent below the same figure in the first quarter
of 2004. Further, the aggregate amount of below investment grade bond debt matur-
ing before 2017 represents less than 75 percent of all such debt outstanding (Guggen-
heim 2014). In other words, interest expense burdens on internal cash flow are low, as
is exposure to refinancing risk. The implications of these statistics are that high yield
companies have never had less burdensome debt capital structures in their history.
This suggests that default rates could remain subdued for some time, although funda-
mental market developments, such as the recent decline in oil prices, could pressure
certain sectors.
118 m a j o r t y p e s o f p r i vat e e q u i t y

While the U.S. market is now fairly efficient, many other markets, both in Europe
and emerging economies, are not as efficient. Some evolution of the legal structures
is necessary in these countries to facilitate distressed-style investing. As these evolve,
decades of opportunities may exist for such investing.

Summary and Conclusions


This chapter reviews the evolution of distressed-asset PE firms following the passage
of the Bankruptcy Act of 1978 and the development of the high yield bond market.
Popular distressed-asset investment strategies are analyzed and the expansion of
the investing scope from primarily the United States to Europe is explored. Finally,
the conflicts between buyout PE sponsors and distressed PEs are outlined. Clearly,
distressed PE has had an important role in making distressed investments an identifi-
able, counter-cyclical, alternative asset class that has attracted growing interest from
institutional investors. Less clear is whether the class can continue to post strong risk-
adjusted returns in an increasingly efficient U.S. market and a substantially different
European market.

Discussion Questions
1. Discuss how the passage of the Bankruptcy Act of 1978 set the stage for the growth
of distressed debt investing.
2. Compare and contrast the main strategies used by distressed debt investors.
3. Explain whether distressed investors can add alpha if their returns tend to depend
on favorable market environments.
4. Explain why distressed debt investors are sometimes characterized as vulture
investors.

Acknowledgments
The authors are grateful for the capable research assistance of Dan Mayer in preparing
this chapter.

References
Altman, Edward I., and Edith Hotchkiss. 2006. Corporate Financial Distress and Bankruptcy, 3d Edi-
tion. Hoboken, NJ: John Wiley & Sons.
Bain. 2012. Global Private Equity Report 2012. Bain & Company. Available at http://www.bain.
com/publications/articles/global-private-equity-report-2012.aspx.
Baribeau, Mark B. 1989. Leverage Risk in the Nonfinancial Corporate Sector. Business Economics
24:3, 3439.
Bernanke, Ben S., John Y. Campbell, and Toni M. Whited. 1990. U.S. Corporate Leverage: Devel-
opments in 1987 and 1988. Brookings Papers on Economic Activity 1990:1, 255278.
Di s t re s s e d De bt I n v e s t m e n t s 119

Brittenham, David A., and Scott B. Selinger. 2014. Everything Old Is New Again: PIK Notes. De-
bevoise & Plimpton, The Private Equity Report 14:1. Available at http://www.debevoise.com/
insights/publications/2014/03/everything-old-is-new-again-pik-notes.
Buffett, Warren E. 2008. Buy American. I Am. New York Times. October 17, A33. Available at
http://www.nytimes.com/2008/10/17/opinion/17buffett.html?_r=0.
Caesars. 2009. Harrahs Entertainment Announces Results of Exchange Offers. Available at
http://investor.caesars.com/releasedetail.cfm?ReleaseID=607322.
DePonte, Kelly. 2010. The Definitive Guide to Distressed Debt and Turnaround Investing, 2d Edition.
London: PEI Media Ltd.
Deva, Saloni. 2010. Determinants of Leveraged Buyouts in Europe: LBO Financing and Country
Legislature. Jonkoping International Business School. Available at http://www.diva-portal.
org/smash/get/diva2:353491/FULLTEXT01.pdf.
Ernst & Young. 2013. Flocking to Europe: Ernst & Young 2013 Non-Performing Loan Report.
Available at http://www.ey.com/Publication/vwLUAssets/Flocking_to_Europe/$FILE/
Flocking_to_Europe.pdf.
Guggenheim. 2014. High Yield and Bank Loan OutlookJanuary 2014. Available at http://
guggenheimpartners.com/perspectives/sectorreport/high-yield-sector-reports/high-yield-
and-bank-loan-outlook-january-2014.
Harner, Michelle M. 2011. Activist Distressed Debtholders: The New Barbarians at the Gate.
Washington Law Review 89:1, 155206.
J. P. Morgan. 2013. 2013 High-Yield Annual Review. New York: J. P. Morgan.
J. P. Morgan. 2014. High-Yield Default Monitor. New York: J. P. Morgan.
Khosla, Victor. 2013. Busting the Myths of European Distressed Debt. Available at http://www.
institutionalinvestor.com/blogarticle/3247436/Blog/Busting-the-Myths-of-European-
Distressed-Debt.html#.U8gpXai0ZcA.
Miller, Harvey R., and Shai Y. Waisman. 2005. Is Chapter 11 Bankrupt? Boston College Law Review
47:129, 129181.
Mincemoyer, R. Jake. 2011. Equity Cures: An Ideal Standard? Available at http://www.whitecase.
com/files/Publication/7b4d81e5-1902-4531-4539d4b-b28a779bd9ce/Presentation/
PublicationAttachment/0fbb427a-15e5-4607-9cd3-deb284e95872/alert_Equity_Cures_
An_Ideal_Standard.pdf.
Miner, Don J. 1979. Business Reorganization under the Bankruptcy Reform Act of 1978: An Anal-
ysis of Chapter 11. Brigham Young University Law Review 1979:4, 961986.
Moyer, Stephen G. 2005. Distressed Debt Analysis: Strategies for Speculative Investors. Boca Raton, FL:
J. Ross Press.
Moyer, Stephen G., David Martin, and John Martin. 2012. A Primer on Distressed Investing:
Buying Companies by Acquiring Their Debt. Journal of Applied Corporate Finance 24:4,
5989.
Oaktree Capital Group, LLC. 2011. Form S-1. Available at http://www.sec.gov/Archives/edgar/
data/1403528/000119312511167852/ds1.htm.
OKeefe, John. 2010. The Effects of Underwriting Practices on Loan Losses: Evidence from the
FDIC Survey of Bank Lending Practices. In Suk-Joong Kim and Michael D. Mckenzie, eds.,
International Banking in the New Era: Post-Crisis Challenges and Opportunities, International Fi-
nance Review, Volume 11, 273314. Bingley, U.K.: Emerald Group Publishing Limited.
Preqin. 2011. Preqin Special Report: Distressed Private Equity. Available at https://www.preqin.
com/docs/reports/Preqin_Special_Report_Distressed_Private_Equity.pdf.
Preqin. 2012. Distressed PE Is the Best Performing Private Equity Strategy. Available at https://
www.preqin.com/docs/press/PrEQIn_Index.pdf.
Preqin. 2014. The Q1 2014 Preqin Quarterly Update: Private Equity. Available at https://www.
preqin.com/docs/quarterly/pe/Preqin_Quarterly_Private_Equity_Update_Q1_2014.pdf.
Rkollp. 2013. Debt Trading in Europe: A Handbook for Trading Debt & Claims in Select Dis-
tressed European Markets. Richards Kibbe & Orbe. Available at http://www.rkollp.com/
assets/attachments/Debt%20Trading%20in%20Distressed%20Europe%202013.pdf.
120 m a j o r t y p e s o f p r i vat e e q u i t y

Ryan, Amy Tucker, and David Freed. 2011. Tearing Down Invisible Walls Substantive Consolida-
tion in Chapter 11. Available at http://www.usfn.org/AM/Template.cfm?Section=Home&
SECTION=Article_Library&TEMPLATE=/CM/HTMLDisplay.cfm&CONTENTID=
19294.
Stus, Richard. 2011. Preqin Special Report: Distressed Private Equity. Available at https://www.
preqin.com/docs/reports/Preqin_Special_Report_Distressed_Private_Equity.pdf.
Wachtel. 2013. Distressed Mergers and Acquisitions. Wachtel, Lipton, Rosen & Katz. Available at
http://www.wlrk.com/webdocs/wlrknew/AttorneyPubs/WLRK.22377.13.pdf.
Weil. 2013. Comparative Guide to Restructuring Procedures. Weil, Gotshal & Manges. Available at
http://eurorestructuring.weil.com/wp-content/uploads/2013/02/Comparative-Guide.pdf.
Wirtz, Matt, and Mike Spector. 2011. Fight for Lehmans Remains Heats Up. Wall Street Journal,
April 26. Available at http://online.wsj.com/news/articles/SB100014240527487038567045
76285483621495232.
Part Three

HOW PRIVATE EQUITY WORKS


8
Valuing Private Equity
H A R L A N P L AT T
Professor of Finance, Northeastern University

EMERY A. TRAHAN
Senior Associate Dean of Faculty and Research and Professor of Finance,
Northeastern University

Introduction
Private equity (PE) is similar to other investments in that the purchaser should realisti-
cally estimate the assets value, persuade the seller to agree on a fair price that does not
overvalue the asset, and develop a plan to exit the investment. This chapter focuses on
issues of PE valuations and provides a clear blueprint for generating a value estimate.
The major components required to value an asset are: (1) a sense of the assets cash-flow
trajectory, (2) an understanding of the riskiness of these cash flows, and (3) knowledge
of the acquirers cost of capital. Properly combining these inputs enables an analyst to
estimate an assets value. The word estimate is critical because actual future values of
an asset are unknown. The value estimated before the purchase may be incorrect for
several reasons including inaccurate cash-flow projections, a failure to understand the
assets riskiness, macroeconomic factors, and an incorrect identification of the compa-
nys cost of capital.
Valuing cash flows is widely employed, but it is not the only process that financial
analysts use to value PE targets. Some other valuation methods include using multiplies
of EBITDA, which refers to earnings before interest and taxes (EBIT) plus depreciation
and amortization, or some measure of value per customer such as dollars per active trad-
ing account if buying a brokerage business. Alternative methods are valuable in situations
in which generating reasonably accurate annual cash-flow projections is more difficult
due to limited historical data, new technologies or product markets, or dealing with un-
certainties surrounding the advantages from rollup or consolidation strategies. Standard
valuation guidelines are based on previous deals and are accepted as the common norm.
For example, some PE firms set limits on the maximum amount they would pay such as
six times EBITDA. Working with such multiples makes exit strategies easier to define. In
this example, the plan might be to exit at eight times the EBITDA multiple.
The purpose of this chapter is to present a detailed discussion of valuation-related
issues and provide a comprehensive example of valuing a PE target company. The rest
123
124 h o w p r i vat e e q u i t y w o r k s

of the chapter is organized in five sections. The next section briefly describes measures
of cash flow. Then the discussion turns to using cash flow to estimate value, including a
discussion on discounted cash flow (DCF) valuation, estimating the appropriate cost of
capital, and using the adjusted present value (APV) method. Third, multiples-based val-
uation is then discussed as an alternative to the DCF method, followed by a brief discus-
sion of how PE firms may increase value. The chapter concludes with a comprehensive
valuation example to illustrate the application of the methods discussed. Other sources
to consider are Arzac (2005), Scharfman (2012), and Stowell (2012).

Measures of Cash Flow


Companies generate cash from their operations, investments, and financing activities.
This structural trifurcation of cash flows is traditional. It is reflected in the statement of
cash flows, which is how accountants summarize incoming and outgoing cash flows.
While some financial analysts use net cash flow, the ending value from the statement
of cash flows, most use more specific measures of cash flow. One reason for not using
net cash flow is that it complicates calculating equity cash flow and enterprise (whole-
firm) cash flow. The difference between equity and enterprise cash flows is that the
former measure looks only at funds available to equity holders while the later measure
determines cash flows available to both debt and equity. Knowing the difference be-
tween equity and enterprise cash flow is especially important in the PE context because
PE firms often purchase an entire enterprise.
The difficulty in the valuation process begins with net income. As its name implies,
net income is determined after netting out interest payments to debt from revenues.
This adjustment takes place because interest payments are not a part of equity cash flow.
However, interest payments need to be reflected in enterprise cash flow. Therefore, in-
terest payments net of tax benefits (interest payments are a deductible expense) must be
returned to cash flows when determining enterprise cash flows. A second issue that pre-
vents some analysts from relying on the statement of cash flows for their cash-flow infor-
mation is capital expenditures. Capital expenditures need to be deducted in the process
of calculating free cash flow and are not part of the operations portion of the statement.
Instead, they fall under the investments category of the statement of cash flows.
The most familiar but perhaps the least informative cash-flow measure is called
simple cash flow (SCF) shown in Equation 8.1. It combines net income plus depreci-
ation and amortization and is often referred to as the sum of net income and noncash
charges. SCF is an equity cash-flow measure. It is a vast improvement over simply using
net income because is reduced by expenses that do not lower the firms cash position.
SCF is probably the most often used cash-flow measure because of its simplicity:

Simple Cash Flow = Net Income + Depreciation and Amortization. (8.1)

The corresponding whole-firm simple cash-flow concept is net operating profit after-tax
(NOPAT). Like SCF, NOPAT begins with net income, but then adds the value of inter-
est payments net of their tax shields back into net income. For example, with a 40 per-
cent tax rate a $1.00 interest payment deduction against income reduces taxes by $0.40.
Val u in g P riv at e E qu it y 125

Hence, the amount added back into net income to create NOPAT in this example is not
the full interest payment of $1.00 since $0.40 had been reflected as lower taxes. Adding
the value of interest payments net of their tax shields to net income creates unlevered
net income (UNI) as described in Equation 8.2:

Unlevered Net Income = Net Income + (1 Tax Rate)( Net Interest Expense ).
(8.2)

UNI represents the net income the firm would report if it had no interest-bearing obli-
gations. That is, UNI determines the combined return to equity and debt holders and is
a more complete concept than net income. In the next step, the change in deferred taxes
is added to UNI; the resulting sum equals NOPAT as seen in Equation 8.3. A firms bal-
ance sheet shows deferred taxes as a long-term liability. Deferred taxes arise because the
firms income statement provided to shareholders reports some taxes that the firm does
not actually pay in the current period due to the difference between the depreciation
reported on financial statements and the amount shown on tax statements. NOPAT
equals UNI plus the change in deferred taxes:

NOPAT = UNI + Deferred Taxes, (8.3)

where shows a change in the value between two time periods.


Moving beyond these simple cash-flow formulas introduces operating activity cash
flows (OACF), as found on the statement of cash flows. OACF is the complex version
of SCF and free cash flow (FCF), which is the complex version of NOPAT. Each layer
of complexity includes additional sources of cash flow. To find OACF, the change in net
working capital (NWC) is netted out of SCF as Equation 8.4 shows:

OACF = SCFNWC. (8.4)

FCF starts with NOPAT but then adds depreciation and subtracts capital expenditures
(CAPEX) and change in NWC, as Equation 8.5 shows:

Free Cash Flow = NOPAT + Depreciation CAPEX NWC. (8.5)

In the PE realm, FCF is the crucial cash-flow measure that financial analysts usually
employ to determine a firms value. FCF is a whole-firm measure. Unlike OACF, the
equity-based complex cash-flow measure, FCF includes the impact on cash flow of net
interest payments, CAPEX, and changes in deferred taxes. Each of these is an impor-
tant additional cash-flow element that explains why PE firms tend not to rely on OACF,
except in unusual situations. If analysts need to select between SCF and NOPAT, the
latter is the more relevant to PE firms. Table 8.1 presents the relationship between the
various cash-flow measures.
PE firms generally want a whole-firm cash-flow measure because they are buying
an entire enterprise. Financial analysts can decide on whether to choose the simple or
the complex whole-firm cash-flow measure by examining the components for each in
126 h o w p r i vat e e q u i t y w o r k s

Table 8.1 Components of the Four Major Cash Flow Definitions


Components Simple-Equity Complex-Equity Simple-Whole Complex-Whole
SCF OACF NOPAT FCF
Net income
Depreciation
Net interest less tax
benefit
Capital
expenditures
Deferred taxes
Accounts
receivable
Inventories
Prepaid expenses
Accounts payable

Denotes included items.

Table 8.1. Analysts typically prefer the complex calculation, but in certain cases such as
technology start-ups a simple method may suffice because these firms may be primarily
financed by equity and have limited working capital. However, analysts may choose to
calculate only a simple cash flow for these newly formed companies due to insufficient
financial information required to determine the complex cash flows.

Using Cash Flows to Estimate Value


Unlike the stock market in which prices or valuations are often based on biased expecta-
tions, PE firms estimate a firms value based on future cash flows, applying a disciplined
and rational assessment of the firms prospects both before and after acquisition. PE firms
typically incorporate estimates of how they might be able to change future cash flows
into their calculations by instituting various revenue-increasing and cost-reducing strate-
gies. Such firms are similar to turnaround companies that try to save distressed compa-
nies except that PE firms intend to reap the rewards of the turnaround effort themselves.
The first step in the valuation effort is to develop detailed analytical spreadsheets
that project the firms income statement and balance sheet. From these statements, ana-
lysts can estimate future cash flows. This phase is when the first differentiation among
PE firms arises as each has varying levels of skill in financial modeling and each has
different assumptions about the firms growth rate, cost structure, and efficiency in
controlling working capital needs. Another difference involves the perceived costs of
under- and overestimating future cash flows. The greater amount of invested funds that
comes from the PE firms partners, the higher is the perceived cost of overestimating
future cash flows because partners will directly absorb the loss. For a PE firm that makes
Val u in g P riv at e E qu it y 127

transactions using solely other peoples money, the cost of underestimating future cash
flows increases because the deal would be underpriced.
The next step in preparing a valuation estimate for a target company is determining
the level of leverage the investment will carry. Greater leverage results in higher interest
payments that reduce net income and cash flow available to the equity holders. Leverage
also increases the riskiness of the investment because excessive debt is often a contribut-
ing factor in bankruptcies. Yet, greater leverage increases the return on equity (ROE) at-
tainable by the PE firm. The trade-offs involved in setting a desired leverage ratio are not
subtle and stand at the nexus of greed and fear. Greed pushes the PE firm to use higher
leverage while fear constrains this impulse and keeps leverage lower.
The final task is to decide how much to pay for the investment. Although price is
often based on the estimate of the investments present value, analysts should consider
other factors such as the size of the PE fund, number of investments the fund desires to
hold, and contractual limits on the proportion of the fund that may be invested in any
one asset.
When Georges Doriat founded American Research and Development Corporation
in 1946, one of the first American PE firms, little competition existed so PE firms could
make safe, low bids on assets. This situation has changed. The PE Growth Capital Coun-
cil (2013) reports the existence of nearly 2,800 PE firms in the United States. As a result,
rising asset values have lowered the likely return for prospective buyers. For that reason,
PE firms need to embrace a turnaround or renewal strategy to get more return out of the
asset than it would yield without a buyout.
These three steps form the core of the PE valuation process: (1) analytical modeling,
(2) leverage optimization, and (3) pricing. To some degree, the process is more of an
art than a science, but the critical elements in a successful PE investment include know-
ledge about the industry in which the target company operates and previous experi-
ence with integrating the three steps. By combining the first two steps, namely analytical
modeling and leverage optimization, the PE firm builds an estimate of the present value
of an investment for the target firm. Comparing this figure with the price the PE firm is
willing to pay leads to an estimate of yield on the investment. Alternatively, the PE firm
may develop its price estimate by applying a return hurdle rate against its estimate of the
investments present value. That is, if the PE firm believes the investment has a $100 mil-
lion present value and wants to earn 25 percent on its investment, then the firm would
bid no higher than $80 million for the target company.

D I S C O U N T E D C A S H F L O W VA L UAT I O N
Analytical modeling is formalized by developing a comprehensive spreadsheet that
provides a detailed examination of the target firms future revenues, costs, assets, and
liabilities. Since many PE buyers may be in discussions with the target company and
have signed nondisclosure agreements, they usually begin by accessing the targets in-
ternal forecasts. PE buyers do not rely on the targets projections, but use them as a
starting point for their own analysis. The main advantage to starting with the targets
own spreadsheets is that they contain details the PE firm might otherwise miss such as
product level revenue and cost details, dates of anticipated costs and new revenues, and
unusual expenditures such as contractual payments or court judgments.
128 h o w p r i vat e e q u i t y w o r k s

The PE buyer is likely to disagree with the target companys forecast assumptions for
many reasons. Consequently, the PE firm creates its own proprietary set of forecasts. The
critical assumptions that underlie this proprietary forecast concern market share and
the associated growth rate in revenues, cost increases arising when long-term contracts
end or as a result of anticipated market forces, variable interest rates, and tax payments
based on differences in tax shields, utilization decisions, and depreciation charges.
The DCF method is defined by the present value calculation, the best-known finan-
cial formula. Financial analysts can use the formula to value a simple project, a target
company, or any identified revenue stream. PE firms are most likely to use FCF as their
cash-flow measure, but if they prefer a different cash-flow measure, they can substitute
it into the formula in place of FCF. As Equation 8.6 shows, the DCF formula discounts
future cash flows back to the present using a proper discount rate:
n
FCFt
Present value = t , (8.6)
t =1 (1+r )
where FCF = free cash flow estimated for a given year; t = time period; n = the terminal
or last year of detailed forecasts; and r = the discount rate. The present value formula
reduces annual cash-flow values to a single number, which is the present value of the
entire stream of cash flows.
No set rule exists on the number of years to include in the detailed forecasts prepared
by the PE firm. Detailed forecasts of five years are likely a minimum requirement. Some
analysts extend the detailed forecasts as far out as reasonably possible. What is essential
is that the forecasts span the time range of various cycles: products, capital equipment,
and contractual arrangements. For example, suppose the target firm has thousands of
point-of-sale terminals that it purchased this year and these expensive capital goods
need to be replaced every 10 years. In that case, the detailed forecast should go out at
least 10 years so the forecast incorporates the abnormally large capital expenditure on
new point-of-sale terminals. Of course, the target company and its cash flows may well
exist beyond the number of years described in the detailed forecasts. To incorporate
these post-detailed forecast cash flows into the PE firms estimate of the target firms
value, financial analysis often generate a terminal value calculation to estimate the addi-
tional value beyond the end of the detailed forecast.
When an investment provides cash flows in the years following the end of the de-
tailed cash-flow forecast period, these cash flows are captured in the present value for-
mula by creating a terminal or wrap-up value for the investment. This alters the basic
present value formula as shown in Equation 8.7:
n
Cash Flow t TV
Present value = + , (8.7)
t =1 (1+ r )t
(1+ r )n
where TV = terminal year value. TV is typically calculated by applying a future growth
rate estimate to the cash-flow forecast for the final year illustrated in Equation 8.8:
(Cash Flow n )(1+ g )
TV = , (8.8)
(r g )
where g = the assumed growth rate in cash flow for years beyond n.
Val u in g P riv at e E qu it y 129

The simplicity of Equation 8.8 conceals several important issues. One issue is the
year selected for the final detailed forecast year (i.e., year n). The final detailed year
should be typical of previous years and should be representative of what is expected
going forward. For example, an investment that generates an extra-large return in year
n of $500,000 in cash flows but only $100,000 of cash flows in every other year would
be misrepresented if year ns cash flows were treated as if they were typical. If the final
detailed year is misleading as an expected future value, replacing it in Equation 8.7 with
a more typical forward-going value would be advisable.
Another issue is that the TV should be determined from a steady state perspective. A
long-term steady state means that future capital expenditures and working capital needs
are exactly matched by future depreciation levels. If that is not the case, the excess cap-
ital expenditures (depreciation) should be subtracted (added) on a present value basis
from (t0) the TV. In other words, presenting the firm in the TV years should be self-
sustaining and complete. This is especially critical for companies that are likely to antic-
ipate a major investment cycle in the future.
The third issue is the assumed future growth, g. One possible value for g is the growth
rate in FCF in the final detailed forecast year. The advantage of this choice is that it reflects
detailed estimates of revenues and costs and accounts for macroeconomic factors, indus-
try competition, technological factors, and market share forecasts. As mentioned previ-
ously, this growth should not be an aberration and should reflect a possible future growth
rate. Another approach sets g equal to a reasonable fraction (e.g., 80 percent) of the firms
recent actual growth rate (e.g., five-year average) or a fraction of the industrys growth rate.
A final problem with the formula for calculating TV occurs when g exceeds the es-
timated discount rate. When this happens, the formula will generate a negative value.
Obviously, the present value is not negative for all future years that the investment is
owned. The forecast period must extend out far enough to reach a point where cash flow
is positive. Additionally, the forecast period must reach a point when the company is at
a steady state and point of maturity in its life cycle such that the long-term growth rate
is less than the discount rate.

C O S T O F C A P I TA L
Albert Einstein is said to have deemed compounding the most powerful force in the
universe (QuotationBest.com 2014). His point is relevant to the present value calcula-
tion as the denominator of the formula compounds the discount rate over time. A dollar
of cash flow earned sooner is more valuable than a dollar earned later. To that point, the
impact of a small change in r, the discount rate, on present value can be sizable. This is
why determining the appropriate discount rate to use in the formula is so important.
The weighted average cost of capital (WACC) is a common choice for the discount
rate to be applied in the formulas above. WACC is calculated by weighting the inde-
pendent costs of after-tax debt and equity by their respective shares in the firms capital
structure. A linear combination of the capital structure weights and the after-tax cost of
each component create the WACC as Equation 8.9 shows:

E D
WACC = ke + kd (1T ), (8.9)
V E
130 h o w p r i vat e e q u i t y w o r k s

where ke = cost of equity; kd = cost of debt measured by dollars of interest paid over
debt; E = market value of equity; D = market value of debt; V = E + D; and T = tax rate.
Although the equity cost of capital can be determined in several ways, the capital
asset pricing model (CAPM) is widely employed in both academic studies and practice.
The CAPM suggests a formula, as in Equation 8.10, for the equity cost of capital, ke:

ke = rF + (rM rF ), (8.10)

where ke = cost of equity; rF = risk-free rate or cost of riskless debt, often the 10-year
Treasury rate; rM = the required rate of return on a stock portfolio, called the market
portfolio; and = the firms beta, often determined with a 60-month regression of the
companys returns on the S&P 500 index representing the market return. If a target
company is not publicly traded, analysts may be able to use the betas of comparable
publicly traded companies to estimate an appropriate beta.
The market risk premium (rM rF) measures the expected higher return on the
market versus risk-free government bonds. Estimates of the market risk premium are
available from many sources such as Ibbotson Associates, which Morningstar acquired
in 2006. Estimates can range between 0 to 8 percent for public companies, with many
analysts using a premium of 5 to 6 percent. , the firms beta as shown in Equation 8.10,
captures a companys risk as related to movements in the general stock market as op-
posed to company-specific factors. A high shows that a companys stock moves up or
down in the same direction as the overall stock market but to a greater extent.

A D J U S T E D P R E S E N T VA L U E
A potential problem with using WACC as the discount rate is that it assumes the firms
capital structure remains unchanged in the future even as the firm eventually begins to
derive FCF, some of which might be used to reduce its indebtedness. The APV model
is an alternative to WACC, addressing the fact that the capital structure may change
over time such as in a highly leveraged deal in which debt will be paid down over time.
This method is an alternative to estimating different WACCs for each period. APV sepa-
rates the value of cash flows into two parts: the first part comes from the investment
itself in this case acquiring a target company and the second part comes from the po-
tentially value-raising effects of the projects financing. Investment cash flows, which
are produced unaided by financial leverage, are discounted to their present value using
the cost of assets, ka, which is determined the same as the CAPM except that beta is for
a debt-free firm. Alternatively, Equation 8.11 shows this relationship as the cost of the
unlevered firm, ku:
D
ku = rF + /1+ (rM rF ), (8.11)
E
D
where A = /1+ = the asset beta.
E
An asset beta reflects risk of the entire firm while an equity beta measures risk to
the firms equity. Differences between ku and WACC arise not only because they rely
on different betas but also because WACC includes the tax shield and therefore must
Val u in g P riv at e E qu it y 131

be recalculated each year while ku, which does not include the tax shield, needs to be
calculated only once.
APV adds to this amount the value derived from financing opportunities and the
value coming from the tax shield, which are discounted using a rate that reflects the
chance the value will be realized. For example, a company with volatile earnings might
experience periods when its net income is negative and financing returns would not be
earned. In that case, an analyst would apply a higher discount rate to returns obtained
from debt financing. Instead, some analysts use the cost of debt, kd, because these cash
flows are related to holding debt. Others discount both cash flows, from the investment
and from the tax shield, using the cost of assets, ka. When ka is the discount rate the
method is called the capital cash flow (CCF) method. CCF contrasts with APV, which
discounts using kd. Because of the similarity between CCF and APV, CCF is not dis-
cussed further in this chapter. Regardless of which discount rate is used, the total pres-
ent value equals the sum of the two parts.

Alternatives to Discounted Cash Flow


While DCF offers a conceptually sound and transparent method for estimating the
value of a company, it relies on forecasting many uncertain variables. Comparing a
DCF valuation estimate against another method is desirable to ensure that the cal-
culations are reasonable. A simple and fairly accurate way to value a PE investment
is by using a technique called multiples-based valuation. This method is sometimes
referred to as relative valuation or using comparables. This method estimates a firms
value by comparing it to the valuations of similar firms in the current market en-
vironment. The technique often relies on an earnings measure called EBITDA as
shown in Equation 8.12:

EBITDA = EBIT + Depreciation and Amortization. (8.12)

EBIT is determined by subtracting operating costs from operating revenues. Costs of


goods sold (COGS); selling, general, and administrative expenses (SG&A); and de-
preciation and amortization are the major components of operating costs. In the
second stage of Equation 8.12, depreciation and amortizationthe two main noncash
expensesare added to EBIT because the PE firm is expected to have access to these
funds. EBITDA is a whole-firm number in that enterprise value to EBITDA estimates
the value of total assets (debt plus equity), whereas the price/earnings ratio, estimates
the value of an equity (assets minus debt). Rather than using EBITDA, some PE firms
base their valuation on sales, FCF, or invested capital.
Generally, financial analysts estimate a future year value of EBITDA based on the
current year and reasonable assumptions about changes in revenues and costs over the
coming 12 months. Then the PE firm multiplies its EBITDA estimate by an earnings-
multiple that PE managers often determine between three (on the low side) and 14 (on
the high side). Investors pay higher multiples for companies experiencing rapid growth
and higher than normal profit margins. The multiple employed may be higher for a pub-
licly listed company because those investments attract more competing PE firms bids
132 h o w p r i vat e e q u i t y w o r k s

than do those in the private sector. The multiple also spontaneously accounts for the
time value of money (i.e., it discounts future earnings back to the present). The earnings-
multiple method is sometimes referred to as the relative value method because the basis of
the size of the multiple is often what others have paid for similar investments or what the
average multiple is for similar assets in the market.
Multiples-based valuation techniques are subject to several criticisms. Perhaps the
most severe criticism is that the earnings-multiple method only estimates a single years
EBITDA rather than stream of future earnings. Advocates contend that the method is
discontinued if future years earnings are unlikely to be at a level similar to that used in
the formula. Further, the method relies on a single number, which may hide troubles in
a specific part of EBITDA, and depends on accurate choices of similar companies from
which the multiple is based.

Affecting Value by Improving Companies


After acquiring a target company, PE firms often try to raise the value of their acquisition
by either hiring restructuring professionals or performing turnaround actions to push
the firm toward a higher performance level. Efforts generally seek to improve manage-
ment, instill an entrepreneurial spirit within the company, and achieve higher levels of
growth and lower costs than previous management was capable of producing. Specific
items confronted might include compensation policies, human resource turnover and
hiring costs, information technology investment, procurement, production efficiencies,
selling and marketing efforts, distribution targets, and any other part of the business that
is underperforming. Efforts to improve the brand and customer attitudes with revenue-
enhancing and cost-reducing efforts are designed to increase EBITDA and net profitability.
A common problem in these efforts is that employees of the acquired firm may resist
change. This action is especially true as employees approach their retirement age and
for employees who have higher compensation/benefits than their productivity justifies.
Efforts to rationalize disparities in compensation can create resentment and undermine
the turnaround program. Perhaps the best way to persuade existing employees to sup-
port the companys restructuring is to provide them with coherent incentives that ap-
propriately share the returns from the turnaround with those whose efforts maximize
the benefits.
In some cases, the turnaround efforts are easy and generate substantial gains. Usually,
these gains come from obvious cost and redundancy reductions. Sales improvement ef-
forts may take longer but they can be even more rewarding than cost reduction efforts
when successful.
A PE firm that has improved a target company will profit when it sells the firm
because firm value generally rises with EBITDA. After the turnaround, the renewed
company could potentially demand a higher EBITDA multiple, which also increases
the selling price. A higher multiple may be justified if the companys growth rate has
increased after the PE firm initiates changes. Lerner, Hardymon, and Leamon (2012)
provide a more detailed discussion of PE firms as active investors. Gompers and Lerner
(1998) provide a discussion of evaluating the performance of PE investments.
Val u in g P riv at e E qu it y 133

A Comprehensive Valuation Example


The previous discussion serves as a basis for estimating the value of a potential PE in-
vestment. The three valuation methods used are DCF, APV, and multiples-based val-
uation. As with all business and financial decisions, valuing a target company involves
decision-making under uncertainty. Each of the three valuation methods has its pros and
cons and the final valuation may draw on each of these methods. Lerner and W illinge
(2002) provide a more detailed discussion of these methods.

DCF METHOD
Using the complex whole cash-flow FCF measure, FCF is computed as EBIT plus de-
preciation expense, minus gross capital expenditures, minus the change in NWC. Cash
flows are projected in detail for a forecast period (e.g., five years), and a terminal (or
horizon) value is calculated at the end of the forecast period. Enterprise value is the
present value of the FCFs over the forecast period plus the present value of the TV, all
discounted back to the present at the WACC appropriate for the target company. Enter-
prise value represents the value of the operating assets of the company. The value of the
companys equity is calculated as the enterprise value less debt.

APV METHOD
Implicit in the DCF method is the assumption that the capital structure for the target com-
pany remains constant in the future. This assumption may not be valid in highly leveraged
transactions in which debt may be high initially, but will be paid down in the future. The APV
method is a DCF method, but it values the operating cash flows and the tax shield on interest
expense as two separate items. Enterprise value is the sum of the present value of the future
cash flows for an all-equity firm, plus the present value of the tax shield on the interest de-
duction. Fenster and Gilson (1994) present a more detailed treatment of the APV method.

M U LT I P L E S - B A S E D VA L UAT I O N
Financial analysts obtain multiples for comparable companies or transactions and apply
them to the relevant inputs for the target company being evaluated. Enterprise value-
to-EBITDA (EV/EBITDA) is a commonly used metric for PE transactions. Analysts
typically assess comparable companies based on similarity of industry and business,
size, risk, and growth potential. The comparable company multiples may be trading
multiples or transaction multiples. Trading multiples are the EV/EBITDA multiples for
similar companies. Transactions multiples are the EV/EBITDA multiples paid for sim-
ilar companies in recent acquisitions. If EV/EBITDA is unsuitable for the target com-
pany (e.g., if EBITDA is negative), then the analyst may use other metrics such as the
number of customers or number of patents. Another consideration is whether a control
premium or liquidity discount is relevant. Trading multiples would not factor in these
variables, whereas transaction multiples for private companies would. Table 8.2 sum-
marizes the pros and cons of each valuation method.
134 h o w p r i vat e e q u i t y w o r k s

Table 8.2 P
 ros and Cons of Discounted Cash Flow, Adjusted Present Value, and
Multiples-Based Valuation Methods

Valuation Method Pros Cons


Discounted cash Conceptually sound: Many assumptions need
flow (DCF) Estimates intrinsic value to be made to forecast the
valuation as the present value of future cash flows and the
future cash flows discount rate
Transparent: The impact Discount rate assumes a
of various assumptions constant capital structure
and drivers of value is Enterprise value is very
easy to see sensitive to the terminal
growth rate and the
discount rate
Adjusted present Conceptually sound: Many assumptions need
value (APV) Estimates intrinsic value to be made to forecast the
valuation as the present value of future cash flows and the
future cash flows discount rate
Transparent: The impact of Enterprise value is very
various assumptions and sensitive to the terminal
drivers of value is easy to see growth rate and the
Does not depend on the discount rate
assumption of a constant
capital structure
Multiples-based Easy to apply and interpret Difficult to find appropriate
valuation Provides a market-based comparable companies or
alternative to DCF transactions
methods Provides a valuation relative
Is common industry to similar companies or
practice transactions at the time, but
not necessarily the intrinsic
value
Need to adjust private
company valuations for a
lack of liquidity
Need to adjust trading
multiples for a control
premium

Analysts can use the three valuation methods to value a potential PE investment.
This case examines Able International (Able), a hypothetical company in the industrial
products industry. Table 8.3 shows Ables historical balance sheets. Table 8.4 presents
a valuation of Able using the DCF method. The top section shows the assumptions for
the various value drivers for the company. Historical data are summarized for 2013 and
2014, followed by pro forma projections for the next 10 years.
Val u in g P riv at e E qu it y 135

Table 8.3 Able International: Balance Sheets for Years Ended December 31
2013(in $) 2014 (in $)
Assets
Cash 70,000 75,000
Accounts receivable 200,000 210,000
Inventory 120,000 130,000
Property, plant, and equipment 510,000 520,000
Total Assets 900,000 935,000
Liabilities and Equity
Accounts payable and accruals 150,000 160,000
Notes payable 50,000 50,000
Long-term debt 200,000 200,000
Common stock 150,000 150,000
Retained earnings 350,000 375,000
Total Liabilities and Equity 900,000 935,000

The assumptions in Table 8.4 are for the company operated by the current manage-
ment. The key drivers include revenue growth, COGS and SG&A costs (which deter-
mine Ables margin), NWC, capital expenditures, and the tax rate. The company has a
capital structure of 20 percent equity and 80 percent debt. The WACC is determined by
plugging the assumed inputs into Equation 8.9, with the cost of equity estimated using
the CAPM shown in Equation 8.10.
The bottom section of Table 8.4 shows the historical FCF calculations, followed by
the projected FCFs for the next 10 years. A terminal value is calculated at the end of
2024 by using the FCF for 2024, WACC, and a terminal growth rate and plugging this
information into Equation 8.8. The value per share estimate at the end of 2014 is cal-
culated by plugging the figures into Equation 8.7 to get the present value of the future
FCFs including the TV. This calculation results in the companys enterprise value or the
value of its operating assets. Subtracting the debt from the enterprise value and dividing
by the number of shares outstanding yields the estimated value per share of the stock.
The $36.08 estimated value per share is in line with the assumed current market value of
$35.10 for this hypothetical company.
A PE investor will typically make changes to the target company. In this case, assume
that by taking Able, Inc. private and combining it into similar companies in the PE firms
portfolio, revenue growth will increase by 100 basis points beginning in 2020; COGS-
to-sales will decrease by 200 basis points in 2015; SG&A to sales will decrease by 300
basis points in 2015; and inventory-to-sales will decrease by 200 basis points beginning
in 2015. Table 8.5 shows the impact of these changes on the estimated value per share.
The new value is $60.65, which is almost 60 percent higher than the initial value of
$36.08.
Table 8.4 Able International: Discounted Cash Flow Assumptions Based on Current Structure and Management
Historical Pro Forma (20152024)
Assumptions 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024
Revenue growth (%) 2.00 5.00 3.00 3.50 3.50 4.00 4.00 4.00 4.00 4.00 4.00 4.00
COGS/Sales (%) 54.00 54.57 54.00 54.00 54.00 54.00 54.00 54.00 54.00 54.00 54.00 54.00
SG&A/Sales (%) 28.50 28.75 28.00 28.00 28.00 28.00 28.00 28.00 28.00 28.00 28.00 28.00
Tax rate (%) 35.00 35.00 35.00 35.00 35.00 35.00 35.00 35.00 35.00 35.00 35.00 35.00
Cash/Sales (%) 7.00 7.14 7.00 7.00 7.00 7.00 7.00 7.00 7.00 7.00 7.00 7.00
AR/Sales (%) 20.00 20.00 20.00 20.00 20.00 20.00 20.00 20.00 20.00 20.00 20.00 20.00
Inventory/Sales (%) 12.00 12.38 12.00 12.00 12.00 12.00 12.00 12.00 12.00 12.00 12.00 12.00
AP and Accruals/Sales (%) 15.00 15.24 15.00 15.00 15.00 15.00 15.00 15.00 15.00 15.00 15.00 15.00
CAPEX/Sales (%) 6.00 5.71 6.00 6.00 6.00 6.00 6.00 6.00 6.00 6.00 6.00 6.00
136

Depreciation sales (%) 5.00 4.86 5.00 5.00 5.00 5.00 5.00 5.00 5.00 5.00 5.00 5.00
Debt/Capital (%) 20.00 20.00 20.00 20.00 20.00 20.00 20.00 20.00 20.00 20.00 20.00 20.00
Equity/Capital (%) 80.00 80.00 80.00 80.00 80.00 80.00 80.00 80.00 80.00 80.00 80.00 80.00
Risk-free rate (%) 3.50
Market risk premium (%) 5.50
Beta 1.25
Cost of equity (%) 10.38
Cost of debt (%) 4.50
WACC (%) 8.89
Terminal growth rate (%) 3.00
Share price $35.10
Shares outstanding (000) 28,500
Table 8.4continued

Historical Pro Forma (20152024)


Free Cash Flow 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024
Revenue $1,000,000 $1,050,000 $1,081,500 $1,119,353 $1,158,530 $1,204,871 $1,253,066 $1,303,189 $1,355,316 $1,409,529 $1,465,910 $1,524,546
COGS 540,000 573,000 584,010 604,450 625,606 650,630 676,656 703,722 731,871 761,145 791,591 823,255
SG&A 285,000 301,875 302,820 313,419 324,388 337,364 350,858 364,893 379,488 394,668 410,455 426,873
EBITDA 175,000 175,125 194,670 201,483 208,535 216,877 225,552 234,574 243,957 253,715 263,864 274,418
Depreciation expense 50,000 51,000 54,075 55,968 57,926 60,244 62,653 65,159 67,766 70,476 73,295 76,227
EBIT 125,000 124,125 140,595 145,516 150,609 156,633 162,899 169,415 176,191 183,239 190,568 198,191
Tax expense 43,750 43,444 49,208 50,931 52,713 54,822 57,014 59,295 61,667 64,134 66,699 69,367
EBT 81,250 80,681 91,387 94,585 97,896 101,812 105,884 110,119 114,524 119,105 123,869 128,824
137

+ Depreciation 50,000 51,000 54,075 55,968 57,926 60,244 62,653 65,159 67,766 70,476 73,295 76,227
Operating cash flow 131,250 131,681 145,462 150,553 155,822 162,055 168,537 175,279 182,290 189,582 197,165 205,051
Change in NWC 14,000 15,000 4,560 9,085 9,403 11,122 11,567 12,029 12,511 13,011 13,531 14,073
CAPEX 60,000 60,000 64,890 67,161 69,512 72,292 75,184 78,191 81,319 84,572 87,955 91,473
Free cash flow 57,250 56,681 76,012 74,307 76,908 78,641 81,787 85,058 88,460 91,999 95,679 99,506
Terminal value 1,741,566
Total cash flows 76,012 74,307 76,908 78,641 81,787 85,058 88,460 91,999 95,679 1,841,072
Enterprise value 1,278,211
Debt 250,000
Equity value 1,028,211
Shares outstanding 28,500
Value per share $36.08
Table 8.5 Able International: Discounted Cash Flow Assumptions with Private Equity Ownership
Historical Pro Forma (20152024)
Assumptions 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024
Revenue growth (%) 2.00 5.00 3.00 3.50 3.50 4.00 4.00 5.00 5.00 5.00 5.00 5.00
COGS/Sales (%) 54.00 54.57 52.00 52.00 52.00 52.00 52.00 52.00 52.00 52.00 52.00 52.00
SG&A/Sales (%) 28.50 28.75 25.00 25.00 25.00 25.00 25.00 25.00 25.00 25.00 25.00 25.00
Tax rate (%) 35.00 35.00 35.00 35.00 35.00 35.00 35.00 35.00 35.00 35.00 35.00 35.00
Cash/Sales (%) 7.00 7.14 7.00 7.00 7.00 7.00 7.00 7.00 7.00 7.00 7.00 7.00
AR/Sales (%) 20.00 20.00 20.00 20.00 20.00 20.00 20.00 20.00 20.00 20.00 20.00 20.00
Inventory/Sales (%) 12.00 12.38 10.00 10.00 10.00 10.00 10.00 10.00 10.00 10.00 10.00 10.00
AP and Accruals/Sales 15.00 15.24 15.00 15.00 15.00 15.00 15.00 15.00 15.00 15.00 15.00 15.00
(%)
CAPEX/Sales (%) 6.00 5.71 6.00 6.00 6.00 6.00 6.00 6.00 6.00 6.00 6.00 6.00
138

Depreciation/Sales (%) 5.00 4.86 5.00 5.00 5.00 5.00 5.00 5.00 5.00 5.00 5.00 5.00
Debt/Capital 20.00 20.00 20.00 20.00 20.00 20.00 20.00 20.00 20.00 20.00 20.00 20.00
Equity/Capital 80.00 80.00 80.00 80.00 80.00 80.00 80.00 80.00 80.00 80.00 80.00 80.00
Risk-free rate 3.50
Market risk 5.50
premium (%)
Beta 1.25
Cost of equity (%) 10.38
Cost of debt (%) 4.50
WACC 8.89
Terminal growth rate (%) 3.00
Share price $35.10
Shares outstanding (000) 28,500
Table 8.5continued

Historical Pro Forma (20152024)


Free Cash Flow 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024
Revenue $1,000,000 $1,050,000 $1,081,500 $1,119,353 $1,158,530 $1,204,871 $1,253,066 $1,315,719 $1,381,505 $1,450,580 $1,523,109 $1,599,265
COGS 540,000 573,000 562,380 582,063 602,436 626,533 651,594 684,174 718,383 754,302 792,017 831,618
SG&A 285,000 301,875 270,375 279,838 289,632 301,218 313,266 328,930 345,376 362,645 380,777 399,816
EBITDA 175,000 175,125 248,745 257,451 266,462 277,120 288,205 302,615 317,746 333,633 350,315 367,831
Depreciation expense 50,000 51,000 54,075 55,968 57,926 60,244 62,653 65,786 $69,075 72,529 76,155 79,963
EBIT 125,000 124,125 194,670 201,483 208,535 216,877 225,552 236,829 248,671 261,104 274,160 287,868
Tax expense 43,750 43,444 68,135 70,519 72,987 75,907 78,943 82,890 87,035 91,387 95,956 100,754
EBT 81,250 80,681 126,536 130,964 135,548 140,970 146,609 153,939 161,636 169,718 178,204 187,114
+ Depreciation 50,000 51,000 54,075 55,968 57,926 60,244 62,653 65,786 69,075 72,529 76,155 79,963
139

Operating cash flow 131,250 131,681 180,611 186,932 193,474 201,213 209,262 219,725 230,711 242,247 254,359 267,077
Change in NWC 14,000 15,000 -17,070 $8,328 8,619 10,195 10,603 13,784 14,473 15,197 15,956 16,754
CAPEX 60,000 60,000 64,890 67,161 69,512 72,292 75,184 78,943 82,890 87,035 91,387 95,956
Free cash flow 57,250 56,681 132,791 111,443 115,344 118,726 123,475 126,998 133,348 140,016 147,016 154,367
Terminal value 2,701,753
Total cash flows 132,791 111,443 115,344 $118,726 123,475 126,998 133,348 140,016 147,016 2,856,120
Enterprise value $1,978,594
Debt 250,000
Equity value 1,728,594
Shares outstanding 28,500
Value per share $60.65

Note: This table assumes a revenue growth increase of 100 basis points beginning in 2020, COGS-to-sales and SG&A-to-sales decrease by 200 and 300 basis points,
respectively, beginning in 2015, and inventory-to-sales decreases by 200 basis points beginning in 2015.
140 h o w p r i vat e e q u i t y w o r k s

Analysts can also use multiples-based valuation techniques to estimate Ables value.
Ables current sales are $1,050,000 and its EBITDA is $175,125. The estimated enter-
prise value to sales and EBITDA ratios implied by the values estimated in Tables 8.4
and 8.5 are 1.22 and 1.88 for sales and 7.3 and 11.3 for EBITDA, respectively. The data
presented in Table 8.6 show multiples data for comparable companies. These compa-
nies are in the same industry as Able and are reasonably close in size. Table 8.6 shows
transactions multiples for recent acquisitions by strategic acquirers and financial acquir-
ers as well as trading multiples for comparable companies.
Applying the mean and median of the multiples for the comparable companies results
in estimated enterprise values for Able ranging from $1.41 million to $1.73 million. The
multiples are lowest for the trading multiples and are highest for the transactions multiples.
This result is not surprising given the control premium paid in acquisitions. The values are
higher for the strategic acquirers than for the financial acquirers. The strategic acquirers
may bring more synergies to the transaction, giving them the ability to pay a higher price.
Another mechanism the PE firm may use to increase the target companys value is
to increase the leverage in its capital structure. Assume that besides making the oper-
ating improvements shown in Table 8.5, the PE firm raises Ables debt level from $250
million to $1 billion. The interest rate on the loan will be 7.5 percent and the loan will
be amortized completely over the next nine years. This scenario creates problems in
applying the DCF method because the WACC discount rate used in the DCF method
assumes a stable capital structure of the company in the future. Table 8.7 shows the fore-
casted levels of debt in the top section along with the interest expense for each year. The
amount of debt and hence the capital structure will vary from year to year.
For this highly leveraged transaction, the APV method is more appropriate because
it does not depend on the assumption that the capital structure remains constant. APV
breaks the valuation into two parts. The first part, as shown in the second section of
Table 8.7, values the companys operating cash flows, assuming that it does not use fi-
nancial leverage. These FCFs are discounted to the present at the cost of equity for an
all-equity company. Equation 8.11 provides an estimate of Ables unlevered or asset
beta. Ables current capital structure is 20 percent debt and 80 percent equity, or a debt/
equity ratio of 25 percent, and its current beta is 1.25. Unlevering this beta results in a
beta of 1.00 and an unlevered cost of equity of 9 percent. The present value of the oper-
ating cash flows discounted at the 9 percent is $1.9 billion.
The next step in the APV valuation is to compute the present value of the interest
tax shield, which appears in the third section of Table 8.7. The interest tax shield is the
interest expense for each year times the tax rate. Taking the present value of these tax
shield cash flows, using the 7.5 percent cost of debt as the discount rate, results in a value
increment of $108.5 million. Summing the value of the operating cash flows and the in-
terest tax shield results in an enterprise value of just over $2 billion. As Table 8.5 shows,
the enterprise value of $2.05 billion exceeds the enterprise value before the leverage
increase of $1.98 billion by about $70 million.
This valuation analysis shows Ables as is value as $1.28 billion or $36.08 per share,
which is reasonably close to its current share price of $35.10 as Table 8.4 shows. With
an EV-to-sales multiple of 1.22 and an EV-to-EBITDA of 7.3, Able is trading some-
what below comparable companies in the industry, which exhibit means (medians) for
these two ratios of 1.37 (1.34) and 8.24 (8.16), respectively, in Table 8.6. A PE firm can
Table 8.6 Able International: Multiples-Based Valuation
Recent acquisitions of companies comparable to Able International
Acquiring Target Date Enterprise Sales EBITDA Enterprise Enterprise Able Able
Company Company Value Value/Sales Value/ Valuation Valuation
EBITDA Based on Based on
Sales EBITDA
Strategic $1,050,000 $175,125
Acquirers
Benning, Inc. Eggleton 6/6/2014 $2,453,978 $1,278,114 $256,692 1.83 7.45
Machine
Cogan Limited Fairfield Corp. 5/4/2014 1,234,986 921,631 170,814 1.34 8.98
Dugan Zyclops, Inc. 9/12/2014 1,899,723 1,347,321 228,058 1.77 9.78
Manufacturing
141

New England Trenton Works 3/30/2014 2,101,350 1,193,949 186,787 1.65 9.85
Partners
Ohio Controls Dayton Man 2/12/2014 1,035,789 870,411 125,855 1.62 10.87
Mean 1,745,165 1,122,285 193,641 1.64 9.39 1,724,100 1,643,723
Median 1,899,723 1,193,949 186,787 1.65 9.78 1,732,500 1,712,723
Financial
Acquirers
BPZ Capital Winston 1/30/2014 $2,100,578 $1,147,857 281,957 1.92 9.56
Machine
Warbash Houghton 5/9/2014 1,130,456 843,624 125,886 1.34 7.23
Partners Corp.

continued
Table 8.6continued

Recent acquisitions of companies comparable to Able International


Acquiring Target Date Enterprise Sales EBITDA Enterprise Enterprise Able Able
Company Company Value Value/Sales Value/ Valuation Valuation
EBITDA Based on Based on
Sales EBITDA
Leon Financial Jackson, Inc. 7/15/2014 1,512,375 854,449 154,640 1.41 8.33
Melba & Co. Klongworth 10/20/2014 2,699,780 1,636,230 274,089 1.76 11.25
Ind.
Newton Gordon 11/5/2014 1,345,987 830,856 123,826 1.19 8.23
Associates International
Mean 1,757,835 1,062,603 192,080 1.52 8.92 1,600,200 1,562,115
142

Median 1,512,375 854,449 154,640 1.41 8.33 1,480,500 1,458,791


Financial data on companies comparable to Able International
Comparable Enterprise Sales EBITDA Enterprise Enterprise
Company Value Value/Sales Value/
EBITDA
Sterling Controls $1,975,645 $1,796,041 $262,370 1.10 7.53
Zolex, Inc. 1,134,589 846,708 134,112 1.34 8.46
Marburton Corp. 1,045,368 995,589 123,976 1.05 NA
Folina 1,679,800 835,721 184,189 2.01 9.12
Manufacturing
Holobuton, Inc. 2,001,445 1,460,909 254,637 1.37 7.86
Mean 1,567,369 1,186,994 142,266 1.37 8.24 1,442,700 1,443,468
Median 1,679,800 995,589 184,189 1.34 8.16 1,407,000 1,429,020
Table 8.7 Able International: Adjusted Present Value Valuation Assuming Private Equity Ownership
Historical Pro Forma (20152024)
Assumptions 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024
Beginning $250,000 $250,000 $1,000,000 $900,000 $800,000 $700,000 $600,000 $500,000 $400,000 $300,000 $200,000 $0
debt
Repayment $0 $0 $100,000 $100,000 $100,000 $100,000 $100,000 $100,000 $100,000 $100,000 $200,000 $0
Ending debt $250,000 $250,000 $900,000 $800,000 $700,000 $600,000 $500,000 $400,000 $300,000 $200,000 $0 $0
Interest rate 4.50% 4.50% 7.50% 7.50% 7.50% 7.50% 7.50% 7.50% 7.50% 7.50% 7.50% 7.50%
Interest $11,250 $11,250 $75,000 $67,500 $60,000 $52,500 $45,000 $37,500 $30,000 $22,500 $15,000 $0
expense
143

Unlevered 1.00
beta
Unlevered cost 9.00%
of equity
Terminal 3.00%
growth rate
Tax rate 35.00%
Value of Operating Cash Flows
Free cash $57,250 $56,681 $132,791 $111,443 $115,344 $118,726 $123,475 $126,998 $133,348 $140,016 $147,016 $154,367
flow
continued
Table 8.7continued

Historical Pro Forma (20152024)


Assumptions 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024
Terminal $2,649,969
value
Total cash $132,791 $111,443 $115,344 $118,726 $123,475 $126,998 $133,348 $140,016 $147,016 $2,804,336
flows
PV of cash $1,940,263
flows
144

Value of Tax Shields on Interest


Interest $75,000 $67,500 $60,000 $52,500 $45,000 $37,500 $30,000 $22,500 $15,000 $0
expense
Interest tax $26,250 $23,625 $21,000 $18,375 $15,750 $13,125 $10,500 $7,875 $5,250 $0
shield
PV of cash $108,484
flows
Enterprise $2,048,747
value

Note: The information shown is that of PE ownership described in Table 8.5.


Val u in g P riv at e E qu it y 145

employ strategies to improve Ables value. For example, operational changes may result
in higher revenue growth, higher margins, and lower levels of inventory, leading to an
enterprise value of $1.98 billion, an increase of 55 percent. Additionally, the PE firm
may be able to create additional value by increasing the value of Ables financial leverage.
The increase in leverage raises the enterprise value to $2.05 billion. The analysis of com-
parable transactions presented in Table 8.6 suggests that Able could be acquired from
$1.46 to $1.73 billion. This leaves substantial room for negotiation between the current
value of $1.28 billion and the restructured value of $2.05 billion.

Summary and Conclusions


PE firms typically generate returns for their investors by purchasing companies and
adding value by making operational and financing changes to those companies. To be
successful, a PE firm must understand the current value of a target company, the poten-
tial value of the company after operational and financing changes, and the value it must
pay to gain control of the company. Sound valuation analysis is a key component of this
business model. The valuation analysis must include a thorough and honest assessment
of the target companys future cash-flow potential, an understanding of the riskiness of
these cash flows, and an estimate of the appropriate cost of capital. Additional insights
may be gleaned from analyzing the values of comparable companies trading in the cur-
rent market and valuations of recent acquisitions of similar companies.
This chapter has presented a discussion of various factors that are a part of valua-
tion analysis along with a detailed numerical example. The content presented provides
a foundation that may be modified and extended to conduct a valuation analysis for PE.

Discussion Questions
1. Explain the difference between simple equity cash flow and complex whole cash
flow.
2. Discuss how to use the DCF method to estimate the value of a PE firm.
3. Explain why multiples-based valuation and the APV method may be appropriate in
evaluating a PE firm.
4. Discuss the means by which PE firms can increase the value of companies they
acquire.

References
Arzac, Enrique. 2005. Valuation for Mergers, Buyouts, and Restructuring. New York: John Wiley and
Sons.
Fenster, Steven R., and Stuart Gilson. 1994. The Adjusted Present Value Method for Capital
Assets. Harvard Business School Background Note No. 294047. November 1993. Revised
July 1994. Harvard Business School Press.
Gompers, Paul and Josh Lerner. 1998. Risk and Reward in Private Equity Investments: The Chal-
lenge of Performance Assessment. Journal of Private Equity 2 (Winter), 512.
146 h o w p r i vat e e q u i t y w o r k s

Lerner, Josh, Felda Hardymon, and Ann Leamon. 2012. Private Equity, Venture Capital, and the
Financing of Entrepreneurship: The Power of Active Investing. New York: John Wiley and Sons.
Lerner, Josh, and John Willinge. 2002. A Note on Valuation in Private Equity Settings. Harvard
Business School Note No. 9297050. Harvard Business School Press.
PE Growth Capital Council. 2013. PitchBook. PEGCC Analysis.
QuotationBest.com. 2014. Albert Einstein Quotes. Available at http://www.quotationbest.com/
search/quotes/?q=einstein.
Scharfman, Jason A. 2012. Private Equity Operational Due Diligence: Tools to Evaluate Liquidity, Valu-
ation, and Documentation. New York: John Wiley and Sons.
Stowell, David. 2012. Investment Banks, Hedge Funds, and Private Equity, 2d Edition. Waltham, MA:
Academic Press.
9
Cost of Capital for Private Equity
ALAIN CON
Professor of Finance, ESG-UQM, University of Quebec in Montreal

AURLIE DESFLEURS
Associate Professor, University of Sherbrooke

Introduction
Many studies have analyzed private equity (PE) funds since the end of the 1990s. The
growth of mutual funds and greater emphasis on asset management have resulted in an
increase in the number and size of PE funds. A current critical topic in finance literature
is the cost of capital for PE firms. This chapter distinguishes among the different classes
and sub-classes of PE firms: buyouts (BOs), mergers, venture capital (VC), incubators,
and mezzanine. The chapter also focuses on their differences when estimating the cost
of capital of PE and the difficulties of obtaining the data needed to make these calcula-
tions. In this context, the analysis of the cost of capital is more difficult than for publicly
traded firms. Similar to hedge funds, PE funds are opaque and illiquid. These character-
istics directly affect risk for potential investors and the cost of capital estimates.
Although measuring the cost of capital for PE funds does not differ sharply from the
standard paradigm established by the modern finance theory, some nuances exist such
as the opacity of financial statements, lack of liquidity, and idiosyncratic risk. These fac-
tors affect the definition of the cost of debt and amplify the risk. This chapter considers
these factors because such nuances have important implications for the cost of equity
estimates for PE.
Because financial literature on PE emphasizes analyzing performance, researchers,
with a few exceptions such as Groh and Gottschalg (2011), often neglect relevant issues
when measuring the cost of capital. Since the late 1990s, an important and growing lit-
erature examines the analysis of risk-return relationships in the PE industry. As Fleming
(2010) reports, performance studies focus on both PE firms and funds-of-funds of PE.
Although using the internal rate of return (IRR) is an accepted performance measure-
ment tool for PE firms and funds-of-funds, this method may not be the most relevant
measurement. Contrary to other asset managers who do not adjust for cash flows, PE
managers can implement opportunistic investment strategies. As Fleming notes, they
can choose timing of drawdowns into the fund distributions of cash.

147
148 h o w p r i v a t e e q u i t y w o r k s

Other performance measures are also available. For example, using the modified in-
ternal rate of return (MIRR), cash flows are assumed to be reinvested at the cost of
capital. Some use the return on an index such as the S&P 500 index as the reinvest-
ment rate (Franzoni, Nowak, and Phalippou 2012), which introduces an opportunity
cost as a benchmark. Ljungqvist and Richardson (2003) suggest using a profitability
index, which is the present value of cash inflows divided by initial outflows. Using this
method, some analysts discount outflows at the risk-free rate and inflows at the return
on a market index such as the S&P 500 index (Phalippou and Gottschalg 2009; Fran-
zoni et al. 2012). Kaplan and Schoar (2005) introduce the public market equivalent
(PME) as a performance metric. As defined by Kaplan and Schoar (2005), the PME
compares an investment in a PE fund with an investment in the S&P 500 index. The
total return of the S&P 500 index is used as the investment rate. The value of the out-
flows of the fund invested at the investment rate is compared at the value of the inflows
invested at the same rate. The PE funds outperformance compared with the return on
the S&P 500 index is illustrated by a PME greater than one.
The main question is whether PE returns (adjusted or unadjusted) are attractive
for investors. The results from market-adjusted return studies are mixed. As Jones and
Rhodes-Kropf (2003), Ljungqvist and Richardson (2003), Lerner, Schoar, and Wong-
sunwai (2007), and Phalippou and Gottschalg (2009) report, the IRRs generated from
PE funds and public equity are similar. However, drawing conclusions on the perfor-
mance of public equity versus PE over a long period is difficult. This problem exists for
both VC and BOs. Moskowitz and Vissing-Jorgensen (2002) show that returns on PE
are not higher than those on public equity. Focusing on U.S. BOs, Groh and Gottschalg
(2011) report contradictory findings in the financial literature.
According to Ljungqvist and Richardson (2003), PE funds generate excess returns of
5 to 8 percent relative to the PME. To compensate investors for holding a 10-year illiquid
investment, the excess value of PE funds is 24 percent relative to the present value of the
invested capital. They find an average beta for the different PE fund portfolios of 1.08,
slightly riskier than the market portfolio. While the S&P 500 index shows a performance
of 14.1 percent, PE fund portfolios show an IRR of return of 21.83 percent. Phalippou
and Gottschalg (2009) find that PE funds underperform the S&P 500 index by about 3
percent, net of fees. Estimated fees are about 6 percent a year. When they adjust returns
for risk, the underperformance increases from 3 to 6 percent. According to Jones and
Rhodes-Kropf (2003), PE funds do not produce positive alphas when considering the
role of idiosyncratic risks in PE valuation. Idiosyncratic risk for BOs is positively related
to higher returns. Examining idiosyncratic risk for BOs, Groh and Gottschalg (2011)
highlight the presence of high Sharpe ratios compared to more diversified portfolios.
Cochrane (2005) studies whether VC investments behave the same way as publicly
traded securities. He estimates the risk and return of VC projects and corrects for selec-
tion bias. The bias is generated by projects that remain private at the end of the sample
period. Therefore, he uses a maximum likelihood estimate that reduces both the esti-
mate of the mean log return from 108 to 15 percent and the log market model intercept
from 92 to 7 percent. Cochrane reports that the selection bias correction significantly
reduces the high average returns initially observed. Finally, comparing VC investments
with smaller NASDAQ stocks, he reports similar large mean log returns as well as high
volatilities and alphas during the same observation period.
Cost of C apit al for P riv at e E qu it y 149

Kaplan and Schoar (2005) suggest using a PME approach to benchmark PE transactions.
Despite the presence of important heterogeneity across funds, they show that average PE
fund returns (net of fees) are similar to the S&P 500 index. VC funds outperform the S&P
500 index while BO returns are significantly smaller than the returns on the S&P 500 index.
Groh and Gottschalg (2011) propose a large-scale analysis on the cost of capital for
BOs that corrects for operating risk, leverage risk, and leverage cost. They show that the
average cost of capital is below that of the companies comprising the S&P 500 index.
Using a database that covers 85 percent of capital raised by U.S. BO funds, Higson and
Stcke (2012) find that BO funds have consistently outperformed the S&P 500 index
since 1980. Groh and Gottschalg (2011) also find that the top-decile performance
drives excess returns rather than the performance of top-quartile funds. Their results
show a decline in absolute returns during the last three decades.
Harris, Jenkinson, and Kaplan (2014) analyze the performance of different types of
PE funds. Their results show that BO funds outperform public equity by 3 percent an-
nually. VC funds outperform public equity during the 1990s and underperform public
equity during the 2000s.
Bchner and Stcke (2014) introduce a new econometric approach to estimate the sys-
tematic risk and abnormal returns of illiquid assets. Their estimations are consistent with
beta coefficients of 2.5 to 3.1. In this case, the price of systematic risk is significantly higher
than previously estimated and generally assumed. After adjusting for high management fees
and carried interest, they report that alphas for BO funds are slightly negative but not statis-
tically different from zero. Net alphas for VC are still positive and range from 2 to 5 percent.
When Bchner and Stcke use the Fama-French all U.S. stock market index, they report
lower market betas around 2.4 and positive alphas for both BO and VC funds.
After illustrating stylized facts on the cost of capital for PE, this chapter contributes
to the literature in two ways. First, it sheds light on the presence of errors-in-variables
(EIVs) in the linear asset pricing model used in the PE industry. The chapter suggests
using the Dagenais and Dagenaiss (1997) instrumental variables estimator to address
the important consequences of EIVs on the cost of capital. Second, following a method
well acknowledged in the hedge fund industry, this chapter recommends using an ad-
justment initiated by Getmansky, Lo, and Makarov (2004) to solve the problem of the
stale valuation observed in the PE industry.
The reminder of the chapter is organized as follows. The next section discusses asset
pricing models used in the PE literature. The data and time-varying performance esti-
mates (betas) for the different styles of PE are then discussed. The theoretical frame-
work is then explained with the Dagenais and Dagenaiss (1997) higher moment
estimator (DDHME) model. The following section reports the empirical results. The
final section presents a summary and conclusions.

Asset Pricing Models and the Cost of Private Equity


In a risk-return environment, as defined by Markowitz (1952), the cost of capital is
directly related to measuring risk. PE asset pricing models are similar to asset pricing
models proposed in financial theory. Analysts and researchers use single and multifactor
models to estimate PE risk by analyzing beta and alpha coefficients.
150 h o w p r i v a t e e q u i t y w o r k s

The estimates of beta and alpha coefficients show sharp contrasts among the
different PE strategies, time periods, market benchmarks, and multifactor models
used. Several important findings emerge. For VC funds, Gompers and Lerner
(1997) find a beta of 1.08 similar to that of Ljungqvist and Richardson (2003),
whereas Cochrane (2005) and Metrick (2007) report estimates of 1.70 and 0.81,
respectively.
PE returns are stale and suffer from illiquidity caused by nonsynchronous trad-
ing. As first reported by Dimson (1979) for mutual funds, this stale valuation issue
is related to the serial correlation of returns. The main effects of stale valuation
are underestimation of risk and overestimation of excess returns. Further, esti-
mates are statistically biased (in the presence of measurement errors). Jones and
Rhodes-Kropf (2003) argue that staleness is in the net asset value (NAV). As re-
cently reported by Ewens, Jones, and Rhodes-Kropf (2013), if NAV is adjusted in a
time-stationary way with one-period lag, the problem is similar to the nonsynchro-
nous trading problem. An adjustment proposed by Dimson can be used with some
difficulty to correct for staleness. A standard solution to this estimation problem
is to use lagged estimators. Cochrane (2005) and Metrick (2007) report higher
beta estimates of 2 and 1.83, respectively, for VC in a capital asset pricing model
(CAPM) framework.
As Cochrane (2005) and Phalippou and Zollo (2005) note, the equity risk premium
is unstable and time-varying. Cochranes results show that beta estimates decline over
time from the initial investment especially for VC funds. Moreover, Cochrane shows
that for VC firms, mature firms are less risky than young firms.
Using the CAPM as the starting point for the development of asset pricing models,
the PE literature offers various multifactor models. The Fama-French three-factor
model (Fama and French 1993, 1997) adds size (small minus big or SMB) and value
(high minus low or HML) factors to the market factor (MKT) and stands as a bench-
mark (Gompers and Lerner 1997; Jones and Rhodes-Kropf 2003; Metrick 2007;
Franzoni et al. 2012; Ewens, Jones, and Rhodes-Kropf 2013; Pedersen, Page, and
He 2014). Lagged market premiums can be used to address the stale data problem.
Kaplan and Ruback (1995), Ljungqvist and Richardson (2003), and Phalippou and
Zollo (2005) suggest using a method based on an unlevered industry beta to capture
the riskiness of PE funds. More recently, Franzoni et al. (2012) in the spirit of Pastor
and Stambaugh (2003) introduce a liquidity premium, defined as the return differ-
ence between portfolios of low liquidity stocks and portfolios of high liquidity stocks.
They find significant variation based on market, value, and liquidity factors but not on
size. With a four-factor model, alpha is zero and the liquidity risk premium is about
3 percent.

ASSET PRICING MODELS


The standard CAPM as developed by Sharpe (1964), Lintner (1965), and Mossin
(1966) and multifactor models including the Fama and French (1993, 1997) model,
Carhart (1997) model, and Pastor and Stambaugh (2003) model are used to determine
asset prices. Thus, a maximum of five risk factors is used to decompose the returns of
PE funds: (1) a market factor (MKT), (2) size factor (SMB), (3) value factor (HML),
Cost of C apit al for P riv at e E qu it y 151

(4) momentum factor (up minus down or UMD), and (5) liquidity factor (LIQ). Equa-
tion 9.1 presents a standard linear asset pricing model:
K
R t = + k Fkt + e t , (9.1)
k=1

where is a constant term defined as the securitys abnormal return, or Jensens alpha
( Jensen 1968), Fkt is factor k realization in period t, k s factor k loading, and e t is a
residual idiosyncratic risk.

T H E D ATA
The data consist of 58 public funds and firms included in the S&P Listed PE Index as
of March 2014. The returns for each fund and firm are obtained from Bloomberg. The
index constituents are divided into four categories following the transaction type de-
scriptions given by the S&P Listed PE Index. Then an equally weighted monthly index
is calculated for each category from January 2000 to December 2012.
The first category (GRO) consists of 31 funds and firms and is devoted to growth
capital and VC but may include some BOs. The second category is devoted to BOs with
10 funds and firms. The third category includes only mezzanine (MEZ) with 13 funds
and firms. Finally, an equally weighted index is computed with all public funds and firms
listed in the S&P Listed PE Index using the acronym PE.
Table 9.1 provides the descriptive statistics of the computed indexes. The statistics in-
clude the risk factors for MKT (market premium), SMB (size factor), HML (value factor),
UMD (momentum factor), and LIQ (liquidity). These risk factors are obtained from
Kenneth Frenchs website (French 2014). Data for LIQ as described by Pastor and St-
ambaugh (2003) are obtained from Pastors website (Pastor 2014). RF is the one-month
T-bill return from Ibbotson and Associates, Inc., available on Kenneth Frenchs website.
Descriptive statistics show sharp contrasts among categories. While the growth and
VC index (GRO) and PE index exhibit negative returns for the full period, with 0.53
and 0.41 percent, respectively, the buyout index (BO) and mezzanine index (MEZ)
report positive returns, with 0.30 and 0.21 percent, respectively. The amplitude of the
distribution is important as highlighted by the minimum return and the maximum
return from 49.16 percent for MEZ to 25.05 percent for GRO. These results should be
compared for the same period to the market premium (0.11 percent) and the risk-free
rate (0.18 percent). More interestingly, an analysis of the higher moments of the distri-
bution of returns show that while the standard deviation ranges from 7 percent (PE) to
7.69 percent (MEZ), negative skewness ranges from 0.56 (GRO) to 2.06 (MEZ),
and kurtosis ranges from 1.98 (GRO) to 12.59 (MEZ). These statistics highlight impor-
tant differences among the categories of PE funds.

A N I L L U S T R AT I O N O F P R I VAT E E Q U I T Y R I S K U S I N G
T H E C A P I TA L A S S E T P R I C I N G M O D E L
Using the CAPM estimated by ordinary least squares (OLS) with a regression of 24
monthly observations, the beta is reported for each computed index. Risk is captured by
152 h o w p r i v a t e e q u i t y w o r k s

Table 9.1Descriptive Statistics: January 2000 to December 2012, Using


Monthly Data

Indexes Mean Standard Skewness Kurtosis Minimum Maximum


Deviation
GRO 0.53 7.39 0.56 1.98 25.28 25.05
BOs 0.30 7.31 1.62 5.59 38.49 19.46
MEZ 0.21 7.69 2.06 12.59 49.16 22.33
PE 0.41 7.00 1.13 4.10 34.26 21.08
Factors
MKT 0.11 4.77 0.51 0.57 17.23 11.34
SMB 0.42 3.66 1.00 10.12 16.39 22.02
HML 0.54 3.51 0.06 3.28 12.68 13.87
RF 0.18 0.17 0.62 0.99 0.00 0.56
UMD 0.15 6.15 1.50 8.05 34.72 18.39
LIQ 0.79 4.24 0.53 3.06 10.14 21.01

Note: This table provides descriptive statistics for indexes and factors. All data are in percent. MKT =
Fama and French market portfolio factor, SMB = Fama and French size factor, HML = Fama and French
book-to-market factor, UMD = Carhart momentum factor, LIQ = Pastor and Stambaugh liquidity factor.
RF = risk-free available on the website of Kenneth French. GLM is the acronym for the Getmansky et al.
(2004) adjustment. The indexes for PE funds are computed from the S&P Listed Private Equity Index
and described as follows: GRO refers to growth capital and some rare BOs; MEZ refers to mezzanine;
PE refers to the equally weighted index computed from the S&P Listed Private Equity Index.

measuring the volatility of the computed beta. The S&P 500 index from January 2000
to December 2012 is used as a benchmark. Results reported in Figures 9.1 to 9.4 reveal
important contrasts.
Figure 9.1 reports that for the growth and VC index (GRO) betas are globally higher
than the S&P 500 index and may reasonably be compared to aggressive stocks during

2
1.8
Value of beta(market)

1.6
1.4
1.2
1
0.8
0.6 GRO
0.4 S&P 500
0.2
0
2002 2004 2006 2008 2010 2012

Figure 9.1 Beta: GRO Index vs. the S&P 500 Index This figure compares beta for GRO
index with that of S&P 500 index. The indexes for PE funds are computed from the S&P Listed
Private Equity Index and described as follows: GRO = Growth Capital and some rare buyouts.
Cost of C apit al for P riv at e E qu it y 153

2.5

Value of beta(market) 1.5

1
MEZ
0.5 S&P 500

0.5
2002 2004 2006 2008 2010 2012

Figure 9.2 Beta: MEZ Index vs. the S&P 500 Index This figure compares beta for
MEZ index with that of S&P 500 index. The indexes for PE funds are computed from the
S&P Listed Private Equity Index and described as follows: MEZ = mezzanine.

the period. The betas of the S&P 500 index range from 0.86 to 1.04 during the full
sample period. For the growth and VC index (GRO), the betas range from 0.77 in 2007
at the early stage of the global financial crisis to 1.71 at the end of 2003. As Figure 9.1
shows, the GRO betas dominate the betas of the benchmark, except for two periods: in
2007 and at the end of 2011.
As illustrated by Figure 9.2, for the mezzanine index (MEZ), low betas occurred
before the global financial crisis with a minimum of 0.18 in 2002 and increased here-
after with a peak at 2.12 in 2010. Mezzanine funds may be reasonably considered as de-
fensive funds before 2008 (with lower betas than the benchmark) and aggressive funds
during the period after the global financial crisis (at least for our sample).
As already illustrated for the growth and VC index (GRO) (Figure 9.1), for BO index
(Figure 9.3), betas are higher than the S&P 500. The trend for the BO index is neverthe-
less quite different. The BO betas are continuously higher than the S&P 500 index from
the end of 2006. During the first period (from 2000 to 2006) the results are contrasted
(the volatility is indeed important). From 2000 to 2012, the results of the estimates are
a minimum of 0.55 in 2005 and a maximum of 1.92 at the end of 2008.

2.5
Value of beta(market)

1.5

1
BUYOUT
0.5
S&P 500
0
2002 2004 2006 2008 2010 2012

Figure 9.3 Beta: Buyouts vs. the S&P 500 Index This figure compares beta for the
BO index with that of the S&P 500 index. The indexes for PE funds are computed from the
S&P Listed Private Equity Index.
154 h o w p r i v a t e e q u i t y w o r k s

1.8
1.6
Value of beta (market) 1.4
1.2
1
0.8
0.6
INDEX PE
0.4
0.2 S&P 500
0
2002 2004 2006 2008 2010 2012

Figure 9.4 Beta: Private Equity Index vs. the S&P 500 Index This figure
compares beta for the PE index with that of the S&P 500 index. The indexes for PE funds
are computed from the S&P Listed Private Equity Index and described as follows: PE = the
equally weighted index computed from the S&P Listed Private Equity Index.

For the PE index (Figure 9.4), the betas tend to be greater than one and greater than
the beta of the S&P 500 index (with a maximum of 1.64 and a minimum of 0.72). These
results confirm the relative aggressiveness of PE funds. The findings are consistent with
the previous literature. High betas and their volatilities directly affect the risk premium.
As illustrated by Figures 9.1 to 9.4, the risk inherent to PE is higher than the S&P 500.
This stylized fact leads to an increase of the cost of equity, which in turn increases the
cost of capital, all things equal. Increased risk also has a direct impact on the cost of debt
that tends to grow (i.e., as acknowledged in the financial literature higher risk tends to
lead to a higher cost of debt, all things equal).
The following section proposes to price this risk through a decomposition of factor
loadings or sources of risk able to explain the variation of PE returns. Asset pricing models,
especially linear asset pricing models, suffer from the problem of EIVs. In this case esti-
mates obtained by OLS regressions are biased, which affects estimating the cost of capital
(Con and Racicot 2007). Using instrumental variables (IV) is required to correct this
bias. Therefore, a possible adjustment technique based on IV is reported. Using the general
method of moments is also suitable but its main drawback is the choice of a relevant IV.

Errors-in-Variables and Linear Asset Pricing Models


Correcting for EIVs is important to accurately estimate asset pricing models. As ac-
knowledged in financial and econometric literature, the main problem with multifac-
tor asset pricing models such as the one shown in Equation 9.1 is the use of observed
factors, Fkt, instead of the (generally unobserved) true factors, Fkt. Con and Racicot
(2007) as well as Carmichael and Con (2008) report that EIVs in linear asset pricing
models lead to biased and inconsistent parameter estimators. EIVs induce partial cor-
relation between residual and regressors. Here, the usual solution is to use IVs.
As Dagenais and Dagenais (1997), Carmichael and Con (2008), Con and Hbner
(2009), and Bodson, Con, and Hbner (2010) find, higher moment estimators are
very accurate and are appropriate candidates to estimate linear asset pricing models.
Cost of C apit al for P riv at e E qu it y 155

This chapter assumes that the true unobserved variables are non-Gaussian and mea-
surement errors are normally distributed. The Dagenais and Dagenaiss (1997) higher
moment estimators are generated and applied with a two-step artificial regression,
as described by Davidson and McKinnon (2004). First, estimates of EIVs, W are com-
puted as the residuals of k OLS regressions with observed factors, F, as dependent vari-
ables, while the instruments are computed as regressors (higher moments of F are used
to define estimates of EIVs, such as F F = W
with F, estimates of the true factors).
Second, estimates of EIVs are added as additional regressors in the dynamic asset
pricing Equation 9.2 as follows:
K K

R t = tHM + ktHM Fkt + kt w kt + t , (9.2)


k=1 k=1

where exponent HM stands for Dagenais and Dagenaiss HME (hereinafter DDHME).
Durbin-Wu-Hausman (hereinafter DWH) type test is used (Hausman 1978) and re-
ported in Tables 9.2 to 9.6 to detect and exhibit the presence of EIVs.

Estimation of the Cost of Private Equity:


Empirical Results
This section reports the main empirical results of the estimation process described pre-
viously. It also analyzes the corrected estimates and their impact on the valuation of the
cost of equity, and thus on the cost of capital of PE.

D E C O M P O S I T I O N O F R I S K I N P R I VAT E E Q U I T Y
To show financial risk decomposition and its impact on the cost of PE, six linear and
standard asset pricing models are used and tested. A correction for EIVs is systemati-
cally applied to all specifications. As reported by the DWH statistic, this correction
is necessary and most relevant because OLS estimates are systematically biased. The
DWH statistic is always statistically significant. Because the cost of capital with OLS
estimates would be biased, a correction for EIVs is needed. The six linear asset pricing
models tested for the four indexes (GRO, BO, MEZ, and PE) are respectively:

The CAPM with risk factor MKT.


The augmented CAPM in the spirit of Pastor and Stambaugh (2003) to capture the
liquidity premium (LIQ): MKT and LIQ.
The standard Fama-French model with three premiums: MKT, SMB (size anomaly),
and HML (book-to-market anomaly).
The Carhart model, an extension of the Fama-French model with the momentum
factor, UMD, and thus four premiums: MKT, SMB, HML, and UMD.
An augmented Fama-French model taking into account the liquidity premium
(LIQ) with four premiums: MKT, SMB, HML, and LIQ.
An augmented Carhart model including a fifth risk factor related to illiquidity (LIQ):
MKT, SMB, HML, UMD, and LIQ.
156 h o w p r i v a t e e q u i t y w o r k s

This decomposition is applied to all four indexes and shows significant results that have
direct consequences on the cost of capital of PE funds, especially on risk valuation.
As mentioned earlier, the necessary correction for EIVs or measurement errors
provides robust estimates that can be used to compute the expected cost of PE. The
DWH statistic is often statistically significant at the 0.01 level. To highlight the ro-
bustness and accuracy of the reported results, all statistics in Tables 9.2 to 9.5 are
computed with a standard correction heteroskedasticity. These adjustments and cor-
rections are a required and necessary first step to analyze the estimates. In the same
spirit, the Fisher test (F), Durbin-Watson test (DW), Schwarz criterion (SC), and
Akaike information criterion (AIC) are reported and all confirm the statistical rel-
evance of the model specifications analyzed in this chapter. The analysis of each index
is reported below.

Growth Index (GRO)


Table 9.2 reports the results of growth index (GRO). For all model specifications, the
R 2 is high, ranging from 0.75 for the two-factor (MKT and LIQ) to 0.82 for the five-
factor model (MKT, SMB, HML, UMD, and LIQ). This result highlights the relevance
of the specific model choice to explain the variation of PE funds returns. All alpha esti-
mates, related to abnormal returns, are negative and statistically significant at the 0.05
level and sometimes at the 0.01 level. Alpha ranges from 0.90 percent for the CAPM
to 1.3 percent for the five-factor model. With this negative estimate, the computed
growth index (GRO) is not attractive for investors during the January 2000 to Decem-
ber 2012 sample period. It could be short sold. The beta of market risk is high with a
value of 1.73 and highly statistically significant. Both PE related to growth and VC are
riskier than the market.
Using the standard linear asset pricing models, a decomposition of the risk is re-
ported with different factor loadings in Table 9.2. The following conclusions can be
drawn. The most relevant asset pricing model is the five-factor model. The beta of
market risk is still high with a statistically significant estimate of 1.49. The size effect,
captured by SMB, and the momentum effect, UMD, are both statistically significant at
the 0.01 level with estimates of 0.47 and 0.28 respectively. The book-to-market effect
and the liquidity effect are not statistically significant at normal levels. The results clearly
show that the main source of risk is market risk, with high statistically significant esti-
mates from 1.49 to 1.73. The liquidity premium for the GRO index is positive but not
statistically significant. These findings should be taken into account to compute the cost
of capital of PE and VC growth.

Buyout Index (BO)


Table 9.3 reports the results of buyout index (BO).The DWH statistic is statistically
2
significant at the 0.01 level for all asset pricing model specifications. The R high, rang-
ing from 0.62 for the CAPM to 0.71 for the five-factor model. All alpha estimates are
negative and statistically significant at the 0.01 level for the augmented Fama-French
model with the liquidity factor (LIQ) (1.92 percent) and for the five-factor model
(1.91 percent). This is a clear illustration of underperformance of GRO index during
the period.
Table 9.2 Growth, Venture Capital, and a Few Buyouts: GRO

R2 F DW SC AIC DWH MKT SMB HML UMD LIQ

0.75 235.19 2.10 5.53 5.47 4.14** 0.90 1.73


(2.60) (6.47)
0.75 117.26 2.09 5.59 5.49 1.81 1.00 1.66 0.14
(2.70) (7.52) (1.30)
0.77 89.55 2.03 5.54 5.40 2.39* 1.10 1.56 0.24 0.22
(2.55) (8.83) (1.05) (0.79)
0.82 90.94 1.86 5.35 5.17 3.78*** 1.23 1.52 0.51 0.33 0.24
(3.29) (8.77) (3.02) (1.14) (3.23)
0.77 67.44 1.99 5.59 5.41 2.04* 1.26 1.55 0.22 0.21 0.23
(2.74) (9.67) (1.39) (1.05) (1.29)
0.82 73.66 1.84 5.39 5.17 3.18*** 1.30 1.49 0.47 0.25 0.28 0.17
(3.04) (8.33) (3.33) (1.11) (4.61) (0.98)

Note: The alpha is expressed in percentages. F and DW are respectively the statistics for the Fisher test and the Durbin-Watson test. SC and AIC are respectively Schwartz
Criterion and Akaike Information Criterion. DWH is a standard q where q is the number of adjustment variables to detect EIVs. The DWH test is a heteroskedasticity
2

robust test. All HME statistics are computed with White (1980) H0 heteroskedasticity-consistent covariance matrix estimators (HCCME). t-statistics are reported in brackets
under the estimates. MKT = Fama and French market portfolio factor, SMB = Fama and French size factor, HML = Fama and French book-to-market factor, UMD =
Momentum factor (Carhart, 1997), LIQ = Pastor and Stambaugh liquidity factor.
*, **, *** Significant at the 0.10, 0.05, and 0.01 levels, respectively.
Source: Available at http://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_library.html and http://faculty.chicagobooth.edu/lubos.pastor/research/liq_data_
1962_2012.txt.
158 h o w p r i v a t e e q u i t y w o r k s

Three points help to explain the variation in PE returns. First, contrary to the growth
index (GRO), the liquidity premium is statistically significant at the 0.01 level: 0.78 for
the two-factor model, 1.17 for the augmented Fama-French model, and 1.16 for the
five-factor model. Second, the momentum is positive but never statistically significant.
The market risk (beta) estimate is high, ranging from 2.01 for the CAPM to 1.29 for the
five-factor model. Third, the size (SMB) and book-to-market factors (HML) are always
statistically significant. These results highlight the difference among different categories
of PE. Taking into account these findings may be useful to compute an accurate cost of
capital of PE.

Mezzanine Index (MEZ)


Table 9.4 reports the results of mezzanine index (MEZ). As reported by the DWH
statistic in Table 9.4, the correction for EIVs is needed for all model specifications to
2
overcome biased OLS estimates. The R ranges from 0.37 for the CAPM to 0.68 for
the augmented Fama-French model with the liquidity premium (LIQ). The contribu-
tion of the Fama-French model with an increase of R 2 to 0.62 is relevant. As previously
observed for the GRO and BO indices, all alpha estimates are negative and statistically
significant at the0.10 level for the augmented Fama-French model and the five-factor
model. For these specifications the liquidity premium is statistically significant at the
0.05 level with a value of 1.30 and 1.22, respectively. The momentum factor is not statis-
tically significant in the Fama-French model, whereas it exhibits a statistically significant
value of 0.23 in the five-factor model specification. The book-to-market factor is statis-
tically significant for all specifications, ranging from 2.39 for the Fama-French model to
1.4 for the five-factor model.
The addition of risk factors tends to decrease the estimate of the market risk from
1.51 for the single factor (i.e., the CAPM) to 0.87 for the five-factor model. This pro-
posed decomposition can be applied to compute the cost of capital of mezzanine
strategies.

Private Equity Index


Table 9.5 reports the results of Private Equity (PE) Index. As previously reported for
the other indices, the DWH statistic is statistically significant. The R 2 is high, ranging
from 0.78 for the simple CAPM to 0.85 for the five-factor model. All alpha estimates
are statistically negative: from 0.8 percent for the CAPM to 1.73 percent for the
augmented Fama-French model (with liquidity premium). The market risk is high and
always statistically significant at the 0.01 level and ranges from 1.84 for the CAPM to 1.4
for the five-factor model. The liquidity premium is statistically significant for the three
asset pricing specifications ranging from 0.52 in the two-factor model to 0.71 in the
augmented Fama-French model. This finding shows that liquidity should be considered
when computing the cost of capital of PE. SMB and HML are statistically significant
(five-factor model). The momentum effect should not be neglected and is significant at
the 0.01 level in the five-factor model specification.
These findings shed new light on the decomposition of PE returns. The parsimoni-
ous and relevant use of the risk factors tested should be considered when computing the
cost of capital of PE.
Table 9.3 Buyouts

R2 F DW SC AIC DWH MKT SMB HML UMD LIQ


0.62 126.36 1.92 5.93 5.88 12.32*** 0.10 2.01
(0.20) (3.56)
0.65 68.70 1.78 5.94 5.84 7.91*** 0.69 1.77 0.78
(1.18) (4.72) (3.56)
0.66 50.67 2.01 5.93 5.79 6.14*** 1.53 1.47 1.38 1.70
(1.39) (2.56) (2.22) (1.86)
0.65 37.63 1.96 5.99 5.81 3.96*** 1.19 1.34 0.97 1.37 0.17
(1.35) (2.35) (3.04) (1.92) (0.73)
0.70 46.24 1.73 5.85 5.67 7.48*** 1.92 1.25 0.90 1.12 1.17
(2.79) (5.23) (3.56) (3.69) (6.91)
0.71 38.77 1.75 5.87 5.65 6.31*** 1.91 1.29 0.89 1.11 0.02 1.16
(2.56) (4.70) (3.72) (3.18) (0.16) (5.28)

Note: The alpha is expressed in percent. F and DW are respectively the statistics for the Fisher test and the Durbin-Watson test. SC and AIC are respectively Schwartz
Criterion and Akaike Information Criterion. DWH is a standard q where q is the number of adjustment variables to detect EIVs. The DWH test is a heteroskedasticity
2

robust test. All HME statistics are computed with White (1980) H0 heteroskedasticity-consistent covariance matrix estimators (HCCME). t-statistics are reported in brackets
under the estimates. MKT = Fama and French market portfolio factor, SMB = Fama and French size factor, HML = Fama and French book-to-market factor, UMD =
Momentum factor (Carhart, 1997), LIQ = Pastor and Stambaugh liquidity factor.
*, **, *** Significant at the 0.10, 0.05, and 0.01 levels, respectively.
Source: Available at http://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_library.html and http://faculty.chicagobooth.edu/lubos.pastor/research/liq_data_
1962_2012.txt.
Table 9.4 Mezzanine: MEZ

R2 F DW SC AIC DWH MKT SMB HML UMD LIQ

0.37 46.05 1.68 6.54 6.48 3.01* 0.14 1.51


(0.23) (1.48)
0.39 26.16 1.60 6.55 6.45 4.58*** 0.81 1.35 0.87
(0.84) (1.67) (1.46)
0.62 43.68 1.92 6.12 5.99 4.14*** 1.88 1.27 1.16 2.39
(1.21) (1.44) (1.41) (1.76)
0.62 32.00 1.93 6.19 6.02 2.13* 1.53 1.08 0.81 2.03 0.04
(1.34) (1.37) (2.23) (2.06) (0.14)
160

0.68 42.39 1.79 6.01 5.83 7.02*** 2.01 0.90 0.36 1.42 1.30
(1.77) (1.91) (0.88) (2.41) (2.22)
0.68 33.27 1.84 6.08 5.86 4.79*** 1.98 0.87 0.56 1.40 0.23 1.22
(1.80) (1.73) (1.58) (2.41) (1.67) (2.06)

Note: The alpha is expressed in percent. F and DW are the statistics for the Fisher test and the Durbin-Watson test, respectively. SC and AIC are the Schwartz Criterion and
Akaike Information Criterion, respectively. DWH is a standard q where q is the number of adjustment variables to detect EIVs. The DWH test is a heteroskedasticity robust
2

test. All HME statistics are computed with White (1980) H0 heteroskedasticity-consistent covariance matrix estimators (HCCME). t-statistics are reported in brackets under
the estimates. MKT = Fama and French market portfolio factor, SMB = Fama and French size factor, HML = Fama and French book-to-market factor, UMD = Momentum
factor (Carhart, 1997), LIQ = Pastor and Stambaugh liquidity factor.
*, **, *** Significant at the 0.10, 0.05, and 0.01 levels, respectively.
Source: Available at http://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_library.html and http://faculty.chicagobooth.edu/lubos.pastor/research/liq_data_
1962_2012.txt.
Table 9.5 Private Equity Index

R2 F DW SC AIC DWH MKT SMB HML UMD LIQ


0.78 275.28 1.90 5.30 5.24 10.51*** 0.80 1.84
(2.20) (5.13)
0.79 145.99 1.78 5.31 5.21 6.58*** 1.20 1.68 0.52
(3.04) (6.79) (4.11)
0.80 101.85 1.96 5.32 5.19 2.79** 1.44 1.55 0.56 0.82
(2.39) (5.20) (1.46) (1.76)
0.83 95.29 1.85 5.20 5.02 2.65** 1.35 1.45 0.55 0.72 0.11
(2.69) (4.56) (2.23) (1.67) (0.85)
161

0.82 87.38 1.75 5.27 5.09 4.59*** 1.73 1.44 0.31 0.54 0.71
(4.01) (9.41) (1.32) (2.58) (5.65)
0.85 88.51 1.64 5.12 4.91 3.86*** 1.72 1.40 0.49 0.53 0.21 0.65
(4.38) (6.80) (2.22) (2.07) (3.06) (5.29)

Note: The alpha is expressed in percent. F and DW are the statistics for the Fisher test and the Durbin-Watson test, respectively. SC and AIC are Schwartz Criterion and
Akaike Information Criterion, respectively. DWH is a standard q2 where q is the number of adjustment variables to detect EIVs. The DWH test is a heteroskedasticity robust
test. All HME statistics are computed with White (1980) H0 heteroskedasticity-consistent covariance matrix estimators (HCCME). t-statistics are reported in brackets under
the estimates. MKT = Fama and French market portfolio factor, SMB = Fama and French size factor, HML = Fama and French book-to-market factor, UMD = Momentum
factor (Carhart, 1997), LIQ = Pastor and Stambaugh liquidity factor.
*, **, *** Significant at the 0.10, 0.05, and 0.01 levels, respectively.
Source: Available at http://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_library.html and http://faculty.chicagobooth.edu/lubos.pastor/research/liq_data_
1962_2012.txt.
162 h o w p r i v a t e e q u i t y w o r k s

Illiquidity and the Cost of Equity in Private Equity


PE returns are often stale and illiquid. To cope with this problem, the financial literature
suggests using either the Dimson (1979) method as initially applied for mutual funds or
lagged estimators. Another option is to use the Getmansky, Lo, and Makarov (2004) ad-
justment. As observed in the hedge fund industry (Okunev and White 2003; Getman-
sky et al. 2004; Cavenaile, Con, and Hbner 2011), the lack of liquidity, measured by
serial correlation, is a well acknowledged risk in the PE industry. Cavenaile et al. (2011)
illustrate that using smoothed hedge fund returns significantly underestimates risk. To
compensate for the biased estimates requires using estimates of true unsmoothed re-
turns. Table 9.6 reports the results of this adjustment for the five-factor model and for
each PE category.
Without adjusting for the stale valuation problem, the DWH statistic is statistically
significant for the GRO, BO, MEZ, and PE coefficients. TheR 2 s high for the four in-
dexes ranging from 0.64 for MEZ to 0.85 for PE. All alpha estimates are negative and
statistically significant ranging from 1.51 percent for GRO to 2.34 percent for BO.
The beta for the market premium is higher with this unsmoothed technique ranging
from 1.81 for GRO to 1.01 for MEZ. This illustrates the impact of the stale valuation
problem in computing the cost of capital of PE. SMB and HML are still statistically
significant and confirm the reported results in the previous section. UMD is statistically
significant for GRO (0.32) and for PE (0.26). The liquidity premium is significant
for BO (1.48), MEZ (1.45), and PE (0.89). This finding empirically confirms the re-
sults reported by Franzoni et al. (2012). As a conclusion, a liquidity premium should be
taken into account when computing PEs cost of capital.

Summary and Conclusions


This chapter provides a reappraisal of the cost of PE. It provides two major adjustments
to address issues in determining the cost of PE. First, implementing an unbiased estima-
tor developed by Dagenais and Dagenais (1997) corrects for EIVs in linear asset pricing
models. Second, using the adjustment approach initiated by Getmansky et al. (2004)
helps to compensate for the illiquidity and stale valuation issue acknowledged in the
literature on PE. Because the results show that OLS estimates are statistically biased as
reported by the DWH test, their use can lead to biased costs of capital.
Using the listed public funds and firms of the S&P Listed PE Index, evidence of
EIVs is apparent in evaluating PE. Therefore, DDHME provides unbiased and robust
estimates. The GLM adjustment seems relevant to generate unsmoothed returns and
unbiased estimates. These two adjustments may be used to compute unbiased and thus
improved cost of capital of PE.
Globally, the results report that various risk factors (MKT, SMB, HML, and LIQ)
are relevant in pricing the risk of PE. The liquidity factor is statistically significant and
should be used to generate a specific risk premium added to the computation of the cost
of equity. This finding is clearly reported for Buyouts, Mezzanines, and PE funds (as il-
lustrated by Tables 9.3 to Table 9.5). For growth PE funds and VC funds, the liquidity
Table 9.6 Illiquidity and the Cost of Equity of Private Equity: GLM Adjustment
R2 F DW SC AIC DWH MKT SMB HML UMD LIQ

GRO
0.82 72.08 2.31 5.92 5.70 2.53** 1.51 1.81 0.70 0.32 0.32 0.26
(2.79) (9.82) (4.79) (1.32) (4.27) (1.17)
BO
0.72 39.57 2.18 6.27 6.05 7.42*** 2.34 1.33 1.11 1.23 0.02 1.48
(2.65) (4.16) (3.40) (2.84) (0.13) (5.34)
MEZ
0.64 28.10 2.27 6.66 6.44 4.30*** 2.25 1.01 0.56 1.63 0.26 1.45
(1.60) (1.46) (1.12) (2.06) (1.55) (1.86)
PE
0.85 88.21 2.16 5.66 5.44 4.12*** 2.06 1.67 0.66 0.64 0.26 0.89
(4.41) (7.84) (2.80) (2.16) (3.36) (5.37)

Note: The alpha is expressed in percent. F and DW are the statistics for the Fisher test and the Durbin-Watson test, respectively. SC and AIC are Schwartz Criterion and
Akaike Information Criterion, respectively. DWH is a standard q where q is the number of adjustment variables to detect EIVs. The DWH test is a heteroskedasticity robust
2

test. All HME statistics are computed with White (1980) H0 heteroskedasticity-consistent covariance matrix estimators (HCCME). t-statistics are reported in brackets under
the estimates. MKT = Fama and French market portfolio factor; SMB = Fama and French size factor; HML = Fama and French book-to-market factor; UMD = Momentum
factor (Carhart, 1997); LIQ = Pastor and Stambaugh liquidity factor; GLM is the acronym for Getmansky, Lo, and Makarov.
*, **, *** Significant at the 0.10, 0.05, and 0.01 levels, respectively.
Source: Available at http://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_library.html; http://faculty.chicagobooth.edu/lubos.pastor/research/liq_data_1962_
2012.txt.
164 h o w p r i v a t e e q u i t y w o r k s

factor is not statistically significant at 0.10 level for the specific sample (Tables 9.2
and 9.6). As reported by the results, the momentum factor (UMD) could be seriously
considered, especially for growth PE funds and VC funds. These different risk factors
have a direct impact on the cost of capital of PE.

Discussion Questions
1. Discuss whether the cost of capital of PE differs from the standard paradigm estab-
lished by the modern finance theory.
2. Identify which asset pricing models should be used to value the cost of capital of PE.
3. Explain the main consequences of illiquidity or stale valuation on PE.
4. Discuss how adjusting for liquidity premiums improves the valuation process.

References
Bodson, Laurent, Alain Con, and Georges Hbner. 2010. Dynamic Hedge Fund Style Analysis
with Errors-in-Variables. Journal of Financial Research 33:3, 201221.
Bchner, Axel, and Rdiger Stucke. 2014. The Systematic Risk of Private Equity. Working Paper,
University of Passau and University of Oxford.
Carhart, Michael. 1997. On Persistence in Mutual Fund Performance. Journal of Finance 52:1,
5782.
Carmichael, Benot, and Alain Con. 2008. Asset Pricing Models with Errors-in-Variables. Journal
of Empirical Finance 15:4, 778788.
Cavenaile, Laurent, Alain Con, and Georges Hbner. 2011. The Impact of Illiquidity and Higher
Moments of Hedge Fund Returns on Their Risk-Adjusted Performance and Diversification
Potential. Journal of Alternative Investments 13:4, 929.
Cochrane, John H. 2005. The Risk and Return of Venture Capital. Journal of Financial Economics
75:1, 352.
Con, Alain, and Georges Hbner. 2009. Risk and Performance Estimation in Hedge Funds Revis-
ited: Evidence from Errors-in-Variables. Journal of Empirical Finance 16:1, 112125.
Con, Alain, and Franois-ric Racicot. 2007. Capital Asset Pricing Models Revisited: Evidence
from Errors in Variables. Economics Letters 95:3, 443450.
Dagenais, Marcel G., and Denyse L. Dagenais. 1997. Higher Moment Estimators for Linear Regres-
sion Models with Errors in the Variables. Journal of Econometrics 76:12, 193221.
Davidson, Russell, and James G. MacKinnon. 2004. Econometric Theory and Methods. New York:
Oxford University Press.
Dimson, Elroy. 1979. Risk Measurement When Shares Are Subject to Infrequent Trading. Journal
of Financial Economics 7:2, 197226.
Ewens, Michael, Charles M. Jones, and Matthew Rhodes-Kropf. 2013. The Price of Diversifiable
Risk in Venture Capital and Private Equity. Review of Financial Studies 26:8, 18541889.
Fama, Eugene F., and Kenneth R. French. 1993. Common Risk Factors in the Returns on Stocks
and Bonds. Journal of Financial Economics 33:1, 356.
Fama, Eugene F., and Kenneth R. French. 1997. Industry Costs of Equity. Journal of Financial Eco-
nomics 43:2, 153193.
Fleming, Grant. 2010. Institutional Investment in Private Equity: Motivations, Strategies, and Per-
formance. In Douglas Cumming, ed., Private Equity: Fund Types, Risks and Returns, and Regu-
lation, 929. Hoboken, NJ: John Wiley & Sons, Inc.
Franzoni, Francesco, Eric Nowak, and Ludovic Phalippou. 2012. Private Equity Performance and
Liquidity Risk. Journal of Finance 67:6, 23412373.
Cost of C apit al for P riv at e E qu it y 165

French, Kenneth R. 2014. Data: Risk Factors. Available at http://mba.tuck.dartmouth.edu/


pages/faculty/ken.french/data_library.html.
Getmansky, Mila, Andrew W. Lo, and Igor Makarov. 2004. An Econometric Model of Serial Corre-
lation and Illiquidity in Hedge Fund Returns. Journal of Financial Economics 74:3, 529609.
Gompers, Paul A., and Josh Lerner. 1997. Risk and Reward in Private Equity Investments: The
Challenge of Performance Assessment. Journal of Private Equity 1:2, 512.
Groh, Alexander Peter, and Oliver Gottschalg. 2011. The Effect of Leverage on the Cost of Capital
of US Buyouts. Journal of Banking and Finance 35:8, 20992110.
Harris, Bob, Tim Jenkinson, and Steve Kaplan. 2014. Performance of Private Equity Funds: What
Do We Know. Journal of Finance 69:5, 18511882 forthcoming.
Hausman, Jerry A. 1978. Specification Tests in Econometrics. Econometrica 46:6, 12511271.
Higson, Chris, and Rdiger Stucke. 2012. The Performance of Private Equity. Working Paper,
London Business School and University of Oxford.
Jensen, Michael C. 1968. The Performance of Mutual Funds in the Period 19451964. Journal of
Finance 23:2, 389413.
Jones, Charles M., and Matthew Rhodes-Kropf. 2003. The Price of Diversifiable Risk in Venture
Capital and Private Equity. Working Paper, Columbia University.
Kaplan, Steven N., and Richard S. Ruback. 1995. The Valuation of Cash Flow Forecasts: An Em-
pirical Analysis. Journal of Finance 50:4, 10591093.
Kaplan, Steven N., and Antoinette Schoar. 2005. Private Equity Performance: Returns, Persistence,
and Capital Flows. Journal of Finance 60:4, 17911823.
Lerner, Josh, Antoinette Schoar, and Wan Wongsunwai. 2007. Smart Institutions, Foolish Choices:
The Limited Partner Performance Puzzle. Journal of Finance 62:2, 731764.
Lintner, John. 1965. The Valuation of Risk Assets and the Selection of Risky Investments in Stock
Portfolios and Capital Budgets. Review of Economics and Statistics 47:1, 1337.
Ljungqvist, Alexander, and Matthew Richardson. 2003. The Cash Flow, Return and Risk Charac-
teristics of Private Equity. Working Paper No. w9454, National Bureau of Economic Research.
Markowitz, Harry. 1952. Portfolio Selection. Journal of Finance 7:1, 7791.
Metrick, Andrew. 2007. Venture Capital and the Finance of Innovation. Hoboken, NJ: John Wiley &
Sons, Inc.
Moskowitz, Tobias J., and Annette Vissing-Jorgensen. 2002. The Returns to Entrepreneurial Invest-
ment: A PE Premium Puzzle? American Economic Review 92:4, 745778.
Mossin, Jan. 1966. Equilibrium in a Capital Asset Market. Econometrica 34:4, 768783.
Okunev, John, and Derek White. 2003. Hedge Fund Risk Factors and Value at Risk of Credit Trad-
ing Strategies. Working Paper, University of New South Wales.
Pastor, Lubos. 2014. Data: Risk Factors. Available at http://faculty.chicagobooth.edu/lubos.
pastor/research/liq_data_ 1962_2012.txt.
Pastor, Lubos, and Robert F. Stambaugh. 2003. Liquidity Risk and Expected Stock Returns. Jour-
nal of Political Economy 111:3, 642685.
Pedersen, Niels, Sbastien Page, and Fei He. 2014. Asset Allocation: Risk Models for Alternative
Investments. Financial Analysts Journal 70:3, 3445.
Phalippou, Ludovic, and Oliver Gottschalg. 2009. The Performance of Private Equity Funds.
Review of Financial Studies 22:4, 17471776.
Phalippou, Ludovic, and Maurizio Zollo. 2005. What Drives Private Equity Fund Performance.
Working Paper, University of Amsterdam and INSEAD.
Sharpe, Williams F. 1964. Capital Asset Prices: A Theory of Market Equilibrium under Conditions
of Risk. Journal of Finance 19:3, 425442.
10
Liquidity Issues in Private Equity
Lessons Learned in the PostFinancial Crisis
ELIF AKBEN SELCUK
Instructor, Kadir Has University

AY S E D I L A R A A LT IO K Y I L M A Z
Assistant Professor, Bahcesehir University

Introduction
Liquidity is an important characteristic that distinguishes private equity (PE) from
public equity (Kleymenova, Talmor, and Vasvari 2012). The academic literature pio-
neered by Amihud and Mendelson (1986) contends that liquidity has a substantial
impact on asset prices. Using bid-ask spreads as a measure of stock-specific liquidity,
the authors document an illiquidity premium. Specifically, they find that a 1 percent
increase in the bid-ask spread increases annual expected returns by 2.4 percent. Fol-
lowing Amihud and Mendelson (1986), many empirical studies examine the impact of
asset-specific liquidity on the performance of stocks. Among others, Brennan and Sub-
rahmanyam (1996), Brennan, Chordia, and Subrahmanyam (1998), and Datar, Naik,
and Radcliffe (1998) document a positive relationship between expected stock returns
and alternative measures of liquidity.
More recent studies, including those of Pastor and Stambaugh (2003), Acharya and
Pedersen (2005), Liu (2006), and Sadka (2006), highlight the role of systemic liquidity
risk in explaining PE returns. These studies provide evidence that higher returns exist for
securities with greater sensitivity to aggregate liquidity in the market. This relationship is es-
pecially true for securities with high asset-specific liquidity as Acharya and Pedersen (2005)
describe by the term commonality-in-liquidity effect. Various authors uncover more
evidence on illiquidity premiums to bond markets (Chordia, Sarkar, and Subrahmanyam
2005; Beber, Brandt, and Kavajecz 2008; Li, Wang, Wu, and He 2009; Acharya, Amihud,
and Bharath 2013) and credit derivative markets (Longstaff, Mithal, and Neis 2005; Bon-
gaerts, De Jong, and Driessen 2010; Longstaff, Pan, Pedersen, and Singleton 2011).
Recently, attention turned to the liquidity of alternative asset classes such as hedge
funds and PE (Maier, Schaub, and Schmid 2011). When researchers document abnor-
mally high returns from PE, they commonly cite compensation for low liquidity as an

166
Liquidit y I s s u e s in P riv at e E qu it y 167

explanation. Ljungqvist and Richardson (2003) address this issue in their study analyz-
ing cash flow, risk, and return properties of PE. Using data from 1981 to 2001 from one
of the largest limited partners (LPs) in the United States, the authors document the ex-
istence of an excess return of 5 to 8 percent for PE firms compared to public firms. Over
a 10-year horizon, the authors calculate a risk-adjusted premium of about 24 percent
and attribute it to compensation for the illiquidity of PE investments.
In a similar study, Cumming, Fleming, and Schwienbacher (2005) analyze the re-
lationship between liquidity of exit markets and investments made by venture capital
(VC), a specific type of PE. Cumming et al. (2005, p. 82) find that venture capitalists
investment types depend on market liquidity and conclude that their findings are con-
sistent with the view that illiquidity is one reason venture capitalists require higher re-
turns on their investments. These results are also consistent with earlier studies on VC
such as those by Gompers and Lerner (1999, 2001). Similarly, Metrick (2007) reports
that returns should be adjusted since PE is largely exposed to liquidity risk. The author
uses an index of VC returns and estimates a time-series regression. Results show a 1 per-
cent yearly premium for liquidity risk.
PE liquidity risk began to receive renewed attention from scholars, practitioners, and
policymakers after the financial crisis of 20072008 that was initially characterized by a
shortage of liquidity in the system. Therefore, the crisis had a strong effect on an illiquid
asset such as PE (Robinson and Sensoy 2013). Before the crisis, when cheap credit was
available, many investors turned to PE as a profitable return source. Especially during the
golden years of PE from 2003 to 2007, liquidity in the system was high and PE firms dis-
tributed high amounts of cash, which LPs in turn reinvested into new PE funds. By mid-
2007, the PE industry had $1 trillion under management (Spangler 2013). However,
when the subprime mortgage bubble burst in 2007, the focus shifted from global excess
liquidity to market illiquidity and funding illiquidity (Gersl and Komarkova 2009).
The objective of this chapter is to discuss liquidity issues in PE within the context
of the financial crisis of 20072008, an important event for PE funds. The rest of the
chapter consists of three sections. The first section discusses the nature of liquidity for
PE investments and introduces two types of liquidity risk: market liquidity risk and
funding liquidity risk. The second section focuses on the liquidity pressures caused by
the financial crisis on PE investors and fund managers. This section also provides both
anecdotal and empirical evidence and discusses the lessons revealed through the finan-
cial crisis for PE investors. The third section provides a summary and conclusions.

Liquidity Risk in Private Equity


In financial economics, liquidity can refer to many different but interrelated concepts.
For example, it can refer to market liquidity (or asset liquidity) and funding liquidity (or
cash-flow liquidity). Simply put, market liquidity is the ease with which an asset can be
converted into an exchangeable medium such as cash. Funding liquidity is the capacity
for a trader to meet obligations and to raise more funds when needed (Gersl and Ko-
markova 2009). These two types of liquidity reinforce each other through a liquidity
spiral (Brunnermeier and Pedersen 2009). For PE investments, both types of liquidity
risk are present as discussed in the following subsections.
168 h o w p r i vat e e q u i t y w o r k s

MARKET LIQUIDITY RISK


For continuously traded financial assets, Kyle (1985) and Harris (2003) suggest four in-
terrelated dimensions of asset liquidity: width, depth, immediacy, and resilience. Width
refers to the bid-ask spread. Depth can be interpreted as the volume traded at given
prices. Immediacy refers to how rapidly a trade can be completed. Resilience refers to
the speed with which prices return to their previous levels following a large transaction.
Based on these four dimensions, Campbell, Lo, and Mackinlay (1997, p. 99) define li-
quidity as the ability to buy or sell significant quantities (width and depth) of a security
quickly (immediacy), anonymously, and with little price impact (resiliency). As Cum-
ming, Fleming, and Schwienbacher (2005) note, the standard definition of liquidity
does not apply to PE investments because PE is traded infrequently.
Market liquidity risk for PE investments refers to exit risk (Cumming et al. 2005).
That is, market liquidity risk is the risk of being forced to sell an investment at a discount
or to be locked up in an investment longer than desired. This risk is mainly due to PE
funds primarily being organized in the form of limited partnerships with finite lives,
typically 10 years with a possible extension of up to three years as defined in the partner-
ship agreement (Laitera 2012). The PE funds managers, called general partners (GPs),
invest the capital raised from investors (LPs) in selected companies. Once invested in a
limited partnership, capital remains locked up for long periods and withdrawing money
before the ex-ante contracted investment term is nearly impossible. Typically, partner-
ship agreements specify restrictions on transferability of ownership and LPs have no
right to execute the sale of a firm. Distributions are made only after the firms in the port-
folio go public or are sold by the fund manager. The proceeds are then divided among
the LPs and GPs in the form of equity or cash (Lerner and Schoar 2004; Timmermans
2009). PE is intended to be a long-term investment for buy-and-hold investors and most
of its returns originate from capital gains in exit transactions (Cumming et al. 2005).
Therefore, PE investors require higher returns partly to compensate for the risk of being
unable to effectively exit (Lerner 2002; Lerner and Schoar 2004, 2005).
According to a theory developed by Lerner and Schoar (2004), illiquidity is a dis-
tinguishing characteristic of PE investments. The authors model illiquidity as a choice
variable used by GPs to screen potential investors and attract those with long-term in-
vestment horizons. Lerner and Schoar justify their theory based on three observations.
First, PE investors in the United States have limited rights to manage the funds. Second,
most LPs prefer exit over voice, and the information they get about the performance of
a given PE fund only influences their decision to participate in future funds of the same
organization. Third, severe limits exist on the transferability of partnership interests,
which are already illiquid because of the large stakes each investor holds.
Limiting this theory is the relatively new concept called private equity secondary mar-
kets. The secondary market is the informal market where LPs can sell their stakes in a
PE fund or a portfolio of funds (Cotton 2012). However, investors are still exposed to
liquidity risk due to the small size and inefficient nature of secondary markets. Since PE
funds are less volatile than public equity, they are valuated less often and net asset values
(NAVs) may not reflect actual market conditions. Also, many factors such as confidence
in economic conditions, anticipated performance of portfolio companies, liquidity of
exit markets, liquidity of credit markets for refinancing, and the average track record of
Liquidit y I s s u e s in P riv at e E qu it y 169

GPs may affect value in the secondary market (Hege and Nuti 2011). As a result, inves-
tors in the secondary market may have to sell their PE interests at a substantial discount
compared to the NAV (Phalippou 2011).

FUNDING LIQUIDITY RISK


Funding liquidity can be explained in many ways. For instance, Drehmann and Niko-
laou (2013) describe funding liquidity as the ability to rapidly settle obligations while
Brunnermeier and Pedersen (2009) define funding liquidity as the ability to raise cash
at short notice through new loans or asset sales.
Funding liquidity risk is present in the PE industry due to its institutional features.
Most apparent is that PE commitments are not fully funded at the beginning. LPs make an
initial capital commitment to provide up to a certain amount of capital to the fund when
needed, rather than making the full investment on the first day. Upon launch, only a certain
portion of the capital is payable, and the balance is drawn over time as the GP performs
capital calls when the PE firm decides to invest in selected companies (Metrick and Yasuda
2010; Sinka 2012). Investors in PE (LPs) are obligated to meet these capital calls. Refusing
a capital call would have severe monetary and reputational results (Laitera 2012).
The unpredictable timing of capital calls over the life of the PE investment creates a
funding liquidity risk for the LPs. Also, these calls are usually accompanied by a short
notice period. For instance, PE investors may be asked to provide funds at times when
liquidity is tight. These capital calls can occur at times when getting cash involves a high
opportunity cost and high liquidity risk for the LPs (Robinson and Sensoy 2013).

The Global Financial Crisis and Private Equity


PE has been a major feature of financial markets long before the global financial crisis
(GFC). Until the late 1990s, the most common choice for institutional investors was
to pursue a balanced strategy with 60 percent equity and 40 percent fixed income al-
locations. Since markets were growing in the 1980s and 1990s, managers gave greater
weight to equity, allowing investors to increase beta risk while maintaining diversifica-
tion benefits.
The bursting of the technology bubble in 2000 and 2001 led many investors to search
for alternative strategies leading to the endowment model pioneered by Yale Univer-
sity (Vaillancourt 2012). The innovation of the endowment model was to allocate a
substantial portion of total assets in a portfolio to alternative investments, especially
hedge funds or PE. The endowment model is based on the premise that these alterna-
tive assets are illiquid and hence provide an illiquidity premium that is unavailable in
traditional asset classes. Also, the model assumes that alternative assets have low per-
formance correlations with public equity and fixed income securities allowing the man-
agers to control risk and provide diversification benefits (Swensen 2000). Due to the
popularity of the endowment model, the share of PE investments in the portfolios of
institutional investors had increased substantially in the years before the financial crisis
of 20072008 (Hege and Nuti 2011). Still, investors questioned the models assump-
tions when several endowments faced liquidity problems during this crisis.
170 h o w p r i vat e e q u i t y w o r k s

The financial crisis was the most severe economic crisis since the Great Depres-
sion and had several effects on PE funds. The industrys overall performance decreased
dramatically with the declining value of portfolio companies because of the credit
squeeze in the financial system. According to Friedman (2010), the one-year internal
rate of return (IRR) was equal to 27.6 percent at December 31, 2008. As a result, the
J-curve (i.e., negative returns in early years and gains in outlying years as portfolio
companies mature) historically depicting PE performance was reversed. As Figure 10.1
illustrates, 2005 and 2006 vintage funds lost their previous gains. With later improve-
ments in market conditions, the returns started to increase again, resulting in a pattern
called the W-curve.
The financial crisis also affected exit, investment, and financing opportunities in the
PE industry. The crisis placed liquidity pressures on both fund managers and PE in-
vestors (Sinka 2012). The following sections focus on the liquidity issues and lessons
learned from the crisis. Empirical investigations of the impact of the crisis on PE are
also discussed.

LIQUIDITY ISSUES DURING THE GLOBAL FINANCIAL CRISIS


The GFC had a major negative effect on the PE industry. Globally, fundraising declined
steeply. As Figure 10.2 shows, 1,435 PE vehicles raised $688 billion in 2008 globally but
948 companies raised only $320 billion in 2009.
The financial crisis affected both PE fund managers and PE investors in terms of
liquidity. Before discussing these effects, a basic understanding of the crisis is neces-
sary. The financial crisis started with credit deterioration in the U.S. subprime mortgage

15.0

10.0

5.0

0.0 Vintage 2004


0 1 2 3 4 5 6 Vintage 2005
5.0 Vintage 2006
Vintage 2007
10.0 Vintage 2008

15.0

20.0

25.0

Figure 10.1 Annual Median Net Internal Rates of Return by VintageThis


graph illustrates the decrease in PE performance due to the GFC and the resulting
W-curve pattern in IRR. The 2005 and 2006 vintage funds lost the gains made earlier,
with returns later rising again. Source: Friedman (2010).
Liquidit y I s s u e s in P riv at e E qu it y 171

1,600
1,435
1,400

1,200
1,027 1,046
1,000 948 951 932

800
688 Number of Funds Closed
600
467 Aggregate Capital Raised
382 ($bn)
400 320 331
295
200

0
2008 2009 2010 2011 2012 2013
Year of Final Close

Figure 10.2 Historical Private Equity Fundraising, 2008 to 2013 This graph
illustrates the amount of capital secured by fund managers between 2008 and 2013 as well
as the number of funds. In 2008, 1,435 PE vehicles raised $688 billion but in 2009, 948
funds raised only $320 billion. Source: Duong (2014).

market and hence was initially a credit, not a liquidity, issue. Yet, market liquidity and
funding liquidity played crucial roles in the developing crisis. Uncertainty about valuing
securities related to the mortgage market quickly increased and led to a decline in the
market liquidity of these instruments (Caruana and Kodres 2008). Moreover, financial
institutions that had large exposure to mortgage-backed securities saw a deterioration
of their balance sheets and faced severe potential losses. As a result, they tried to sell
their assets to expand their cash buffers. This action caused a further decline in asset
prices and created a downward spiral (Brunnermeier 2009; Brunnermeier and Peder-
sen 2009). To correct their balance sheets, financial institutions also decreased the loans
they made to borrowers among which were PE houses, which reduced their funding
liquidity.
These issues affected GPs in PE in different ways. One particular feature of PE invest-
ments is their high degree of leverage. This feature is especially true for buyout (BO)
funds (Axelson, Stromberg, and Weisbach 2009; Kaplan and Strmberg 2009). Since
PE firms usually borrow short or medium term, the loans need to be regularly rene-
gotiated. During the crisis, fund managers had difficulty refinancing their investments
because of the decrease in the liquidity that had affected the whole system including
providers of debt such as banks or other financial institutions. As a result, PE firms had
to liquidate their investments or accept higher costs of borrowing (Franzoni, Nowak,
and Phalippou 2012). For instance, the share of BOs decreased from 66 percent in 2008
to 57 percent in 2009 since BOs are the most leveraged transactions.
On the investor side, the financial crisis also affected many LPs, especially those
based on the endowment model with high exposure to alternative investments such as
hedge funds and PE. Many investors faced over allocation and liquidity problems due
to capital calls, decreased distributions, and suspended redemptions in other types of
assets (Cornelius 2011).
172 h o w p r i vat e e q u i t y w o r k s

The impact of the crisis on portfolio composition is commonly described by the


denominator effect faced by long-term investors. In general, these investors create
their target asset allocation as a percentage of their total assets under management. Due
to the decrease in the values of other asset classes such as public equity, the investors
became superficially over-allocated in alternative investments such as hedge funds or
PE (Spangler 2013). Many LPs sought relief from this imbalance and tried to decrease
their exposure to PE. A huge increase occurred in the number of LPs attempting to sell
their PE fund interests (Kleymenova et al. 2012). Effectively exiting was difficult be-
cause PE investments are illiquid long-term investments.
As a result, the best option for LPs was to sell their PE funds in the PE secondary
market. The high number of sellers led to a substantial decrease in prices in the second-
ary market and negatively affected the markets effective liquidity. The deal volume fell
to a level between $1 and $2 billion in the first half of 2009, temporarily causing the
PE secondary market to freeze at the beginning of the year. A large gap grew between
buyers and sellers price expectations. Adding to this problem was NAVs delayed re-
sponse in adjusting to decreases in the performance of underlying companies. At the be-
ginning of 2009, the bid-ask spread had increased substantially and reached 85 percent
of the bid levels (Hege and Nuti 2011). As previously mentioned, developing effective
exit strategies for PE investments is difficult.
Another characteristic of PE investments making them problematic for liquidity is
the exclusive right of the GPs to make capital calls asking investors for more money.
As Figure 10.3 shows, the crisis caused a decline in exit activity as illustrated by lower
distributions between 2008 and 2010. During the crisis, the GPs facing liquidity issues
continued to exercise their right to place capital calls but at slightly lower amounts.
These capital calls placed further liquidity pressures on the LPs. To be able to fulfill
these calls and avoid severe penalties that would result from default, PE investors had to
sell other assets in their portfolio when market values were low. The result was a deterio-
rating liquidity profile of investors portfolios because LPs who were able to sell more

500 476 470


450
404 409 392
400 366 373
345
350
288 292 304 Capital Called
300 253
250 210 ($bn)
201 209
200 165 174
188 Capital Distributed
161 143
146
150 115 ($bn)
120
100 88 91 78
59 57
50
0
3
00

01

02

03

04

05

06

07

08

09

10

11

12
01
20

20

20

20

20

20

20

20

20

20

20

20

20
n2
ec

ec

ec

ec

ec

ec

ec

ec

ec

ec

ec

ec

ec
Ju
D

Figure 10.3 All Private Equity Annual Amount Called Up and Distributed,
2003 to 2013 This graph illustrates the decline in distributions between 2008 and 2010
due to the GFC. During that time, GPs continued to call capital but at slightly lower
amounts. This pattern caused liquidity pressures on PE investors. Source: Duong (2014).
Liquidit y I s s u e s in P riv at e E qu it y 173

liquid assets reallocated these funds to PE investments. The most dramatic impact of
liquidity risk was on institutions with spending targets that consistently used up liquid-
ity. More capital calls on top of these spending targets led to major problems for the LPs
(Vaillancourt 2012).
Among institutional investors that tried to sell their PE investments, U.S. univer-
sity endowments were particularly important because the share of PE in their portfolio
was sometimes as high as 20 percent of total assets under management. An example
attracting considerable attention was Harvard Management Corporation (HMC) that
managed Harvard Universitys endowment. As explained in a report by the European
Private Equity and Venture Capital Association (EVCA 2013), 13 percent of HMCs
total portfolio consisted of PE in mid-2008. Before the financial crisis, Harvard Univer-
sity derived one-third of its total operating income from contributions by HMC. The
university relied on HMC even in planning decisions such as hiring or expansions based
on the assumption that HMC would produce constant cash flows through harvested PE
investments. During the GFC, this assumption was proven wrong when exit markets
shut down and distributions plummeted. As a result, HMC had to liquidate some public
equity and fixed income investments in a declining market to fulfill its commitments to
PE funds. HMC also decided to sell some of its PE investments in the secondary market
and issue debt in capital markets to overcome the liquidity shock (EVCA 2013). This
unanticipated liquidity risk forced Harvard University to make both budget and per-
sonnel cuts. Several administrators lost their jobs or were forced into early retirement
(Vaillancourt 2012).
Another example is the California Public Employees Retirement Systems (CalP-
ERS), which is the largest U.S. pension fund and one of the most important PE players
in the world. During the GFC, CalPERS faced severe liquidity pressures due to its ob-
ligations from PE deals and had to sell equity in a declining market. The share of equity
in its portfolio declined from 60 percent in 2007 to 44 percent in 2009 and the fund lost
$70 billion during the crisis (Ang and Kjaer 2011).
Recent studies in asset pricing literature suggest that investors prefer to receive cash
distributions from their assets in times of low liquidity rather than in times of high li-
quidity. The financial crisis was a clear illustration of this argument. Large PE investors
such as Harvard or CalPERS would have preferred to receive distributions from their
PE holdings in 2008 rather than in 2006 (Franzoni, Nowak, and Phalippou 2009). Be-
sides showing the importance of liquidity, these examples reveal additional lessons for
investors discussed in the next subsection.

LESSONS LEARNED FROM THE GLOBAL FINANCIAL CRISIS


The GFC brought into question the validity of the endowment model. Under this
model, PE investors usually divide their funds to PE on the basis of a mean-variance
framework such as the capital asset pricing model (CAPM). This model assumes in-
vestors can diversify away the unsystematic risk inherent in each investment as long
as assets are not perfectly positively correlated. Returns of different asset classes can
therefore be explained by their systematic risk or beta risk (Cornelius 2011). Yet, the
crisis revealed that some of the restrictive assumptions of the CAPM do not apply to
PE investments.
174 h o w p r i vat e e q u i t y w o r k s

First, no market prices exist for true PE. Due to the lack of a market trading platform,
managers determine their valuation. Investors can get price information by using prox-
ies such as recently introduced indexes or public stock offerings made by the GPs. These
proxies tend to perform poorly during a crisis (Vaillancourt 2012). Second, the CAPM
assumes that market risk is static. The GFC revealed that risk in the whole system may
increase and the correlations among asset classes may increase (Cornelius 2011). With
many investors trying to deleverage their portfolios during short periods in 2007, 2008,
and 2009, the correlation between PE and other asset classes was much higher than
previously assumed in the endowment model. As a result, institutions following the
endowment model experienced liquidity constraints (Vaillancourt 2012). The high cor-
relations between PE and other assets during a deleveraging event such as the GFC
brought the endowment model into question. Additionally, when applying more realis-
tic correlation assumptions, the optimal allocation to PE is lower than predicted by the
endowment model. The GFC shows that PE and alternative asset classes do not provide
enough risk diversification to justify such high allocations.
In the aftermath of the GFC, investors realized that liquidity has a cost. They also
need to take liquidity risk into consideration when evaluating the optimal allocation in
PE or other asset classes. Liquidity crises have occurred since the emergence of modern
financial markets. Yet, investors had a sense of security that the system would keep li-
quidity crises from becoming systemic due to its innate features and actions by regula-
tory authorities (Shafer 2013). This false sense of security magnified the impact of the
crisis.
Another lesson learned from the GFC is that investors should keep both illiquid and
complementary liquid investments in their portfolios to avoid cash-flow distress and
have flexibility in response to rising systemic risk (Cornelius 2011). Although LPs rec-
ognize that long-term strategies that lock up capital for extended periods are critical to
increasing the PE investment returns, liquidity has also become a key concern. Today,
investors need to stage their investments in PE so the supply of cash flow from all in-
vestments aligns with the need for cash (Shafer 2013).
Investors are also becoming more selective and risk averse, looking for more estab-
lished PE fund managers with extensive track records. This leads to a concentration
in the industry because established fund managers by nature raise larger funds. As
Figure10.2 shows, the amount of capital raised in 2013 was at its highest level since
2009. Yet, the number of funds raising this capital actually decreased from previous
years, suggesting an increase in the average size of PE funds and increased activity from
larger fund managers (Duong 2014).
The GFC also highlights the need for more robust and effective liquidity risk man-
agement tools and procedures that consider the specific characteristics of PE. Along
with investor expectations, regulatory expectations have also increased. Accordingly,
regulatory authorities imposed new rules. The major legal rules affecting PE are the
DoddFrank Wall Street Reform and Consumer Protection Act (Dodd-Frank Act),
which was signed into law in the United States on July 21, 2010, and the Alternative
Investments Fund Managers Directive (AIFM Directive) in the European Union, which
took effect on July 22, 2013.
Broadly, the objective of the Dodd-Frank Act is to control systemic risk, which is the
risk that the entire financial system might become unstable as a result of a localized
Liquidit y I s s u e s in P riv at e E qu it y 175

economic shock due to contagion effects. As such, if a PE fund is considered to be sys-


temically important, its manager will be subject to heightened standards including li-
quidity and leverage requirements. Other standards include overall risk management,
enhanced risk-based capital, concentration limits, and early remediation requirements
(Laitera 2012).
The AIFM Directive regulates managers of alternative investment funds, essentially
hedge funds and PE. Liquidity risk is an important area of focus within the directive,
reflecting the increase in liquidity risk following redemption requests during the crisis.
According to the AIFM Directive, managers are required to adopt appropriate liquidity
management procedures for each alternative investment fund, except for unleveraged
closed-end funds. As explained in a report by Ernst & Young (2013), they should ensure
that the liquidity profiles of the investments within the fund comply with the obliga-
tions and that investment strategy, liquidity profile, and redemption policy are consist-
ent. Managers are also required to conduct stress testing under normal and exceptional
liquidity conditions. Also, they should be able to show authorities that proper liquidity
management systems are in place.

EMPIRICAL EVIDENCE
As the anecdotal evidence discussed in the previous section shows, the GFC had
a major impact on the PE industry. Nonetheless, the number of academic studies
systematically examining PE during the crisis is limited. This gap is largely due to
the difficulty of getting recent and reliable data on PE because this asset class is not
subject to disclosure requirements (Robinson and Sensoy 2013). Thus, any valuation
and cash flow data are based on voluntary reporting by GPs and LPs (Kaplan and
Schoar 2005).
Among these studies, Franzoni et al. (2012) use data from the pre-crisis period to ex-
amine the impact of the crisis on PE. The authors investigate whether aggregate market
liquidity affects PE performance. Using data from 7,198 BO investments between 1975
and 2006, they apply Pastor and Stambaughs (2003) four-factor model, which includes
a liquidity risk factor among others. The Pastor-Stambaugh liquidity measure is con-
structed for each stock by estimating the return reversal effect associated with a given
volume. These liquidity estimates are then aggregated to measure market liquidity and
liquidity betas are derived following the methodology used to estimate the Fama and
French (1992) three-factor model.
Franzoni et al. (2012) find that PE returns depend on their exposure to the Pastor-
Stambaugh liquidity risk factor and report a significant liquidity beta of 0.64. This trans-
lates into an annual liquidity premium of 3 percent. This premium is an important part
of the cost of capital and explains a major portion of excess returns in PE. These results
suggest that liquidity risk is present for PE investments and the contribution of PE to
diversifying investors portfolios is lower than assumed by earlier studies. The authors
further hypothesize that the relationship between aggregate market liquidity and PE
returns occurs through funding liquidity. At times of low liquidity such as the GFC,
GPs may have difficulty refinancing their highly leveraged investments and may have
to accept higher borrowing rates, resulting in lower returns for PE investors. In other
words, PE returns depend on the availability of capital to financial institutions (funding
176 h o w p r i vat e e q u i t y w o r k s

liquidity), which in turn depends on overall market liquidity. The authors also note that
if PE investors have larger losses during crises, these losses will happen at a slow pace,
lengthening the duration of the crisis.
In a related study, Laitera (2012) uses data on 119 funds raised between 1990 and
2011, provided by two LPs who operated independently of each other. Related to per-
formance, the author finds that PE outperformed public equity in the 2000s but the
relationship reversed during the crisis. Laitera suggests that this reversal might be due
to higher risk of PE. The evidence also shows that PE funds scaled down their invest-
ment activity during the GFC suggesting a possible lack of financing or risk aversion of
the GPs. Finally, a decrease in exit activity occurred during the GFC. Laitera explains
this finding by the following two observations. First, merger and acquisition (M&A)
as well as initial public offering (IPO) activity, which are the main exit avenues, slowed
down during the GFC. Second, some GPs postponed selling their investments due to
depressed market prices. This rationing hypothesis is also consistent with the findings of
Kaplan and Schoar (2005), who report that exits during an economic downturn result
in lower returns.
Robinson and Sensoy (2013) use a proprietary database of 837 VC and BO
funds from 1984 to 2010. The dataset includes the quarterly cash flows between
PE funds and investors, resulting in nearly 35,000 fund-quarter observations. The
authors calculate Kaplan and Schoars (2005) public market equivalents (PMEs),
which are defined as the ratio of the present value of distributions from the PE fund
to the present value of capital calls, both discounted at the realized market return.
PMEs greater than 1 indicate that the fund outperformed the market index. Results
show that PE funds, especially BO funds, outperform the S&P 500 index on a net-
of-fee basis by 15 percent. The authors also document that returns on PE invest-
ments are correlated with changes in broad market conditions. Finally, Robinson
and Sensoy analyze the liquidity properties of PE cash flows and their behavior
during the GFC. Outside the GFC, their evidence shows that PE is a source of li-
quidity when market conditions are good and a modest liquidity sinks when market
valuations are low. During the GFC and ensuing recession, a large increase in unex-
plained capital calls and a large decrease in distributions occurred. Robinson and
Sensoy suggest that the GFC resulted in a greater abnormal liquidity demand by
PE funds.

Summary and Conclusions


Liquidity is an important characteristic that distinguishes PE from public equity. PE
investments are exposed to both market liquidity risk and funding liquidity risk. Market
liquidity risk for PE refers to exit risk. Since PE funds are mostly organized in the form
of limited partnerships with finite lives of typically 10 years, they are difficult to effec-
tively exit. Also, the PE secondary market is small and investors usually have to sell their
assets at substantial discounts. On the other hand, funding liquidity risk stems from
the ability of GPs to make capital calls and the obligation of the LPs to fulfill them. The
unpredictable timing of capital calls over the life of the PE investment creates funding
liquidity risk for the LPs.
Liquidit y I s s u e s in P riv at e E qu it y 177

PE liquidity risk started to receive renewed attention after the financial crisis of
20072008, which was characterized by a shortage of liquidity in the financial system.
The crisis resulted in a large increase in the LPs requests to liquidate their PE interests.
Effective exit strategies were difficult to achieve because PE investments were illiquid
and long-term. As a result, the best option for LPs was to sell their PE holdings in the
PE secondary market. This high number of sellers led to a large decrease in prices in the
secondary market and effective liquidity in the market deteriorated. A large decline oc-
curred in deal volume along with a large increase in bid-ask spreads.
Besides the difficulty in developing effective exit strategies, PE investments gave GPs
the right to place additional capital calls. In the depths of the crisis, many funds faced
problems refinancing their investments and continued to place capital calls. These cap-
ital calls, on top of reduced distributions, deprived PE investors of liquidity when they
most needed it. Many LPs had to sell more liquid assets in their portfolios to fulfill these
capital calls and to avoid the penalties that can result from default.
The liquidity problems during the GFC revealed the risks of the endowment model
of investing, which institutional investors extensively used before the crisis. Specifically,
the GFC highlighted the potential dangers of a large allocation to illiquid asset classes
such as hedge funds and PE and prompted institutional investors, including pension
funds and university endowments, to evaluate their low liquidity/high-return portfo-
lios. In the aftermath of the GFC, investors need to consider liquidity risk when evaluat-
ing the optimal allocation to PE. The GFC also highlights the need for more robust and
effective liquidity risk management procedures that consider the specific characteristics
of PE. Consequently, new rules and regulations have been imposed. Liquidity risk is an
important area of focus within the Dodd-Frank Act and the AIFM Directive.
In summary, the GFC had a major impact on the PE industry. Recalling liquidity
problems during the crisis, investor and regulatory expectations increased. Investors are
becoming more selective and risk averse while looking for more established PE fund
managers with extensive track records.

Discussion Questions
1. Identify and discuss the two types of liquidity risk investors face in PE.
2. Explain the term denominator effect for PE and the GFC and discuss its impact on
the PE secondary market.
3. Describe the impact of the GFC on PE GPs in terms of liquidity.
4. Explain the avenues through which PE LPs faced liquidity problems during the
GFC.
5. Explain how the DoddFrank Act and AIFM Directive relate to PE liquidity risk.

References
Acharya, Viral V., Yakov Amihud, and Sreedhar T. Bharath. 2013. Liquidity Risk of Corporate Bond
Returns: A Conditional Approach. Journal of Financial Economics 110:2, 358386.
Acharya, Viral V., and Lasse H. Pedersen. 2005. Asset Pricing with Liquidity Risk. Journal of Finan-
cial Economics 77:2, 375410.
178 h o w p r i vat e e q u i t y w o r k s

Amihud, Yakov, and Haim Mendelson. 1986. Asset Pricing and the Bid-Ask Spread. Journal of Fi-
nancial Economics 17:2, 223249.
Ang, Andrew, and Knut N. Kjaer. 2011. Investing for the Long Run. Working Paper, Columbia
University. Available at http://ssrn.com/abstract=1958258 or http://dx.doi.org/10.2139/
ssrn.1958258.
Axelson, Ulf, Per Strmberg, and Michael Weisbach. 2009. Why Are Buyouts Levered? The Finan-
cial Structure of Private Equity Funds. Journal of Finance 64:4, 15491582.
Beber, Alessandro, Michael W. Brandt, and Kenneth A. Kavajecz. 2008. Flight to Quality or Flight
to Liquidity? Evidence from the Euro-Area Bond Market. Review of Financial Studies 22:3,
925957.
Bongaerts, Dion, Frank De Jong, and Joost Driessen. 2010. Derivative Pricing with Liquidity
Risk: Theory and Evidence from the Credit Default Swap Market. Journal of Finance 66:1,
203240.
Brennan, Michael J., Tarun Chordia, and Avanidhar Subrahmanyam. 1998. Alternative Factor Spec-
ifications, Security Characteristics, and the Cross-section of Expected Stock Returns. Journal
of Financial Economics 49:3, 345373.
Brennan, Michael J., and Avanidhar Subrahmanyam. 1996. Market Microstructure and Asset Pric-
ing: On the Compensation for Illiquidity in Stock Returns. Journal of Financial Economics
41:3, 441464.
Brunnermeier, Markus K. 2009. Deciphering the Liquidity and Credit Crunch 20072008. Jour-
nal of Economic Perspectives 23:1, 77100.
Brunnermeier, Markus K., and Lasse H. Pedersen. 2009. Market Liquidity and Funding Liquidity.
Review of Financial Studies 22:6, 22012238.
Campbell, John Y., Andrew W. Lo, and A. Craig Mackinlay. 1997. The Econometrics of Financial Mar-
kets. Princeton, NJ: Princeton University Press.
Caruana, Jaime, and Laura Kodres. 2008. Liquidity in Global Markets. Financial Stability Review,
Special Issue on Liquidity 11:2, 6574.
Chordia, Tarun, Asani Sarkar, and Avanidhar Subrahmanyam. 2005. An Empirical Analysis of
Stock and Bond Market Liquidity. Review of Financial Studies 18:1, 85129.
Cornelius, Peter. 2011. The Varied Approaches to Risk in Private Equity. Quarterly Review 7:1, 37.
Cotton, Ryan. 2012. The Benefits of Secondary Funds in a Private Equity Portfolio. Available at
https://www.harrismycfo.com/pdf/secondary-funds-benefit.pdf.
Cumming, Douglas J., Grant Fleming, and Armin Schwienbacher. 2005. Liquidity Risk and Ven-
ture Capital Finance. Financial Management 34:4, 77105.
Datar, Vinay T., Narayan Y. Naik, and Robert Radcliffe. 1998. Liquidity and Stock Returns: An Al-
ternative Test. Journal of Financial Markets 1:2, 203219.
Drehmann, Mathias, and Kleopatra Nikolaou. 2013. Funding Liquidity Risk: Definition and Mea-
surement. Journal of Banking and Finance 37:7, 21732182.
Ernst & Young. 2013. Risk and Liquidity Management for Private Equity and Real Estate Funds.
Available at http://www.ey.com/Publication/vwLUAssets/Presentation_AIF-Club_17-oct-
2013/$FILE/Presentation_AIFC-Risk-and-Liquidity.pdf.
European Private Equity and Venture Capital Association. 2013. Risk Measurement Guidelines.
Available at http://www.evca.eu/uploadedfiles/evca_risk_measurement_guidelines_january_
2013.pdf.
Fama, Eugene F., and Kenneth R. French. 1992. The Cross-Section of Expected Stock Returns.
Journal of Finance, 47:2, 427465.
Franzoni, Francesco, Eric Nowak, and Ludovic Phalippou. 2009. Private Equity and Liquidity
Risk. Working Paper, Goethe University. Available at http://www.wiwi.uni-frankfurt.de/
professoren/schlag/dgf2009/contribution232.pdf.
Franzoni, Francesco, Eric Nowak, and Ludovic Phalippou. 2012. Private Equity Performance and
Liquidity Risk. Journal of Finance 67:6, 23412373.
Friedman, Tim. 2010. Private Equity Performance: The W-Curve. Preqin Private Equity Spotlight
6:7, 67. Available at https://www.preqin.com/docs/newsletters/PE/Preqin_Private_
Equity_Spotlight_July_2010.pdf.
Liquidit y I s s u e s in P riv at e E qu it y 179

Gersl, Adam, and Zlatuse Komarkova. 2009. Liquidity Risk and Banks Bidding Behavior: Evi-
dence from the Global Financial Crisis. Finance a Uver-Czech Journal of Economics and Finance
59:6, 577592.
Gompers, Paul A., and Josh Lerner. 1999. The Venture Capital Cycle. Cambridge, MA: MIT Press.
Gompers, Paul A., and Josh Lerner. 2001. The Money of Invention: How Venture Capital Creates New
Wealth. Cambridge, MA: Harvard Business School Press.
Harris, Larry. 2003. Trading and Exchanges. New York: Oxford University Press.
Hege, Ulrich, and Allessandro Nuti. 2011. The Private Equity Secondaries Market during the Fi-
nancial Crisis and the Valuation Gap. Journal of Private Equity 14:3, 4254.
Kaplan, Steven N., and Antoinette Schoar. 2005. Private Equity Returns: Persistence and Capital
Flows. Journal of Finance 60:4, 17911823.
Kaplan, Steven N., and Per Strmberg. 2009. Leveraged Buyouts and Private Equity. Journal of Ec-
onomic Perspectives 23:1, 121146.
Kleymenova, Anya, Eli Talmor, and Florin P. Vasvari. 2012. Liquidity in the Secondaries Private
Equity Market. Working Paper, Coller Institute of Private Equity, London Business School.
Available at http://www.collerinstitute.com/research/paper/235.
Kyle, Albert S. 1985. Continuous Auctions and Insider Trading Source. Econometrica 53:6,
13151335.
Laitera, Mikko. 2012. Private Equity Funds and the Financial Crisis. Saarbrucken, Germany: LAP
LAMBERT Academic Publishing GmbH & Co. KG.
Lerner, Josh. 2002. Boom and Bust in the Venture Capital Industry and the Impact on Innovation.
Federal Reserve Bank of Atlanta Economic Review 87:4, 2539.
Lerner, Josh, and Antoinette Schoar. 2004. The Illiquidity Puzzle: Theory and Evidence from Pri-
vate Equity. Journal of Financial Economics 72:1, 340.
Lerner, Josh, and Antoinette Schoar. 2005. Does Legal Enforcement Affect Financial Transactions?
The Contractual Channel in Private Equity. Quarterly Journal of Economics 120:1, 223246.
Li, Haitao, Junbo Wang, Chunchi Wu, and Yan He. 2009. Are Liquidity and Information Risks
Priced in the Treasury Bond Market? Journal of Finance 64:1, 467503.
Liu, Weimin. 2006. A Liquidity-Augmented Capital Asset Pricing Model. Journal of Financial Eco-
nomics 82:3, 631671.
Ljungqvist, Alexander, and Matthew P. Richardson. 2003. The Cash Flow, Return and Risk Char-
acteristics of Private Equity. Working Paper No. 03-001, New York University. Available at
http://ssrn.com/abstract=369600 or http://dx.doi.org/10.2139/ssrn.369600.
Longstaff, Francis A., Sanjay Mithal, and Eric Neis. 2005. Corporate Yield Spreads: Default or
Liquidity? New Evidence from the Credit Default Swap Market. Journal of Finance 60:5,
22132253.
Longstaff, Francis A., Jun Pan, Lasse H. Pedersen, and Kenneth J. Singleton. 2011. How Sovereign
is Sovereign Credit Risk? American Economic Journal: Macroeconomics 3:2, 75103.
Maier, Tobias, Nic Schaub, and Markus Schmid. 2011. Hedge Fund Liquidity and Performance:
Evidence from the Financial Crisis. Working Paper, University of Mannheim. Available at
http://efmaefm.org/0efmsymposum/germany2012/papers/030.pdf.
Metrick, Andrew. 2007. Venture Capital and the Finance of Innovation. Hoboken, NJ: John Wiley
& Sons.
Metrick, Andrew, and Ayako Yasuda. 2010. The Economics of Private Equity Funds. Review of Fi-
nancial Studies 23:6, 23032341.
Duong, Jessica.2014. Cautious Optimism: Outlook for Private Equity in the Year Ahead. Preqin
Private Equity Spotlight 10:2, 35. Available at https://www.preqin.com/docs/newsletters/
pe/Preqin_PESL_Feb_14_Cautious_Optimism.pdf.
Pastor, Lubos, and Robert F. Stambaugh. 2003. Liquidity Risk and Expected Stock Returns. Jour-
nal of Political Economy 111:3, 642685.
Phalippou, Ludovic. 2011. An Evaluation of the Potential for GPFG to Achieve above Average Re-
turns from Investments in Private Equity and Recommendations Regarding Benchmarking.
Working Paper, University of Oxford. Available at http://ssrn.com/abstract=1807569 or
http://dx.doi.org/10.2139/ssrn.1807569.
180 h o w p r i vat e e q u i t y w o r k s

Robinson, David T., and Berk A. Sensoy. 2013. Cyclicality, Performance Measurement, and Cash
Flow Liquidity in Private Equity. Working Paper No. 2010-2021, Charles A. Dice Center;
Working Paper No. 2010-2003-021, Fisher College of Business. Available at http://ssrn.com/
abstract=1731603 or http://dx.doi.org/10.2139/ssrn.1731603.
Sadka, Ronnie. 2006. Momentum and Post-Earnings-Announcement Drift Anomalies: The Role
of Liquidity Risk. Journal of Financial Economics 80:2, 309349.
Shafer, Jeffrey. 2013. Five Years Later: Lessons from the Financial Crisis. Working Paper, McGraw
Hill Financial Global Institute. Available at http://www.mhfigi.com/wp-content/uploads/
2013/09/Five-Years-Later-Paper-by-Jeff-Shafer1.pdf.
Sinka, Michael. 2012. Regulation of Hedge Funds and Private Equity: Lessons from the Global Financial
Crisis. Saarbrucken, Germany: LAP LAMBERT Academic Publishing GmbH & Co. KG.
Spangler, Timothy. 2013. One Step Ahead: Private Equity and Hedge Funds after the Global Financial
Crisis. London: Oneworld Publications.
Swensen, David. 2000. Pioneering Portfolio Management: An Unconventional Approach to Institutional
Investment. New York: Free Press.
Timmermans, Xavier. 2009. Investing in Illiquid Assets. Risk and Rewards 53:2, 1116.
Vaillancourt, Jason R. 2012. Reducing Liquidity Risk in Plan Management. Available at https://
www.putnam.com/literature/pdf/su827.pdf.
11
Private Equity Portfolio Management
Challenges, Approaches, and Implementation
THOMAS MEYER
Director, LDS Partners

TOM WEIDIG
Risk Modelling Expert, Commissariat Aux Assurances

Introduction
Designing and implementing a private equity (PE) portfolio is often described as more
an art than a science. PE investments do not fit naturally into the standard risk-return
framework underlying the modern portfolio theory (MPT), which is anchored in the
efficient market hypothesis (EMH). MPT relies on efficient markets in which many
participants continuously price assets and investors can buy and sell at reasonable costs,
within a reasonable amount of time. Hence, this chapter examines these challenges for
portfolio management by reviewing return, risk, time horizon, taxes and fees, liquidity,
legal, and unique circumstances (RRTTLLU) framework, which represents a list of the
crucial factors when constructing an investment portfolio.
One major aspect of portfolio management is how to combine several asset classes.
In the environment of extreme uncertainty that characterizes PE, the search for opti-
mal solutions is futile and simpler methods work better than sophisticated ones. Market
practices heavily rely on heuristics, such as structuring portfolios into a core and a
satellite. This technique violates MPT dogmas, but is justified in the wider context
of behavioral finance. The adaptive market hypothesis (AMH) also goes beyond the
EMH. While the AMH is rarely discussed in PE, the AMH provides a satisfying frame-
work because it aligns the core-satellite approach with the EMH. Hence, these different
frameworks are discussed in this chapter.
Designing a successful PE portfolio is often about finding a balance between di-
versification and concentration and discovering how to assure a high degree of inde-
pendence among various positions held. This chapter describes tools to implement
portfolios, such as top-down versus bottom-up, as well as the limits to active portfo-
lio management. It also provides an argument for the growing importance of portfolio
management in PE.

181
182 h o w p r i vat e e q u i t y w o r k s

W H AT I S P R I VAT E E Q U I T Y ?
Private equity refers to an institutionalized way of owning a share of a company that is
not registered and not publicly traded on an exchange market. Institutional investors
typically focus on organized PE market investing either directly or through funds as un-
quoted vehicles. Understanding PE investing requires clarification on the nature of this
asset class, in particular for a distinction between privately held and private equity
as Table 11.1 shows.
Privately held companies broadly have the same intrinsic risk characteristics as their
public market peers (Cornelius 2011); therefore, traditional portfolio management
techniques can be used. Valuing these assets is based on periodic, subjective apprais-
als. While quotes for specific privately held assets are unavailable, prices observed in
financial markets for comparable assets are assumed to be representative and a suitable
input for a quantitative portfolio model. However, some adjustments are needed. For
instance, the illiquidity of PE is modeled by appropriate discount rates as investors re-
quire a higher rate of return for assets that cannot be easily sold. Due to the so-called ap-
praisal value effect, arising from the fact that an assets value is appraised only periodically
by an appraiser and infrequently by the market, time series valuations tend to understate
volatility. This effect, which is also called stale pricing, comes with the time-lag effect.
For example, a traded price might start at 100, fluctuate considerably, and end up at 100
again at the end of the month. Due to constant trading, this price shows its underlying
volatility unlike an appraised price with an unchanged appraisal value of 100 that hap-
pens once at the beginning and once at the end of the month.
To compensate for stale pricing, the derived variance can be adjusted by re-adding
volatility through factoring in public market returns such as the methodologies pro-
posed by Geltner, MacGregor, and Schwann (2003) and Getmansky, Lo, and Makarov
(2004). Many practitioners find these assumptions acceptable for their investment ap-
proach and thus treat PE as if it were embedded into a wider allocation of equity.

Table 11.1 Distinction between Privately Held and Private Equity

Privately Held Private Equity


Typically buy-to-keep Buy-to-sell
Direct investment Investment typically through limited
partnership funds and co-investments
Valuations based on comparables to Cash-flow asset
public market peers as proxy for prices
Risk Uncertainty (i.e., non-quantifiable risk)
Traditional portfolio management Heuristics and cash-flow based approaches
techniques

Note: This table indicates differences between privately held assets and PE. Privately held assets are
essentially identical to PE equity in terms of their role in a companys capital structure, except for their
substantially higher degree of illiquidity. PE assets have unique traits that go beyond a simple definition
of being non-quoted. PE investments are only held temporarily where during an intervention phase
and a value creation plan is implemented toward a specific three- to five-year goal.
Pr ivate E qu it y Port fol io M an ag e m e n t 183

P R I VAT E E Q U I T Y I N T E R V E N T I O N
PE assets have unique traits that go beyond a simple definition of being non-quoted.
In its core, PE is a strategy to search for arbitrage opportunities in under-researched or
overlooked niches in which information is proprietary and little competition exists. The
excess profits available to PE investors can be seen as a function of these opportuni-
ties for extreme arbitrage. When defining PE, investments are only held temporarily,
with a pre-planned or contractually fixed maximum term with intentions to profitable
divestment.
PE investing follows a buy-to-sell modus operandi in which during an intervention
phase portfolio companies are shielded from adverse market influences. During this in-
tervention phase, market value and market risk are not fully applicable, which requires
PE investors to refine how they perceive risk (Cornelius, Diller, Guennoc, and Meyer
2013). This framework is opposed to the modus operandi of many strategic buy-to-
hold (also called buy-to-keep). Some institutional investors such as sovereign wealth
funds, pension funds, and corporations follow the PE modus operandi and implement
a value creation plan. However, here after the intervention phase instead of a sell-off
follows an indefinite holding phase in which investors become rather passive and the
portfolio companies are privately held and again operate in the same market environ-
ment as publicly quoted assets.

I N T E R M E D I AT I O N
Clear limits exist when investing directly and with increasing capital allocated to the
PE asset class. Investors have no choice but to seek intermediation through fund in-
vestments. Funds are pooled, privately organized investment vehicles administered by
professional managers. PE funds are organized on behalf of qualified investors and are
not open to the general public. Because of their structure they can take advantage of
exemptions in regulations and legitimately exploit opportunities associated with lack of
transparency. PE funds are generally structured as asymmetric limited partnerships in
which limited partners (LPs) relinquish their ability to manage the business in exchange
for limited liability for the partnerships debts.
While terms and conditions as well as investor rights and obligations have long
been defined in specific nonstandard partnership agreements, the limited partnership
structureor comparable structures used in the various jurisdictionshas evolved
over recent decades into a quasi-standard. The fund usually has a contractually limited
life of 7 to 10 years.
The fund managers objective is to realize all investments before or when liquidating
the partnership. Often a provision exists for an extension of two or three years. A sub-
stantial part of the committed capital may remain undrawn and stays in the hands of the
LP, which only generates unattractive public market returns or even Treasury returns
unless the capital is put to work somewhere else where it can be easily taken out to sat-
isfy capital calls.
The choice between investing directly or through funds depends not only on the size
of the investment program but also on how much investors want to diversify through
incorporating more companies in a PE fund portfolio. For a small investment program
184 h o w p r i vat e e q u i t y w o r k s

focusing on local and later-stage companies, investing directly is advantageous. A


medium-sized investment program requires increasing the commitments to funds. For
a large program, the solo deals need to be replaced by co-investments, with third-party
fund managers supporting sourcing, monitoring, and existing portfolio companies. For
many institutional investors, investing through funds becomes a cost-effective solution.

P R I VAT E E Q U I T Y A S C A S H - F L O W A S S E T S
While direct investments can either be modeled as buy-to-sell or buy-to-keep, in-
vestment by PE funds are buy-to-sell because the fund structure is self-liquidating.
Funds and PE assets that are buy-to-sell could be seen as cash-flow assets. Such assets
usually cannot be traded profitably, create cash flows, and need to be sustained through
a timely provision of liquidity because the opportunity costs associated with undrawn
commitments require explicit or implicit over-commitment strategies. Making trada-
ble assets and cash-flow assets comparable involves either mapping market prices on
a cash-flow model or using a cash-flow model to determine a fair value, assuming that
this would fetch the same price in the market. This poses complications for portfolio
management approaches based on MPT, which relies on market prices and the ability
to transact at any point in time.

Portfolio Management
MPT is based on the Markowitzs (1952) insight that not fully or weakly correlated
assets can be combined in a way that maximizes return for any given level of risk. MPT
relies on the fundamental principle of diversification and suggests that allocation
choices in efficient markets are simple. Investors choose the appropriate combination
of the risk-free asset and the market portfolio that is in line with their risk aversion level.
This approach can then be adapted to include tax considerations and other investment
constraints.

A POWERFUL IDEA
Since Markowitz (1952) set forth MPT, this powerful idea has dominated the thinking
on portfolio design. To apply MPT, the return must be defined as the relative change of
a market price from the beginning to the end of a period. To find a portfolio fitting the
investment target of mean return and risk, investors need to input the expected mean,
expected standard deviation, and expected correlation among all considered assets
return to determine an efficient asset allocation. Models based on MPT may be suitable
for the publicly traded instruments, but for PE the traditional concept of return cannot
be as easily defined. Mean, standard deviation, and correlation cannot be reliably esti-
mated from historical data, which is essential for forecasting expected return. Moreover,
many of MPTs critical underlying assumptions such as normally distributed returns
are not met. Empirical evidence shows that PE returns are not normally distributed be-
cause of many small losses but few extraordinary home runs. Also, this non-normality
is much greater than what can be observed empirically in publicly traded markets.
Pr ivate E qu it y Port fol io M an ag e m e n t 185

S AT I S F I C I N G R AT H E R T H A N O P T I M I Z I N G
Arguably, MPTs main attraction lies in the optimization based on few inputs, provid-
ing an efficient frontier, which is the line at which a portfolio gives the highest return for
a given level of risk or the lowest risk for a given level of return. An individual assets
risk level does not matter as long as its returns vary from other assets risk levels in the
portfolio. The fundamental concept does not relate to individual assets; what is impor-
tant is how each asset changes prices relative to how every other asset in the portfolio
changes prices. The practices in PE are at odds with the idea of building an optimum
portfolio, as investors mainly rely on their selection skills, which trump overriding all
other considerations.
Despite its importance in finance, MPTs powerful mathematical apparatus is dif-
ficult to apply in PE. Here, quality data needed for precise calculations simply do not
exist. Risks, associated with investments in limited partnership funds, need to be mod-
eled based on qualitative as well as quantitative data. Realistically, portfolio manage-
ment in PE is about satisficing, a term coined by Simon (1978) that contrasts with
optimum decision-making. Instead, investors rely on heuristics to search through avail-
able alternatives until they find an acceptable solution.

Challenges
Discussing the challenges associated with PE portfolio management involves following
the RRTTLLU list. Investors typically approach their portfolio design based on invest-
ment targets in terms of return and risk, in addition to investment constraints such as
the time horizon, taxes and fees, liquidity, legal issues, and unique circumstances.

RETURN
In MPT, return is defined as the relative change of a market price from the beginning
to the end of a period while taking into account cash flows within the period. However,
PE is at odds with financial markets, which generate rich historical information and de-
pends on high precision and frequently updated data. For PE, the information base is
limited and investments do not have a market price. Thus, investors rely on proxies such
as valuation figures. In contrast to public market valuations, PE valuations are infre-
quent (typically quarterly), and based on either an experts appraisal or the occasional
third-party transaction related to the investment.
Measuring returns also depends on whether portfolio companies are directly held or
included in funds. Information on directly held PE investments in companies is difficult
to find. Performance data on funds are available from commercial databases, but typi-
cally on an aggregated level and without individual cash flows. The total value to paid-in
(TVPI), also called the money multiple, is one way of gauging a funds performance. The
TVPIs limitation is that it does not provide information on the investment periods du-
ration. A funds internal rate of return (IRR) is the most commonly used performance
measure, but it has well-known drawbacks, does not fit the return concept for traded
assets, and also does not always give the full picture.
186 h o w p r i vat e e q u i t y w o r k s

Distortions, such as the J-curve pattern of a funds annual returns, contribute to the
confusion. Here, the computed returns tend to be very poor in early years, giving the
impression that a funds performance is disastrous. However, this impression is an illu-
sion as companies are first valued at cost, fixed costs are deduced, and badly performing
companies are quickly liquidated or their value is written down. After a few years, this
trend reverses as the valuation of portfolio companies increases (e.g., due to third-party
transactions such as a new financing round), and others experience successful exits via
an initial public offering (IPO) or sale. The fund gives a picture of higher than normal
annual returns before leveling off again toward the end of the funds lifetime. The true
performance is usually steadier and less extreme, which raises the question whether
such returns really compensate for the risk involved.
Finally, correct but not necessarily precise valuation and consideration of the complete
picture are the cornerstones of effective risk modeling. To compute the effective return
requires examining not only the invested capital but also the return on all other resources
dedicated to PE investments. Notably the role of the funds undrawn commitments is
not fully understood at this point. Academic research into commitment risk is still in its
infancy (Cornelius et al. 2013). The performance of a portfolio of funds should be meas-
ured taking all resources dedicated to PE into account, including undrawn commitments
regardless of whether investors have sufficient liquidity at any given point in time.

RISK
Modeling risk similar to public equity on a time series basis has the clear advantage that
this framework addresses risk in the language with which the risk management units of in-
stitutional investors and regulators are familiar. This process appears straightforward and
attractive, as quarterly returns are thought to be easily comparable to public indices. Using
time series of valuations may look simple and straightforward, but technical complica-
tions arise. Taking historical risk figures to determine an expected risk for portfolio opti-
mization fails. Reasons for failure are similar to the ones discussed in measuring returns.
Investors often use value-at-risk (VaR) as a basis for determining their overall eco-
nomic and regulatory capital adequacy and measuring traded risk. Without market
prices, applying this concept to PE is difficult. Also, non-financial firms find that VaR is
difficult to apply to risk management as value mainly takes the form of real investments
in fixed assets that cannot be easily monetized. Industrial companies look at cash-flow-at-
risk (CFaR) as a more relevant measure for their risk exposures. CFaR is the maximum
deviation between actual cash flows and a set level (e.g., a budget figure) due to changes in
the underlying risk factors within a given time period for a given confidence level. Thus,
the Risk Measurement Guidelines of the European Private Equity and Venture Capital As-
sociation (EVCA) suggest that a VaR calculated based on cash-flow scenarios can be a
useful way of looking at PE-related risks (European Private Equity and Venture C apital
Association 2013). Correlations are another aspect of risk, which are later discussed.

TIME HORIZON
PE is a structurally illiquid asset class that carries a compensatory risk premium for
institutional investors. To benefit, they must have a liability profile that allows them
Pr ivate E qu it y Port fol io M an ag e m e n t 187

to lock capital in for a minimum of 10 years. For example, large university endow-
ments have the ability to sustain long investment horizons, which makes them eli-
gible to take on PEs illiquidity. However, pay-out policies and associated liquidity
constraints may dictate a more conservative investment strategy. Similar to univer-
sity endowments, sovereign wealth funds and family offices (i.e., funds managing a
familys fortune) can control their liabilities, which give them more flexibility when
investing in illiquid assets. Insurers need to be careful about their asset liability man-
agement (ALM) looking at both assets and liabilities when trying to optimize the
trade-off between risk and reward.

TA X E S
Tax issues can affect the optimal portfolio design. Many long-term oriented institutions
such as pension plans, foundations, and endowments are the most important source of
capital for PE funds. Historically, the tax-exempt status of PE investors in the United
States under the Internal Revenue Code has been a catalyst to their investments in PE
in which most income is gained from sales of portfolio companies rather than dividends
and interest. For investors, tax-exempt status permits more flexibility and allows them
to focus on total return instead on current income. Tax and regulatory requirements also
drive structure limited partnership funds with the additional objectives of tax transpar-
ency (i.e., investors are treated as if investing directly in the underlying portfolio compa-
nies). In many countries, the tax treatment of PE has increasingly come under discussion
and changes are likely to have an adverse effect on allocations to the asset class.

LIQUIDITY
An investors liquidity poses another constraint to the portfolios composition. In fact,
unforeseen liquidity needs are a hallmark of short-term oriented investors. Pension
funds and life insurance companies can match part of their long-term liabilities with
assets that mature within a similar time horizon. By contrast, non-life insurance com-
panies have an annual renewal cycle and less predictable pay-outs. Thus, sizable allo-
cations to PE are impossible. Banks are subject to comparable restrictions. Repeated
attempts to make PE more liquid, thus easier to price and more suitable for short-term
oriented investment strategies have not met with great success. This change can be seen
as counterproductive because it brings in precisely the problems PE techniques such as
the limited partnership structure have successfully overcome: the inability to put a price
on extreme uncertainty and the lack of investor commitment over the long horizons
necessary for success.

LEGAL ISSUES
Legal issues can also affect portfolio design. The most obvious one is regulation that
can restrict allocations to illiquid and non-marketable assets but, conversely, also makes
investing in alternative assets more attractive for unregulated institutions. In recent
years, a tightening and convergence of regulations made investing in PE more difficult
due to the closing of loopholes and narrowing of the scope for regulatory arbitrage.
188 h o w p r i vat e e q u i t y w o r k s

This trend is even true for nonprofits, which historically are viewed as largely unreg-
ulated in the United States. The Uniform Prudent Management of Institutional Funds
Act (UPMIFA) and the Sarbanes-Oxley Act created a more complex and restrictive
environment.
Both regulation and perceptions on regulation create a vicious circle regarding al-
locations to PE. Many perceive the asset class as risky, resulting in negligible alloca-
tions. Thus, investors put little effort into modeling risks. The tools employed tend to
be simple, but are often misleading, thus further reinforcing a sense of riskiness. These
issues concern how to structure the fund management company or the contractual ob-
ligation between GPs and LPs. Legal issues vary among different countries. Chapter3
provides a discussion of some regulatory developments involving PE in the United
States and Europe.

U N I Q U E C I R C U M S TA N C E S
Unique circumstances also drive allocations to PE. Maintaining pre-set target invest-
ment allocation levels is an important part of any institutional investors prudential
apparatus. The problem with many definitions of allocation targets is that they largely
ignore PEs illiquidity. Indeed, for unquoted assets, a fire sale mentality is problematic
because the market might fail to work resulting in the inability to sell assets no matter
the price. This situation can cause problems during public market downturns in which
prices of liquid assets fall along with the value of the overall portfolio. As the total port-
folio value is put into the denominator to determine the percentage of PE in the total
portfolio, the share of PE might now violate a set threshold although its value remains
unchanged. This so-called denominator effect can also happen during periods in which
PE valuations are appreciating and thus lead to breaches imposed by internal rules or
regulation. To address such breaches, investors may be forced to sell some holdings,
typically at a substantial loss.
Other investment constraints are self-imposed restrictions from stakeholders of the
investor, mostly the so-called environment, social, and governance (ESG) restrictions.
Examples include excluding investments in industries such as alcohol, tobacco, por-
nography, gambling, or arms. Due to regulations and organization-specific restrictions
and preferences, portfolio compositions often deviate from what institutional investors
would consider ideal.

Private Equity Allocation


Asset allocation shows the asset classes in which to invest for an optimal PE mix. Much
research suggests that asset allocation is the main driver of investment performance and
the added value provided by the selection of a particular asset within the class is small
(Brinson, Hood, and Beebower 1986, 1995; Brinson, Singer, and Beebower 1991; Ib-
botson and Kaplan 2000; Ibbotson 2010). Its fundamental justification is the notion
that asset classes perform differently in dissimilar market and economic conditions.
Consequently, this type of diversification is sometimes described as the only free
lunch available in financial markets.
Pr ivate E qu it y Port fol io M an ag e m e n t 189

NAVE APPROACHES
Swensen (2000), Fraser-Sampson (2006), and Cornelius (2011) discuss how PE is
embedded in a broader asset allocation. In recent years, the Yale model or endowment
model of investing has attracted attention from both academics and practitioners. Ac-
cording to Swensen, this model embraces the principles of MPT, though in a simplistic
and robust way. It is based on diversification across asset classes with low correlations to
maximize risk-adjusted investment return. The Yale model is essentially a nave diver-
sification approach, broadly dividing a portfolio into five or six roughly equal parts and
investing each in a different asset class. The portfolio should be regularly rebalanced to
the original weightings of the asset classes. Such nave diversification recognizes that the
risk-return relationship for long-term oriented assets cannot be effectively quantified.
By rebalancing, investors in theory would be selling PE positions on secondary markets
when prices are high and discounts are low, and buying low (i.e., committing to funds
when a scarcity of funding exists and portfolio companies can be acquired at attractive
prices). An equity orientation to the detriment of asset classes with low expected re-
turns such as fixed income and commodities is a key part of this model. The novelty of
this model is that liquidity is to be avoided rather than sought out because it comes at a
heavy price through lower returns and has relatively high exposure to alternative asset
classes including PE compared to more traditional portfolios. Swensen provides no in-
depth guidance on how to build the PE portfolio.

BACKGROUND RISK
In practice, allocations to illiquid asset classes such as PE are not as nave. They are often
those dictated by an investors pay-out policies and liquidity constraints. Research by
Dimmock (2008) suggests that if an economic agent is endowed with a non-tradable
risk (i.e., background risk), this should decrease his appetite for other risks even if all
sources of risk are independent. Market imperfections should have a large impact on
portfolio choice. In situations exposed to background risk, investors should optimally
choose unique portfolios that best hedge their personal risks. Dimmock examines back-
ground risk for university endowments, which he defines as the volatility of a univer-
sitys non-financial income. His results strongly support the hypothesis that this risk
drives the endowment fund portfolio composition. He suggests that endowment man-
agers choose portfolios that lower the overall risk of the university entity comprising
both its endowment fund and its non-investment operations.
Because universities have theoretically infinite lives, investment strategies for their
endowments should not be excessively risk-averse. Yet, the endowment needs regular
cash inflows to meet the operating needs of the university. The same argument applies
to foundations and family offices (i.e., a private company that manages investments
and trusts for a single family). Foundations can scale back their grant-making, but mu-
seums or schools depend on endowments to cover their budgets. Universities with
greater income risk or high debt-to-assets ratios tend to avoid alternative assets such
as venture capital (VC). Public universities also invest more in fixed income as they are
mainly funded by public means through a national or sub-national government. Thus,
some could argue that such universities receive their endowment from the outside.
190 h o w p r i vat e e q u i t y w o r k s

Research-intensive universities need to hold much safer portfolios than liberal arts col-
leges because their funding needs are higher and more volatile. Therefore, their endow-
ment funds need to be more liquid.

Portfolio Management Frameworks for Private Equity


How can portfolio managers manage a portfolio of PE assets? Here market practices are
based on heuristics. Too many of the MPTs assumptions do not apply for buy-to-sell
PE assets, which makes considering other frameworks such as behavioral finance or the
AMH worthwhile.

B E H AV I O R A L F I N A N C E
Layered portfolios such as those reflected in the rules of core-satellite make sense in
the context of behavioral finance, which also considers behavioral human aspects in the
investment process (e.g., aversion to loss) that are important when allocating portfolio
assets. The MPT only considers return, risk, and correlation, rendering layered portfo-
lios as sub-optimal. The core-satellite method aims to increase risk control and lower
costs. For example, the portfolio can be divided into two layers. The core portfolio
consists of institutional quality funds that can raise large pools of capital and are ex-
pected to generate a predictable base return. The satellite portfolio has niche strategies
funds that fall outside of the mainstream (e.g., emerging markets, new teams, and spe-
cialist funds). The core-satellite approach provides a framework for targeting and con-
trolling those areas where investors believe they can better control risks or are willing
to take more risk.
This strategy may be effective for institutions that want to diversify their portfolios
without sacrificing the potential for higher returns generated by selected active man-
agement strategies. Another advantage is the flexibility to customize a portfolio to meet
specific investment objectives and preferences. It also provides the framework for tar-
geting and controlling those areas where investors believe they can better control risks
or are willing to take more risk. What makes up core versus satellite depends on the
investors focus and expertise. Some see VC as satellite, while others view a balanced
BO and VC fund portfolio as core. Finally, this approach also allows more time on the
satellite portfolio, which is expected to generate the highest returns and less time on the
core portfolio, as it is less risky.
Behavioral models use social, cognitive, and emotional factors to understand in-
vestment decisions and are primarily concerned with the bounds of rationality (e.g.,
selfishness and self-control) of market participants. They help explain layered pyramid
portfolios, often seen with private small investors. These investors view their portfolios
not as a whole, as prescribed by MPT, but as distinct layers in a hierarchy of assets, in
which layers are associated with particular goals and attitudes toward risk vary across
layers.
However, behavioral finance models do not offer a satisfactory explanation. Why
should investors protect their downside within an allocation to PE if they can achieve
Pr ivate E qu it y Port fol io M an ag e m e n t 191

the same effect much cheaper with allocations to more conventional assets? The core-
satellite is also called a core and explore (Pietranico and Riepe 2002) approach and
therefore appears to be more similar to concepts put forward under the AMH than to
behavioral finance.

A D A P T I V E M A R K E T H Y P OT H E S I S
The AMH originated in the hedge fund world to reconcile the raison dtre of many
hedge fundsdoing arbitragewith the EMH, which assumes that no arbitrage is pos-
sible. The AMH addresses the market inefficiencies that hedge funds exploit as fleeting
opportunities. Good investors of some hedge funds may be the first to spot them and
have a time window during which they can profit, but eventually others detect the in-
efficiency and it finally disappears. To stay ahead in the game, hedge fund managers
need to continuously look out for new inefficiencies. However, even an excellent track
record does not guarantee that the manager can find comparable opportunities again,
and indeed many fail to do so.
The AMH embraces the principle of evolution that allows hedge fund managers
to flourish within changing environments characterized by extreme uncertainty. The
AMH assumes that market participants make mistakes, but also learn. Competition
drives adaptation and innovation, natural selection shapes market ecology, and evolu-
tion determines market dynamics. Speculative opportunities do exist in the market, but
appear and disappear over time, so innovation in the form of continuous search for new
opportunities is the key to survival. Unsuccessful investors, who continue to make mal-
adaptive decisions, are eventually eliminated from the market. The AMH is a relatively
recent framework developed by Lo (2005) and Lo and Mueller (2010), who argue
that the practical implementation of the theory would be more difficult than investing
under the EMH framework. They put forth arguments mainly in the context of hedge
funds, which are the Galapagos Islands of the financial globe but, unlike PE, operate
in a near-efficient market. Academia is struggling to reconcile several theoretical frame-
works such as AMH and EMH.
The underlying dynamics of PE are more difficult to sense as the pace of evo-
lution is much slower than in the hedge fund world, but the observable structures
clearly support Los (2005) hypothesis. Taking a long-term perspective and looking
at players such as endowments and family offices indicate evolutionary processes.
The industry is experimenting with other structures, terms, and conditions but not
always successfully. What is best practice today can be obsolete tomorrow. Such de-
velopments certainly do not happen with the extraordinary speed of the hedge fund
industry. In PE, things progress more slowly, but the same dynamics appear to be
at work.

A P P L I C AT I O N TO P R I VAT E E Q U I T Y P O R T F O L I O S
Successful investing in PE is largely based on an informational advantage associated with
markets that are neither efficient nor regulated. Here intelligence and networks form the
basis of an investment strategy; the opacity of this market makes approaches based on
secrecy and proprietary insights feasible. Portfolio management in PE has to answer
192 h o w p r i vat e e q u i t y w o r k s

how to best search for opportunities and identify the best proposals ahead of competing
investors. Chances to outperform exist, but are only temporary and cannot necessarily
be identified based on historical data. Once a market matures and data are readily avail-
able, the opportunities to exploit inefficiencies will often already have passed, and the
investor can only hope to earn a risk or illiquidity premium.
For portfolio composition, investors need to strike a balance between stability
and experimentation or between diversification and concentration. Investors explore
through diversification and exploit through building concentrated portfolios or increas-
ing their allocations to niches with identified potential.

Diversification Management
The view that risk for PE is very high may be true for individual investments. Yet, for
well-selected and diversified portfolios, the standard deviation of returns is smaller
than intuitively expected. Diversification protects the portfolio amid uncertainty and
change, but increasing the number of assets restricts the range of possible portfolio
returns, which converge to the markets average performance. Investors can also do
the opposite and use the principle of concentration to achieve growth. This behavior
raises several questions. Does an optimum diversification level exist? Should investors
concentrate their portfolios in periods when the economy is favorable to PE? To what
degree can diversification address exposure to foreign exchange related risks? Finally,
how can investors assure a sufficient degree of independence among PE assets, so that
diversification effects really take place?

D I V E R S I F I C AT I O N L E V E L
Pinpointing the optimal diversification level is not straightforward. For portfolios of PE
funds, Weidig and Mathonet (2004) find that the maximum diversification benefit is
achieved with just 20 to 30 positions as Figure 11.1 shows. The question of how much
to diversify depends on an investors perceived selection skills and the returns targeted
for the overall PE portfolio. Investors who are confident in their ability to identify and
access opportunities with returns above the targeted return should aim for concentrated
portfolios that comprise the best opportunities.
If investors are unsure of their selection skills but are optimistic about the PE asset
class, the universe of available opportunities should promise an average return above
the targeted return. This attitude is certainly reasonable, at least temporarily in booming
market niches, and a highly diversified portfolio would offer protection against a perfor-
mance below the target.
In many situations, neither one nor the other can be assumed. How can investors
deal with a situation of high uncertainty in which they are not confident of their ability
to identify winners and average returns may be insufficient? This situation describes, for
example, the VC market in many countries. Such a diversified portfolio will typically
be too expensive. Therefore, investors need to look at how to amplify a positive impact
(e.g., through real options) as will be discussed later in the chapter.
Pr ivate E qu it y Port fol io M an ag e m e n t 193

30%
Direct Funds FoFs
25%

20%
Probability

15%

10%

5%

0%
0 1 2 3 4 5 6 7 8 9 10
Multiple

Figure 11.1 The Risk Profile of Venture Capital: Direct, Funds, and Fund-
of-Funds The histogram shows the distribution of multiple returns for three different
investment types of VC: direct, fund, and fund-of-funds investments. While direct
investments can lead not only to total losses but also to extreme wins, the fund and fund-
of-funds investments are distributed more normally and have no extreme multiple.
Source: Weidig and Mathonet (2004).

MARKET TIMING
When discussing diversification versus concentration, the appropriateness of market
timing is specific to each case. Market timing is part of an allocation strategy that aims
to exploit market cyclicality. This is based on the observation that in PE marked dif-
ferences in performance occur from one vintage year to another. Figure 11.2 shows
the average fund performance per vintage year for VC and PE worldwide from 1990
to 2000. The boom-to-bust-and-back years of the 1990s are representative of the ex-
treme average returns shown per vintage years, ranging from low single digit numbers
to 20percent plus.

%
25

20

15

10

0
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000
Vintage Year

Figure 11.2 Average Fund Performance per Vintage Year, 1990 to 2000This
figure shows the average fund performance per vintage year for VC and PE worldwide
between 1990 and 2000.
194 h o w p r i vat e e q u i t y w o r k s

Many practitioners view market timing as a losing strategy because no evidence sug-
gests that large institutions can consistently enter the market when it is low and exit
when it is high (Swensen 2000; Cornelius 2011). The proven modus operandi is to con-
sistently invest a fixed amount throughout the vintage years, which leads to an anti-
cyclical exposure to the ups and downs of the PE market. The investment managers
underinvest during periods of exuberance and take bolder bets in other years as com-
pared to their peers. Leaving the herd behind may increase the subjective risk felt by
managers, although it is the right strategy.

FOREIGN EXCHANGE
As PE has developed into an international asset class, investors are increasingly exposed
to currency movements. Thus, the currency decision needs to be explicitly incorporated
into the portfolio management process. In the case of PE, uncertainty concerns not only
the foreign exchange rate but also the amount and the timing of cash flows, all of which
are unrelated.
In theory, hedging instruments can be customized, although this is not an option
in practice due to unacceptable costs. Instead, investors tend to rely on natural hedges
(i.e., techniques to lessen risk in an investment by diversifying with investments that
offset, at least to some degree, the original one). Here portfolios are engineered ex-ante
by modeling cash-flow characteristics and by trying to achieve a targeted risk profile.
In many situations, hedging in any form may not even be desirable. In fact, foreign ex-
change can be seen as one of several assets comprising the PE portfolio. The impact of
investing in various currencies can be beneficial to the portfolios risk and return as well
as its liquidity profile.

REDUCING INTERDEPENDENCE
The question of what diversification level is meaningful also depends on the correla-
tions of selected positions. Correlation is a cornerstone of portfolio optimization. In-
vestors in search of diversification seek assets that behave differently from those in the
rest of their portfolios. In theory, correlations can be measured by looking at historical
return data that serve as a reference for the expected correlation. Correlations cannot
be measured this way for PE, as the approximated return gives too few data points with
which to perform an analysis. Even the richest possible set of relevant data (e.g., based
on quarterly net asset values and daily cash flows) is insufficient due to stale pricing and
mechanical valuation rules and thus leads to far too high autocorrelation.
Instead, PE portfolios need to be designed for low correlation. Experience suggests
that over vintage years, PE shows markedly different performance levels. Portfolios re-
semble the age pyramid of a human population and can be unbalanced when the time
dimension is ignored: a population consisting of 50 percent males and 50 percent fe-
males may appear to be a good balance, but it is a serious problem if the population only
comprises females over the age of 90 and males under the age of 10. The label chosen for
assets is irrelevant, and so a proper classification of PE investments along independent
risk dimensions is an important tool for managing diversification. With some degree of
subjectivity, this goal can be achieved.
Pr ivate E qu it y Port fol io M an ag e m e n t 195

Ideally, a portfolio remains diversified across all risk dimensions even under eco-
nomic stress. One tool for managing this objective is cluster analysis, a technique that is
widely applied in different research areas such as medicine, marketing, and economics.
It is used to classify similar objects into relatively homogeneous groups (or independent
risk dimensions) and dissimilar objects into different groups. This technique cannot be
applied wholesale to PE, as interpreting the results is too subjective. Mostly, looking at
the degree to which, under different stress levels, a portfolio tends to form clusters of
sub-portfolios that cannot be seen as at least partially independent is sufficient.
The stress level is the minimum distance two assets need to have in their classifica-
tions to be considered independent. For example, two funds with different classifica-
tions for fund manager and vintage year and all other things equal have a distance
of two. One fund can be considered to be in reach of another and be in the same cluster
if the distance is lower than the set stress level. Ideally, a portfolio is diversified across
all risk dimensions and clog incrementally with increasing stress levels, thus allowing
investors to incur the same risk for lower levels of costly diversification.

Implementation
Even when the broad approach to diversification is clarified, questions about the practi-
cal implementation remain. Investors usually construct PE portfolios either top-down
or bottom-up depending on their own expertise and constraints.

TO P - D O W N A N D B OT TO M - U P A P P R O A C H E S
The top-down approach is based on a construction of the overall portfolio by deter-
mining allocation ranges and then searching for opportunities that fit these allocations.
It gives priority to selected sectors, countries, trends, and vintage years as opposed to
individual assets. Investors who follow a top-down approach typically emphasize the
overall composition of their portfolio. Others doubt whether a top-down approach is
meaningful for PE markets. Apart from the questions associated with determining allo-
cation weights, the major shortcoming of a top-down approach is that a strict adherence
to these ranges is impossible in reality. When investing in funds, finding and accessing a
sufficient number of superior managers to fill in each pre-determined sub-class alloca-
tion can be difficult. Indeed, often only one or two superior fund managers operate in a
particular sector, and they raise capital only every three or four years.
The bottom-up approach focuses on screening all investment opportunities with
quality being the overriding criterion, irrespective of portfolio considerations such as
sector or geographical diversification, which here are assumed to have a lesser impact.
The first step is identifying suitable investments, followed by analyzing opportunities
in order to rank them based on attractiveness. The bottom-up approach has several at-
tractive features and arguably is the most widely used: it is intuitively appealing, easy to
understand, and robust because it depends solely on ranking. However, this approach is
not without problems. Because it is opportunistic, the bottom-up approach can lead to
an unbalanced portfolio, carrying more risk than expected.
196 h o w p r i vat e e q u i t y w o r k s

In practice, bottom-up and top-down approaches are typically used in tandem, with
most investors following a combined approach. Investors would not consider assets that
are not of a minimum quality just to fulfill a targeted allocation. Similarly, even investors
who are convinced of their selection skills are aware of the importance of effective diver-
sification. The bottom-up approach could also be more representative of a period during
which some LPs indeed had strong selection skills, which is unlikely to work nowadays.

M O N I TO R I N G A N D R E B A L A N C I N G
Even though a portfolios design is crucial, investors also need to monitor its imple-
mentation and development via monitoring key measures summarizing important
aspects of the portfolio. For example, Weidig and Grabenwarter (2005) propose the
following measures: total drawdowns, total NAV, total exposure to sectors, stages, ge-
ographic areas and individual teams, average lifetime, performance on an overall and
sub-portfolio basis, and diversification measures such as exposure across vintage years,
region, or sector. Most measures are exclusively used for reporting, while others such
as diversification measures can be used to decide which new PE investments should
be made. For example, if the sub-portfolio of bio-tech exited earlier than expected and
the investment period is not over yet, the portfolio has a lack of bio-tech exposure and
rebalancing toward more bio-tech makes senses. An investor could also undertake a sec-
ondary purchase if the opportunity presents itself. Modeling the future drawdowns and
exits of a PE portfolio is important for understanding short-term and long-term liquid-
ity needs.

ACTIVE PORTFOLIO MANAGEMENT


For an illiquid asset class, portfolio rebalancing is impractical. Sometimes investors
need to exit the PE markets (e.g., to implement a change in investment strategy, satisfy
regulatory changes, and free up capital for their own funds to support regulatory capi-
tal). In situations in which investors are forced to trade in declining secondary markets
result in high exiting costs. Mechanistically administered portfolio allocation rules can
be too simplistic and can be at odds even with MPT, which does not assume a rebalanc-
ing through such distressed sales with potentially large losses. As demonstrated by the
denominator effect previously discussed, such rules often have unintended and unde-
sirable outcomes.
Instead, PE portfolios need to be largely designed up-front and implemented accord-
ingly: improved portfolio management mostly results from improved portfolio design.
Additionally, investors need to continuously monitor and model the future evolution of
their existing PE portfolios to understand the impact asset classes have on their wider
portfolio, implement cash-flow management, and decide the nature of future private
investments. While a target allocation may exist, a flexible allocation approach allows
for a bandwidth within which deviations from the target are tolerated, and thus helps to
avoid or at least mitigate the denominator effect.
Exiting PE mostly consists of letting the funds run out or waiting for all underlying
companies to exit. Because this process can take many years, investors need to have a
sufficiently long time horizon and cannot easily move investments in and out of the PE
asset class.
Pr ivate E qu it y Port fol io M an ag e m e n t 197

Outlook
Few players initially participated in a largely underexplored PE market that offered rich
opportunities. In such circumstances, access and due diligence were critical. Now the
landscape has become increasingly competitive and these traditional tools often do not
lead to the desired outcomes any longer. Increased inflows of capital into the asset class
have largely neutralized the institutional learning about PE investments and improve-
ments of skills in assessing such opportunities. Research evidence shows that average
PE returns are mediocre (Cornelius et al. 2013), but still many investors appear to suc-
cumb to the natural human tendency to overestimate their own capabilities, which in
behavioral finance is called overconfidence bias. This is also called the Lake Wobegon fal-
lacy, named after a fictional town where all children are above average. Investors must
not rely on above average selection skills and instead look for other ways to improve
their chance of success such as improved portfolio management.
In PE, investments are frequently made in conditions of extreme uncertainty and,
due to the illiquidity of this asset class, decisions are often effectively irreversible. In
fact, the asset class flourishes in an environment of extreme uncertainty. Rather than
being an obstacle, uncertainty is a continuous source of new opportunities and the very
engine of innovation. Making irreversible all-or-nothing decisions in the face of uncer-
tainty is risky, but investors can at least prepare to manage the uncertain outcomes once
they happen. The real option method recognizes the value of deferring investment deci-
sions and of learning over time.
Real options are useful during situations involving a high degree of uncertainty, some
managerial flexibility, and incomplete ex-ante information. Co-investments, secondary
transactions, and side-funds are established parts of the PE market, but are rarely viewed
as such. For instance, co-investments tend to be seen as a way to cut costs. LPs often refuse
to pay and fund managers charge little or no fees for access to co-investment opportu-
nities and claim no carried interest, only transaction fees without set-off. However, the
perception that all co-investment deals come without fees and carried interest is wrong.
Indeed, some LPs incentivize GPs by paying the so-called promote, a jargon that com-
prises carried interest and annual management fees on co-investments.
While portfolios of funds are effectively a diversification strategy, co-investments
and buying existing investments (so-called secondaries) allow for concentrations of
portfolios and LPs with a highly diversified primary fund program to amplify exposure
to industries they expect to have above-average upside potential given the stage of the
cycle. Co-investments can also address the fact that fundraising cycles do not necessar-
ily correspond to industry trends. The real option framework remains a comparatively
new paradigm with a potential that is rarely fully realized, as PE investors still tend to
underestimate its relevance to value creation.

Summary and Conclusions


While this chapter has highlighted the challenges of fitting PE into at standard frame-
work, building a PE portfolio is accomplished primarily using simple heuristics;
whereas, MPT is difficult to apply and in many situations inapplicable. Therefore, issues
198 h o w p r i vat e e q u i t y w o r k s

particularly arise when the investor wants to include PE in the general portfolio design
and management. For institutional investors managing multi-asset portfolios, risk
management systems and economic and regulatory capital models require a consist-
ent approach. The growing importance of portfolio management in PE requires that
institutional investors move beyond the heuristics described in this chapter and employ
such techniques that embrace the cash-flow characteristics of PE and take real options
into deeper consideration than in the past. Tackling these questions is likely to pose
challenges for both practitioners and academics.

Discussion Questions
1. List and describe differences between PE investments and tradable assets such as
common stock and bonds.
2. Explain why PE does not fit into the standard Markowitz portfolio optimization.
3. Describe the two alternative frameworks to MPT and how they are relevant for PE
investment.
4. Discuss the pros and cons of bottom-up and top-down portfolio constructions.
5. List and explain measures that can be used to monitor a portfolios evolution.

References
Brinson, Gary P., L. Randolph Hood, and Gilbert L. Beebower. 1986. Determinants of Portfolio
Performance. Financial Analysts Journal 42:4, 3948.
Brinson, Gary P., L. Randolph Hood, and Gilbert L. Beebower. 1995. Determinants of Portfolio
Performance. Financial Analysts Journal 51:1, 133138.
Brinson, Gary P., Brian D. Singer, and Gilbert L. Beebower. 1991. Determinants of Portfolio Perfor-
mance II: An Update. Financial Analysts Journal 47:3, 4048.
Cornelius, Peter. 2011. International Investments in Private Equity. Burlington, MA: Academic Press.
Cornelius, Peter, Christian Diller, Didier Guennoc, and Thomas Meyer. 2013. Mastering Illiquidity
Risk Management for Portfolios of Limited Partnership Funds. Chichester, U.K.: John Wiley &
Sons.
Dimmock, Stephen G. 2008. Portfolio Choice, Background Risk, and University Endowment
Funds. Working Paper, Michigan State University.
European Private Equity and Venture Capital Association. 2013. Risk Measurement Guidelines
2013. Available at http://www.evca.eu/uploadedfiles/EVCA_Risk_Measurement_
Guidelines_January_2013.pdf.
Fraser-Sampson, Guy. 2006. Multi Asset Class Investment Strategy. Chichester, U.K.: John Wiley &
Sons.
Geltner, David, Bryan D. MacGregor, and Gregory M. Schwann. 2003. Appraisal Smoothing and
Price Discovery in Real Estate Markets. Urban Studies 40: 56, 10471064. Available at
http://usj.sagepub.com/content/40/5-6/1047.abstract.
Getmansky, Mila, Andrew W. Lo, and Igor Makarov. 2004. An Econometric Model of Serial Corre-
lation and Illiquidity in Hedge Fund Returns. Journal of Financial Economics 74:3, 529609.
Ibbotson, Roger G. 2010. The Importance of Asset Allocation. Financial Analysts Journal 66:2,
1820.
Ibbotson, Roger G., and Paul D. Kaplan. 2000. Does Asset Allocation Policy Explain 40, 90, or 100
Percent of Performance? Financial Analysts Journal 56:1, 2633.
Pr ivate E qu it y Port fol io M an ag e m e n t 199

Lo, Andrew W. 2005. The Adaptive Market Hypothesis. Journal of Investment Consulting 7:2,
2144.
Lo, Andrew W., and Mark T. Mueller. 2010. WARNING: Physics Envy May be Hazardous to Your
Wealth. Available at http://ssrn.com/ abstract = 1563882.
Markowitz, Harry. 1952. Portfolio Selection. Journal of Finance 7:1, 7791.
Pietranico, Paul, and Mark Riepe. 2002. Core & ExploreDetails. Charles Schwab and Company.
Available at http://www.schwab.com.
Simon, Herbert A. 1978. Rationality as Process and Product of Thought. American Economic
Review 68:2, 116.
Swensen, David. 2000. Pioneering Portfolio Management: An Unconventional Approach to Institutional
Investment. New York: Simon & Schuster.
Weidig, Tom, and Ulrich Grabenwarter. 2005. Exposed to the J-curve. London: Euromoney Books.
Weidig, Tom, and Pierre-Yves Mathonet. 2004. The Risk Profile of Private Equity. January. Euro-
pean Venture Capital Association. Available at http://ssrn.com/abstract=495482.
12
The Role of Private Equity in Initial
Public Offerings
The Case of Venture Capital Firms

S H A N TA N U D U T TA
Associate Professor, University of Ottawa

A R U P GA N G U LY
PhD candidate, University of Pittsburgh

LIN GE
PhD candidate, University of Pittsburgh

Introduction
Venture capital (VC) involves providing funds to companies in their early stages. VC
firms are critical drivers of economic growth, both in developed and developing coun-
tries. Many prominent publicly traded companies such as Amazon Inc., Starbucks Corp.,
and Twitter, Inc. are household names partly because VC firms backed them in their
early stages. The influence of venture capitalists (VCs) is such that sometimes investee
companies are valued in billions of dollars even before they are taken public. For exam-
ple, VC-backed Uber Technologies Inc.s valuation was $18.2 billion in June 2014. VC
firms do not typically make long-term investments in companies. Instead, they invest
with the intent of exiting companies once they receive their expected returns.
Common exit strategies for VC firms include an initial public offering (IPO), trade
sale (i.e., acquisition), management buyout (MBO), secondary sale (i.e., refinancing),
and liquidation (i.e., write-off). In recent years, the most common exit strategy has been
trade sales (Preqin 2014). However, historically IPOs have long been the standard for
VC exits due to their excellent performance.
With the passage of the Jumpstart Our Business Start-Ups ( JOBS) Act by the
Obama administration on April 5, 2012, which reduced the required regulations for
IPOs of emerging growth companies and the improving market conditions during the
post-financial crisis of 20072008, IPOs are expected to return to being the preferred

200
The Rol e of P riv at e E qu it y in I P Os 201

choice of VCs. With the introduction of the JOBS Act, some expect that access to public
money will be less restrictive and costly for private companies. Start-up and smaller
companies can take advantage of the crowdfunding provision of the Act, which makes
raising funds easier for these firms. Crowdfunding is a form of financing a new project
or venture that relies on smaller monetary contributions from many individuals. The
JOBS Act also increased the shareholder limit from 500 to 2,000 before a company is
required to register with the Securities and Exchange Commission (SEC). This change
allows a company to grow further with its private company tag avoiding a premature
public offering. Further, the the IPO On-Ramp provision of the JOBS Act requires
that an emerging growth company (i.e., one with gross revenues of less than $1 billion
in the last fiscal year) needs to make fewer disclosures for a public offering and can enjoy
this limited disclosure provision for up to five years after the IPO. Thus, the JOBS Act is
likely to help relatively smaller companies raise funds and promote more IPO activity.
Figure 12.1 shows the number of financial buyers, including both PE (private
equity) and VC firms that exited through IPOs during the three-year period ending
May 26, 2014. This exit strategy is still a popular choice in countries such as the United
States and Canada and in industries such as information technology and consumer
discretionary.
As Figure 12.2 shows, the number of VC-backed IPOs in the United States from
1980 to 2013 consists of a large fraction of the total number of IPOs in the United
States every year. For this period, more than 35 percent of the total IPOs, on average,
were VC-backed IPOs according to Jay Ritters IPO database (Ritter 2014). Kaplan,
Sensoy, and Strmberg (2009), however, find that a much higher percent of IPOs are
VC-backed. For example, they document that in 2004, 83 percent of nonfinancial IPOs
were VC-backed.

Latin America and Utilities, 8


Telecommunication No Primary Industry
Caribbean, 22 Services, 9 Assigned, 1
Energy, Materials,
Europe, 81 45 39
Information
Technology,
Asia/Pacific, 144 Industrials,
111 87
United States Financials,
and Canada, 389 64
Consumer
Healthcare, Discretionary,
61 136
Africa/Middle
East, 14

Consumer Staples, 23

Figure 12.1 Number of Financial Buyer Exits via Public Offerings in


Different Geographic Regions and Industries This figure shows the number
of financial buyer exits via public offerings in different geographic regions and industries
from May 26, 2011, to May 26, 2014. The following screening criteria are used: sellers
(all investment firms/financial buyer exits); time period (May 26, 2011, to May 26,
2014); transaction types (public offerings); and transaction status (announced, closed, or
effective). Source: S&P Capital IQ Database (2014).
202 h o w p r i vat e e q u i t y w o r k s

800
700
600
Number of IPOs

500
400
300
200
100
0
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
2099
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
13
19

Years
Number of All IPOs Number of VC-backed IPOs

Figure 12.2 Number of Venture-Capital-backed Initial Public Offerings


in the United States, 1980 to 2013 This figure shows the number of all IPOs and
VC-backed IPOs from 1980 to 2013. The data exclude IPOs with an offer price below $5
per share, unit offers, American depository receipts (ADRs), closed-end funds, oil and gas
limited partnerships, acquisition companies, real estate investment trusts, bank and saving
and loan IPOs, and firms not listed on CRSP. Source: Jay Ritters IPO database (http://bear.
warrington.ufl.edu/ritter/ipodata.htm).

As Figure 12.3 shows, both the short- and long-term performance of VC-backed
IPOs completed from 1980 to 2013 appear better than non-VC-backed IPOs. While
the average first-day return of non-VC-backed IPOs was 12.60 percent, the VC-backed
IPOs have an average first-day return of 27.40 percent. Even the longer term returns in-
cluding the average three-year buy-and-hold return (both market- and style-adjusted)
are higher for VC-backed IPOs.

30% 27%

20%
13%
Average 3-year Buy-and-Hold Average 3-year Buy-and-Hold
10%
Return (Market-Adjusted) Return (Style-Adjusted)
Returns

1%
0%
Average First-day Return

10%
11% 11%
20%
23%
30%
VCbacked Non VCbacked

Figure 12.3 Comparison of the Short- and Long-Term Returns of


Venture-Capital-Backed and Non-Venture-Capital-Backed Initial Public
Offerings in the United States, 1980 to 2013 This figure shows that during the
period from 1980 to 2013, the short-term and long-term performance of VC-backed IPOs
is better, on average, than that of non-VC-backed IPOs. Source: Data extracted from Jay Ritters
IPO database (http://bear.warrington.ufl.edu/ritter/ipodata.htm).
The Rol e of P riv at e E qu it y in I P Os 203

Brav and Gompers (1997) reach a similar conclusion that VC-backed IPOs usually
outperform non-VC-backed IPOs when they compare the long-run performance of 934
VC-backed IPOs from 1972 to 1992 to 3,407 non-VC-backed IPOs from 1975 to 1992.
More recently, Puri and Zarutskie (2012) use more than 25 years of data from different
government and proprietary data sources to study the life cycle of VC- and non-VC-
financed firms. The authors document that the cumulative failure rate of VC-financed
firms (39.7 percent) is significantly lower than that of non-VC-financed firms (78.9 per-
cent). Thus, the evidence shows that VC-backed IPOs have generally been more suc-
cessful, especially in the United States.
The objective of this chapter is to discuss the vital roles played by VC firms in IPOs.
The extant literature presents three distinct roles of a VC firm: certification, underpric-
ing, and monitoring. All three roles can affect the outcome of an IPO. VC involvement
in an IPO deal is considered certification by a venture capitalist. It signals the potential
that a company will seek additional funding through a public listing and instills confi-
dence in prospective investors.
Another important issue in the IPO literature is the degree of underpricing asso-
ciated with the new security on a stock exchange. Underpricing refers to the first-day
return of a newly listed firm. Newly listed firms and institutional investors prefer to have
higher first-day returns to signal better prospects for the firm and to implant more con-
fidence in investors. By contrast, VCs with a greater level of financial involvement might
not prefer a high level of underpricing because that would deprive them of higher finan-
cial gains.
The third important aspect of the role of VCs is monitoring their investments. Stud-
ies show that VCs have incentives to monitor the entrepreneurs to reduce information
asymmetry and lend managerial support to the relatively inexperienced management
team of a new firm. However, gaining a deeper understanding of the role of VCs in an
IPO involves reviewing the interests and interactions among various players, such as
the entrepreneur, lead investment banker, VC firm, and investors. Sometimes, VCs have
conflicts of interest that may affect the structure and outcome of an IPO. Accordingly,
this chapter focuses on both the conflicts of interest and the roles played by a VC firm
in an IPO.
The rest of this chapter is organized as follows. The next section provides an over-
view of the conflicts of interest in a VC-backed IPO. The following section discusses the
three distinct roles played by VC firms: certification, underpricing, and monitoring. The
final section provides a summary of the findings and conclusions.

Conflicts of Interest
A VC-backed IPO has four distinct economic agents with potentially conflicting inter-
ests: (1) the entrepreneur, (2) lead investment banker, (3) VC firm, and (4) investors.
Although the entrepreneur wants to promote company success, the lead investment
banker strives for a successful deal. The VC firm is interested in maximizing the com-
panys selling price and money raised in the public capital markets providing for a
more profitable exit. Investors seek to minimize the purchase price to increase their
returns.
204 h o w p r i vat e e q u i t y w o r k s

The conflicts of interest for VCs stem from VC firms being financial buyers and
not strategic buyers. VC firms want to get the highest price possible in the IPO and do
not want to leave money on the table due to underpricing. Additionally, VCs usually
prefer taking the firm public sooner than the entrepreneur because VCs like to raise
money and show returns in order to raise subsequent funds (Gompers 1996). In con-
trast, entrepreneurs prefer taking more time and are often hesitant to go public too soon
because they do not like to face strict public scrutiny. Also, a principalagent problem
may exist. The entrepreneur, who can be considered the agent in such a set-up, usually
has inside information about the business, which may be unknown to the venture cap-
italist, who is the principal. This arrangement may cause agency costs due to over- or
underinvestment by the entrepreneur.
Another conflict of interest in VC-backed IPOs is when the lead underwriters
are also the VC partners who own the company before the IPO. An example is
Dollar General Corporations IPO in 2009 in which the PE divisions of the lead
underwriters for the deal (i.e., Citigroup, Goldman Sachs, and Kohlberg Kravis
Roberts (KKR)) were also investors. Although JPMorgan was brought in as an in-
dependent underwriter to meet National Association of Securities Dealers, Inc.
(NASD) Rule 2720 requirements, the mere presence of an independent investment
bank probably did not entirely mitigate the potential conflicts of interest arising
from such affiliations.
Researchers have studied this conflict of interest between investment banks and
VCs. Evidence shows that such affiliated investment bankers and VCs can exploit
unwary buyers and investors correctly foresee such conflicts and price the securities
accordingly. The first finding is consistent with the nave investor hypothesis of Kroszner
and Rajan (1994), which says that investors do not consider conflicts of interest or past
performance when investing. The second finding agrees with their rational discounting
hypothesis, which claims that investors are rational enough to correctly anticipate such
conflicts of interest and hence discount prices.
Gompers and Lerner (1999) empirically test the effect of such affiliations and dis-
tinguish between these two hypotheses. The authors use a four-pronged empirical anal-
ysis in which they examine the IPO offerings characteristics, post-issuance long-run
performance, underpricing, and information sensitivity of such conflicts of interest
ridden issues. Their evidence shows compelling support for the rational discounting
hypothesis.
Several later empirical papers report opposite results supporting the nave investor
hypothesis. For example, Li and Masulis (2004), who use a sample of 1,500 VC-backed
IPOs between 1993 and 2000, find that VC investments by investment bankers signifi-
cantly reduce IPO underpricing. This evidence is consistent with both the nave inves-
tor hypothesis and the certification hypothesis, which is explained in the next section.
Outside the United States, several researchers offer similar empirical results support-
ing the nave investor hypothesis including in France (Chahine and Filatotchev 2008),
Japan (Arikawa and Imadeddine 2010), and Italy (Pennacchio 2014). However, when
sophisticated investors are considered, especially institutional investors, studies show
that such investors prefer a lower offering price (Rock 1986; Ritter 1987; Tini 1988;
Ljungqvist, Jenkinson, and Wilhelm 2003) because such investments provide higher
returns when they are sold.
The Rol e of P riv at e E qu it y in I P Os 205

Although several issues regarding conflicts of interest in a VC-backed IPO are still
unsettled, the importance of the role played by VC firms in IPOs cannot be denied or
ignored. VCs expertise, professionalization, and economic impact before, during, and
after an IPO are crucial and are discussed in greater detail later in the chapter.

The Role of Venture Capital in Initial Public Offerings


Following Jensens (1993) influential paper, several studies examine the active role that
VCs play in managing, mentoring, and governing the firms in which they invest. Jensen
identifies three separate roles played by VCs in IPOs: certification, underpricing, and
monitoring.

C E R T I F I C AT I O N
VC involvement in a company often signals to other potential investors that the com-
pany is legitimate, which results in the ability to raise more money for the entrepre-
neurs. Such a phenomenon is known as the certification hypothesis. Megginson and
Weiss (1991) first test this hypothesis by conducting a comparative study between
VC-backed IPOs and non-VC-backed IPOs between 1983 and 1987. The authors doc-
ument that the involvement of VC firms acts as certification and lowers the total costs
of going public. Nahata (2008, p. 127), who confirms this result by analyzing VC invest-
ments between 1991 and 2001, notes: Companies backed by more reputable VCs by
IPO capitalization share (based on cumulative market capitalization of IPOs backed
by the VC), are more likely to exit successfully, access public markets faster, and have
higher asset productivity at IPOs.
Even long after a firm goes public, Celikyurt, Sevilir, and Shivdasani (2014) find
positive announcement returns and better operating performance associated with
appointments of the VC directors to the companys board of directors. This finding
offers evidence that involvement of high-quality VCs acts as a signaling mechanism.
To be an effective signal, the cost of such a signal should be high enough that less
reputable firms cannot replicate it as predicted by the models of Akerlof (1970)
and Spence (1973). Given the potential benefits of being associated with a repu-
table venture capitalist, start-up firms are likely to be more willing to take an offer
from a reputable VC firm. Using hand-collected data, Hsu (2004) provides empir-
ical evidence similar to the signaling models that entrepreneurs are willing to pay
by accepting a discount to get access to VCs with better reputations. Hsus findings
reinforce the view that VCs not only provide financial support to a start-up firm but
also lend extra-financial services such as access to the VCs information network
and managerial experience. Start-up firms are willing to pay for such extra-financial
support.
As discussed in the following paragraphs, other studies find evidence against the
certification hypothesis and support Gomperss (1993) grandstanding hypothesis. This
hypothesis states that new VC firms with little past performance tend to undertake
costly actions such as taking the company public earlier to signal that they can invest
successfully. According to the grandstanding hypothesis, VC-backed IPOs are likely
206 h o w p r i vat e e q u i t y w o r k s

to experience more underpricing than non-VC-backed IPOs and this underpricing is


more applicable for new VCs with shorter track records.
Gompers (1996) studies a sample of 433 VC-backed IPOs from 1978 through 1987
and finds that corporations backed by young VC firms are underpriced relative to those
backed by well-known and established VC firms. Lee and Wahal (2004) support the
grandstanding hypothesis using a sample of 6,413 IPOs from 1980 to 2000 while con-
trolling for the endogeneity of VC funding. The authors document a positive relation-
ship between the first-day returns and future fundraising.
Despite continuing debate on the certification role of VCs in IPOs, the certification
hypothesis appears more applicable for reputable VCs. Less reputable VCs or VCs with
shorter track record are more aggressive in taking a firm public and are willing to accept
a discount in an IPO. Thus, more support exists for the grandstanding hypothesis for
less reputable VCs.

UNDERPRICING
VCs play another important role in underpricing issued securities. Researchers use
underpricing as a proxy for the presence and extent of the certification effect. To fully
appreciate the part played by VCs in underpricing requires an understanding of under-
pricing and why it occurs with IPOs.
The IPO literature identifies three basic costs associated with issuing stock in an IPO:
(1) the underwriting spread, which is the difference between the proceeds the issuer
receives and the total amount raised in the offering; (2) out-of-pocket expenses, which
are exemplified by investment banking fees, legal fees, accounting expenses, printing
costs, and filing fees; and (3) the implicit cost of underpricing, which is the amount by
which an issue is underpriced.
Underpricing is defined as the returns experienced by newly listed firms at the clos-
ing of the first-day of their IPO (Habib and Ljungqvist 2001; Aggarwal, Krigman, and
Womack 2002). Researchers observe this phenomenon of underpricing in both the
United States and other financial markets (Boulton, Smart, and Zutter 2010). The find-
ings on IPO underpricing show that first-day returns averaged 16.9 percent for the U.S.
markets between 1960 and 2013 (Ritter 2014).
The current literature offers multiple explanations of this IPO underpricing puzzle.
Some models assume information asymmetry exists with an IPO issuance. For exam-
ple, Rock (1986) assumes that some investors are better informed of the true intrinsic
value of the offer than others including, for example, the issuing firm or even its under-
writing bank. This assumption gives informed investors an advantage over uninformed
investors, thereby imposing a winners curse on the uninformed investors.
Others believe that issuing firms have better knowledge of the present value or risk
of their future cash flows than do investors. Therefore, firms might use underpricing as
a signal of their quality or to differentiate themselves from low-quality firms. Assume
two types of firmshigh quality and low qualityappear indistinguishable to outside
investors. In such a case, the high-quality firms might use underpricing to communicate
their quality. Their degree of underpricing might be higher than in the low-quality firms
because such firms find mimicking behavior more costly. Ibbotson (1975, p. 264) first
contributed to this IPO signaling literature by stating that issuers underpriced to leave
The Rol e of P riv at e E qu it y in I P Os 207

a good taste in investors mouths. Later, Allen and Faulhaber (1989), Grinblatt and
Hwang (1989), and Welch (1989) also contribute to this theory. While these studies
present theoretical models with different assumptions and set-ups, the predictions of
the models are similar. These theoretical models predict that IPO underpricing gener-
ates greater investor interest. As a result, the low-priced new issues subsequently help
the newly listed firms to raise more funds by issuing high-priced seasoned shares. These
models suggest that IPO firms generally adopt a multiple issue strategy while they
decide on the IPO price and the quantity of share issues. Welchs research further shows
that an underpricing strategy mainly works for high-quality firms and can be a costly
strategy for low-quality firms to mimic.
Some researchers offer institutional explanations for IPO underpricing (Logue
1973; Ibbotson 1975; Jenkinson 1990; Beller, Terai, and Levine 1992; Keloharju 1993;
Kunz and Aggarwal 1994; Ljungqvist 1997). One such explanation is IPO price stabili-
zation, a service that underwriters provide when taking firms public. Price stabilization
by investment bankers is legal in many countries including the United States and it is a
common phenomenon.
Another institutional explanation first given by Logue (1973) and Ibbotson (1975)
is that firms deliberately underprice themselves to reduce the likelihood of future law-
suits from shareholders disappointed with the post-IPO performance of their shares.
Other studies refute this convincing explanation ( Jenkinson 1990; Beller et al. 1992;
Keloharju 1993; Kunz and Aggarwal 1994; Lee, Taylor, and Walter 1996; Ljungqvist
1997) and provide evidence from international financial markets in which firms under-
price themselves without a large risk of being sued.
The third institutional explanation for IPO underpricing is the tax advantage for
managers holding employee stock options. Here, the line of argument is that in the
United States, holders of employee stock options pay taxes in two steps. First, they pay
an income tax based on the strike price and the fair market value when they exercise their
options. Second, they pay capital gains tax on the difference between the fair market
value and the sale price when they eventually sell their stocks. Since the capital gains tax
can be deferred and is lower than the income tax, managers who hold employee stock
options have an incentive to have the fair market value as low as possible and hence
favor underpricing. Taranto (2003) finds results consistent with this argument.
Another branch in this literature focuses on the behavioral aspects of underpricing.
The most popular one is Welchs (1992) informational cascades argument. As Welch
notes, investors make their investment decisions sequentially, meaning that they disre-
gard their own private information and follow the actions of earlier investors believing
that the earlier investors have better information. Based on Welchs theory, early IPO
investors have more power to demand underpricing in return for committing to the
IPO, thus starting a positive cascade. Another behavioral explanation for IPO under-
pricing involves hot issue markets. According to Ibbotson and Jaffe (1975, p. 1027),
hot issue markets are periods in which the average first month performance (or after-
market performance) of new issues are abnormally high.
Certain time periods exist in which the mean return on all IPOs has been much
higher compared to other periods. For example, Ritter (1984) notes that during Jan-
uary 1980 and the following 15 months, the mean return on all IPOs in the United
States was 48.4 percent compared to 16.3 percent for the rest of the period between
208 h o w p r i vat e e q u i t y w o r k s

1977 and1982. Another hot period was the dot-com bubble of the 1990s. Such phe-
nomena can be attributed to general market sentiments during those periods. Finally,
Loughran and Ritter (2002) combine prospect theory (Kahneman and Tversky 1979)
and mental accounting (Thaler 1980, 1985) to argue that issuing companies fail to get
upset about leaving money on the table in the form of large first-day returns because
they attribute the wealth loss to underpricing with the wealth gain on retained shares as
prices jump in the aftermarket. Such complacent behavior also benefits the underwrit-
ers because investors can increase their chances of being allocated underpriced stock
both in the present and the future.
The following question remains: What role do VCs play in underpricing? Taking a
company public simply means a separation will occur between ownership and control.
Sometimes this becomes a difficult reality for the companys original owners or entre-
preneurs. Although they want to raise money in the public equity markets, owners may
have difficulty relinquishing some of their control. According to Grossman and Hart
(1980), the original owners prefer underpricing because it generates excess demand
and enables them to ration the investors so they can have greater ownership dispersion
and less threat of a hostile takeover.
Yet, VCs prefer less underpricing (Megginson and Weiss 1991) to ease information
asymmetry and to avoid leaving too much money on the table. Barry, Muscarella, Peavy,
and Vetsuypens (1990) examine VC-backed IPOs between 1978 and 1987. They docu-
ment that underpricing in VC-backed IPOs decreases with the number of VCs owning
shares in the equity, length of time that the lead venture capitalist has served on the
companys board, age of the VC firm, prior experience of the lead venture capitalist in
taking companies public, and the fraction of shares owned by VCs. Both studies provide
compelling evidence that markets react approvingly to the presence of VCs at the time
of IPOs.
Other studies find that IPOs that are backed by VCs and supported by top tier in-
vestment bankers have greater underpricing. For example, Kraus (2002) studies the
role of VCs in underpricing in the German financial markets. The author conducts a
comparative study between 124 VC-backed and 184 non-VC-backed firms that went
public on Germanys Neuer Markt between March 1997 and May 2001. Findings show
that VC-backed firms with top tier underwriters are underpriced more than their non-
VC- backed counterparts. Loughran and Ritter (2004) analyze IPOs from 1980 to
2003 and document the average first-day returns of the VC-backed IPOs are signifi-
cantly larger than those of non-VC-backed IPOs. Specifically, from 1980 to 1989, the
average first-day returns of the VC-backed IPOs were 8 percent versus 7.1 percent for
non-VC-backed IPOs. This difference was even greater in the consecutive periods, with
the highest difference occurring during the peak of the dot-com bubble in 19992000
when the average first-day return of VC-backed IPOs was 82.2 percent as compared to
38.5 percent for non-VC-backed IPOs.
In a more exhaustive and recent study on IPOs, Autore, Boulton, Smart, and Zutter
(2014) analyze a sample of 10,783 IPOs listed in 37 countries between 1998 and 2008.
They find that VCs in developed markets backed about 21 percent of IPOs as compared
to 5 percent in emerging financial markets. The authors further confirm the results of
Loughran and Ritter (2004) when they find that initial returns are 10.4 to 11.2 percent
higher if the IPOs are VC-backed.
The Rol e of P riv at e E qu it y in I P Os 209

Another more recent study based on a survey by Ernst & Young (2013) asks what
institutional investors think about PE- and VC-backed IPOs. Surveying more than
300 institutional investors across the globe, Ernst & Young concludes that institutional
investors have different perceptions on PE- and VC-backed IPOs. While 42 percent
of institutional investors believe that PE- and VC-backed IPOs are more expensive,
28 percent of institutional respondents believe that they are appropriately priced with
no discount. The remaining 30 percent believe that PE- and VC-backed IPOs are rela-
tively cheaper (i.e., they are more underpriced).
The survey also asks the institutional investors about their views on post-IPO perfor-
mance of PE- and VC-backed IPOs in comparison with non-PE/VC-backed IPOs. Of
the institutional investors responding, 40 percent think that PE- and VC-backed IPOs
perform worse than IPO offerings not backed by PE or VC firms; 30 percent believe
that they performed better; and the remaining 30 percent feel that no difference exists
in post-IPO performance between the PE/VC-backed IPOs and non-backed IPOs. As
far as the differences in regional perceptions worldwide with respect to the value created
in IPOs by PE- and VC-backing is concerned, the Ernst & Young (2013, p. 22) report
finds that:

Differences in regional perceptions regarding the value that PE/VC brings to


IPOs are quite startling. Those in Central and South America are overwhelm-
ingly positivein direct contrast to investors in another rapid-growth region, the
Middle East and Africa. In North America, the birthplace of PE and the VC in-
dustry, expectations of a better price performance post-listing are alarmingly low.

Hence, debate continues on the effect that VCs have on IPO underpricing. Current re-
search has not yet exhausted the important questions about this issue. The underpricing
results for VC-backed IPOs differ depending on the period of study and the geographi-
cal coverage. With the passage of the JOBS Act, the number of IPOs is expected to rise.
Thus, revisiting this issue should be of interest to researchers.

M O N I TO R I N G
VCs need to constantly monitor their investments because the presence of information
asymmetry causes agency costs such as adverse selection and moral hazard. Given that
entrepreneurs are insiders of the firm, they usually have more information than VCs and
hence can use this information to their personal advantage. Therefore, VCs have com-
pelling incentives to monitor the behavior of entrepreneurs. Much literature exists on
this monitoring role of VCs.
Gorman and Sahlman (1989) find indirect evidence of the monitoring role of VCs
based on a questionnaire mailed to 100 VCs in 1984. The survey asks the VCs how
much time they spend with their portfolio companies and how that time was distrib-
uted. The findings show that a venture capitalist typically spends 80 hours on-site and
30 hours on the phone per year with each company it manages. These findings are likely
the first survey evidence showing that VCs also act as active monitors. In a later study,
Barry et al. (1990) examine VC-backed IPOs from 1978 to 1987 and find evidence of
the monitoring role of VCs in the companies they back for an IPO.
210 h o w p r i vat e e q u i t y w o r k s

Lerner (1995) elicits a similar point on the monitoring role of VCs. He finds that
board representation of VCs increases around the time of chief executive officer
(CEO) turnover. The argument is that if VCs play a monitoring role, their board
of directors representation should increase when the need of monitoring and over-
sight is greatest. Using a random sample of 794 VC-financed companies, Gomp-
ers (1995) finds that with increasing expected agency costs, VCs also increase the
frequency of monitoring. Robbie and Wright (1998) highlight a key difference be-
tween traditional corporate finance and VC. The authors note that in traditional
corporate finance theory, monitoring management by shareholders is passive or in-
direct, whereas in VC the monitoring by the VCs is active and direct. Kaplan and
Strmberg (2000) also document direct evidence of the monitoring and advisory
role of VCs. They find that VCs not only shape and monitor the management team
before investing in a particular company but also expect to continue monitoring
post-investment. In at least half of their sample, VCs expect to play a role in recruit-
ing top management.
In a later empirical study, Baker and Gompers (2003) document a positive associa-
tion between VC-backed IPOs and the fraction of independent outsiders on the board,
which in turn increases monitoring. Boone, Field, Karpoff, and Raheja (2007) find a
similar result and show that VC presence during the IPO has a significant effect on the
independence of the board even 10 years after going public. Campbell and Frye (2009)
also find strong evidence that VC-backed firms have better governance structures
during their IPOs, enabling them to have higher levels of monitoring both at the time
of an IPO and four years afterward. Additionally, high-quality VC-backed firms have
higher levels of monitoring during an IPO than those backed by low-quality VC firms.
High-quality VC-backed firms also use equity-based compensation at a higher rate than
do firms backed by low-quality VC firms.
Hochberg (2012) studies the effect of VC involvement pre-IPO on the governance
path taken for some period after the IPO. Using both commercially available and hand-
collected data sources, as well as a selection model framework, the author conducts
three different sets of tests for comparing VC-backed and non-VC-backed firms. She
corroborates the monitoring role of VC firms by finding that the VC-backed firms have
more independent board structures, lower levels of management earnings, and a more
positive market reaction to adopting shareholder rights agreements.
A major challenge in studying the role of VCs is endogeneity. One might argue that
VCs tend to invest systematically in a certain firm. Therefore, establishing a causal re-
lationship in such studies is moot. Bernstein, Giroud, and Townsend (2014) try to cir-
cumvent this problem by using an exogenous variation in the VC involvement, which
results from introducing new airline routes that reduce the travel times for the VCs to
travel to their investee companies. Using this clever empirical design, the authors docu-
ment that the decline in travel time lowers monitoring costs, thereby resulting in greater
engagement and monitoring by the VCs further leading to a higher likelihood of an IPO
and better innovation.
The monitoring role of VCs, pre-, post- and during IPOs is well established. Fast-
paced globalization, technological advancements, and innovation drastically reduce
monitoring costs (Chemmanur and Fulghieri 2014).
The Rol e of P riv at e E qu it y in I P Os 211

Summary and Conclusions


This chapter presents detailed discussions on three main roles played by VCs in IPOs:
certification, underpricing, and monitoring. Newly listed firms appear to incur a lower
total cost of going public if reputable VCs back them. This is the certification role of
VCs. However, an ongoing debate exists on how VCs affect the underpricing of an IPO.
While no consensus exists on the role that VC firms play in underpricing, general em-
pirical results suggest that IPOs backed by more reputable VCs experience lower un-
derpricing. By contrast, VCs with shorter track records or worse reputations are more
willing to set the IPO price at a discount (i.e., higher underpricing) to attract more
investments in the future. Further, research shows that VCs play an important role in
monitoring start-up firms to mitigate information asymmetry and manage entrepre-
neurial behavior.
Despite the substantial body of research on the role of VCs in IPOs, some unan-
swered questions still remain. First, not much is known as to why VCs choose IPOs as
an exit route instead of alternative exit mechanisms. Another question is whether VCs
choose to take a firm public based on certain characteristics that are systematically dif-
ferent from firms that VCs do not take public, but instead use other methods to exit such
as trade sales and secondary sales.
Second, little research exists on the effect of VC exits once the companies become
public. If VCs play an important role during the IPO, their eventual exits should also
have substantial impact on the firms in which they invest. After VCs exit, firms must rely
on their own managerial skills to operate the business and decide on financing issues.
This questions how firms perform in the post-VC era, whether they can maintain good
operating performance, and whether they can effectively raise funds through secondary
equity offerings.
Third, some researchers such as Cable and Shane (1997) contend that in an entre-
preneur-venture capitalist set up, the venture capitalist acts as an agent for the entrepre-
neur who is the principal and not vice versa. The reasoning is that VCs are opportunistic
in nature and can sometimes underinvest in their investee companies if they find a better
opportunity elsewhere. Cable and Shane use the prisoners dilemma model to make
their case that VCs often act as agents rather than as principals. However, more research
needs to be done in this area.
Finally, researchers need to explore the roles played by VC firms during IPOs, which
are different from certification, underpricing, and monitoring. Some evidence shows
that VCs provide different kinds of support to their portfolio companies (Hellman and
Puri 2002). Analyzing how the roles of VCs have evolved and expanded over the years
is a topic of future research.

Discussion Questions
1. Explain the common exit routes taken by VC firms and identify the most popular
exit route.
2. Discuss the role of VCs in IPO underpricing.
212 h o w p r i vat e e q u i t y w o r k s

3. Describe the monitoring role of VCs in their investee companies.


4. Discuss the certification and grandstanding hypotheses in relation to IPO
underpricing.
5. Identify several unanswered research questions about the role of PE and VC firms in
IPOs.

References
Aggarwal, Rajesh K., Laurie Krigman, and Kent L. Womack. 2002. Strategic IPO Underpricing,
Information Momentum, and Lockup Expiration Selling. Journal of Financial Economics 66:1,
105137.
Akerlof, George. 1970. The Market for Lemons: Quality Uncertainty and the Market Mechanism.
Quarterly Journal of Economics 84:3, 488500.
Allen, Franklin, and Gerald R. Faulhaber. 1989. Signaling by Underpricing in the IPO Market.
Journal of Financial Economics 23:2, 303323.
Arikawa, Yasuhiro, and Gael Imadeddine. 2010. Venture Capital Affiliation with Underwriters and
the Underpricing of Initial Public Offerings in Japan. Journal of Economics and Business 62:6,
502516.
Autore, Don M., Thomas J. Boulton, Scott B. Smart, and Chad J. Zutter. 2014. The Impact of Insti-
tutional Quality on Initial Public Offerings. Journal of Economics and Business 73: MayJune,
6596.
Baker, Malcolm, and Paul A. Gompers. 2003. The Determinants of Board Structure at the Initial
Public Offering. Journal of Law and Economics 46:2, 569598.
Barry, Christopher B., Chris J. Muscarella, John W. Peavy III, and Michael R. Vetsuypens. 1990.
The Role of Venture Capital in the Creation of Public Companies: Evidence from the Going-
Public Process. Journal of Financial Economics 27:2, 447471.
Beller, Alan L., Tsunemasa Terai, and Richard M. Levine. 1992. Looks Can Be Deceiving: A Com-
parison of Initial Public Offering Procedures under Japanese and US Securities Laws. Law and
Contemporary Problems 55:4, 77118.
Bernstein, Shai, Xavier Giroud, and Richard Townsend. 2014. The Impact of Venture Capital Mon-
itoring: Evidence from a Natural Experiment. Working Paper, Stanford University, Massachu-
setts Institute of Technology (MIT), and Dartmouth College. Available at http://ssrn.com/
abstract=2341329.
Boone, Audra L., Laura Casares Field, Jonathan M. Karpoff, and Charu G. Raheja. 2007. The De-
terminants of Corporate Board Size and Composition: An Empirical Analysis. Journal of Fi-
nancial Economics 85:1, 66101.
Boulton, Thomas J., Scott B. Smart, and Chad J. Zutter. 2010. IPO Underpricing and International
Corporate Governance. Journal of International Business Studies 41:2, 206222.
Brav, Alon, and Paul A. Gompers. 1997. Myth or Reality? The LongRun Underperformance of
Initial Public Offerings: Evidence from Venture and Nonventure CapitalBacked Companies.
Journal of Finance 52:5, 17911821.
Cable, Daniel M., and Scott Shane. 1997. A Prisoners Dilemma Approach to Entrepreneur-Venture
Capitalist Relationships. Academy of Management Review 22:1, 142176.
Campbell, Terry L., II, and Melissa B. Frye. 2009. Venture Capitalist Monitoring: Evidence from
Governance Structures. Quarterly Review of Economics and Finance 49:2, 265282.
Celikyurt, Ugur, Merih Sevilir, and Anil Shivdasani. 2014. Venture Capitalists on Boards of Mature
Public Firms. Review of Financial Studies 27:1, 56101.
Chahine, Salim, and Igor Filatotchev. 2008. The Effects of Venture Capitalist Affiliation to Under-
writers on Short- and Long-term Performance in French IPOs. Global Finance Journal 18:3,
351372.
The Rol e of P riv at e E qu it y in I P Os 213

Chemmanur, Thomas J., and Paolo Fulghieri. 2014. Entrepreneurial Finance and Innovation: An
Introduction and Agenda for Future Research. Review of Financial Studies 27:1, 119.
Ernst & Young. 2013. Institutional Investor Survey. London: Ernst & Young.
Gompers, Paul A. 1993. The Theory, Structure, and Performance of Venture Capital. PhD disserta-
tion, Harvard University.
Gompers, Paul A. 1995. Optimal Investment, Monitoring, and the Staging of Venture Capital.
Journal of Finance 50:5, 14611489.
Gompers, Paul A. 1996. Grandstanding in the Venture Capital Industry. Journal of Financial Eco-
nomics 42:1, 133156.
Gompers, Paul, and Josh Lerner. 1999. Conflict of Interest in the Issuance of Public Securities: Ev-
idence from Venture Capital. Journal of Law & Economics 42 (April), 1831.
Gorman, Michael, and William A. Sahlman. 1989. What Do Venture Capitalists Do? Journal of
Business Venturing 4:4, 231248.
Grinblatt, Mark, and Chuan Y. Hwang. 1989. Signalling and the Pricing of New Issues. Journal of
Finance 44:2, 393420.
Grossman, Sanford J., and Oliver D. Hart. 1980. Takeover Bids, the Free-Rider Problem, and the
Theory of the Corporation. Bell Journal of Economics 11:1, 4264.
Habib, Michel A., and Alexander P. Ljungqvist. 2001. Underpricing and Entrepreneurial Wealth
Losses in IPOs: Theory and Evidence. Review of Financial Studies 14:2, 433458.
Hellmann, Thomas, and Manju Puri. 2002. Venture Capital and the Professionalization of Startup
Firms: Empirical Evidence. Journal of Finance 57:1, 169197.
Hochberg, Yael V. 2012. Venture Capital and Corporate Governance in the Newly Public Firm.
Review of Finance 16:2, 429480.
Hsu, David H. 2004. What Do Entrepreneurs Pay for Venture Capital Affiliation? Journal of Fi-
nance 59:4, 18051844.
Ibbotson, Roger G. 1975. Price Performance of Common Stock New Issues. Journal of Financial
Economics 2:3, 235272.
Ibbotson, Roger G., and Jeffrey F. Jaffe. 1975. Hot Issue Markets. Journal of Finance 30:4,
10271042.
Jenkinson, Tim J. 1990. Initial Public Offerings in the United Kingdom, the United States, and
Japan. Journal of the Japanese and International Economies 4:4, 428449.
Jensen, Michael C. 1993. The Modern Industrial Revolution, Exit, and the Failure of Internal Con-
trol Systems. Journal of Finance 48:3, 831880.
Kahneman, Daniel, and Amos Tversky. 1979. Prospect Theory: An Analysis of Decision under
Risk. Econometrica 47:2, 263292.
Kaplan, Steven N., Berk A. Sensoy, and Per Strmberg. 2009. Should Investors Bet on the Jockey or
the Horse? Evidence from the Evolution of Firms from Early Business Plans to Public Compa-
nies. Journal of Finance 64:1, 75115.
Kaplan, Steven N., and Per Strmberg. 2000. How Do Venture Capitalists Choose Investments?
Working Paper, University of Chicago.
Keloharju, Matti. 1993. The Winners Curse, Legal Liability, and the Long-Run Price Performance
of Initial Public Offerings in Finland. Journal of Financial Economics 34:2, 251277.
Kraus, Tilo. 2002. Underpricing of IPOs and the Certification Role of Venture Capitalists: Evi-
dence from Germanys Neuer Markt. Working Paper, University of Warwick.
Kroszner, Randall S., and Raghuram G. Rajan. 1994. Is the Glass-Steagall Act Justified? A Study of
the US Experience with Universal Banking before 1933. American Economic Review 84:4,
810832.
Kunz, Roger M., and Reena Aggarwal. 1994. Why Initial Public Offerings Are Underpriced: Evi-
dence from Switzerland. Journal of Banking & Finance 18:4, 705723.
Lee, Peggy M., and Sunil Wahal. 2004. Grandstanding, Certification and the Underpricing of Ven-
ture Capital Backed IPOs. Journal of Financial Economics 73:2, 375407.
Lee, Philip J., Stephen L. Taylor, and Terry S. Walter. 1996. Australian IPO Pricing in the Short and
Long Run. Journal of Banking & Finance 20:7, 11891210.
214 h o w p r i vat e e q u i t y w o r k s

Lerner, Josh. 1995. Venture Capitalists and the Oversight of Private Firms. Journal of Finance, 50:1,
301318.
Li, Xi, and Ronald W. Masulis. 2004. Venture Capital Investments by IPO Underwriters: Certifica-
tion, Alignment of Interest or Moral Hazard? Working Paper, Hong Kong University of Sci-
ence & Technology, and University of New South Wales, Available at http://ssrn.com/
abstract=559105.
Ljungqvist, Alexander P. 1997. Pricing Initial Public Offerings: Further Evidence from Germany.
European Economic Review 41:7, 13091320.
Ljungqvist, Alexander P., Tim Jenkinson, and William J. Wilhelm Jr. 2003. Global Integration in
Primary Equity Markets: The Role of US Banks and US Investors. Review of Financial Studies
16:1, 6399.
Logue, Dennis E. 1973. Premia on Unseasoned Equity Issues. Journal of Economics and Business
25:3, 133141.
Loughran, Tim, and Jay R. Ritter. 2002. Why Dont Issuers Get Upset about Leaving Money on the
Table in IPOs? Review of Financial Studies 15:2, 413444.
Loughran, Tim, and Jay R. Ritter. 2004. Why Has IPO Underpricing Changed over Time? Finan-
cial Management 33:3, 537.
Megginson, William L., and Kathleen A. Weiss. 1991. Venture Capitalist Certification in Initial
Public Offerings. Journal of Finance 46:3, 879903.
Nahata, Rajarishi. 2008. Venture Capital Reputation and Investment Performance. Journal of Fi-
nancial Economics 90:2, 127151.
Pennacchio, Luca. 2014. The Causal Effect of Venture Capital Backing on the Underpricing of Ital-
ian Initial Public Offerings. Venture Capital 16:2, 131155.
Preqin. 2014. Preqin Global Private Equity Report. Available at https://www.preqin.com/item/2014-
preqin-global-private-equity-report/1/8194.
Puri, Manju, and Rebecca Zarutskie. 2012. On the Life Cycle Dynamics of VentureCapital and
NonVentureCapitalFinanced Firms. Journal of Finance 67:6, 22472293.
Ritter, Jay R. 1984. The Hot Issue Market of 1980. Journal of Business 57:2, 215240.
Ritter, Jay R. 1987. The Costs of Going Public. Journal of Financial Economics 19:2, 269281.
Ritter, Jay R. 2014. IPO Database. Available at http://bear.warrington.ufl.edu/ritter/ipodata.htm.
Robbie, Ken, and Mike Wright. 1998. Venture Capital and Private Equity: A Review and Synthe-
sis. Journal of Business Finance & Accounting 25:56, 521570.
Rock, Kevin. 1986. Why New Issues Are Underpriced. Journal of Financial Economics 15:1, 187212.
Spence, Michael. 1973. Job Market Signaling. Quarterly Journal of Economics 87:3, 355374.
Taranto, Mark. 2003. Employee Stock Options and the Underpricing of Initial Public Offerings.
Working Paper, University of Pennsylvania.
Thaler, Richard. 1980. Toward a Positive Theory of Consumer Choice. Journal of Economic Behav-
ior & Organization 1:1, 3960.
Thaler, Richard. 1985. Mental Accounting and Consumer Choice. Marketing Science 4:3, 199214.
Tini, Seha M. 1988. Anatomy of Initial Public Offerings of Common Stock. Journal of Finance
43:4, 789822.
Welch, Ivo. 1989. Seasoned Offerings, Imitation Costs, and the Underpricing of Initial Public Of-
ferings. Journal of Finance 44:2, 421449.
Welch, Ivo. 1992. Sequential Sales, Learning, and Cascades. Journal of Finance 47:2, 695732.
13
Exit Strategies in Private Equity
DIDIER FOLUS
Professor of Finance, University of Paris Ouest Nanterre La Dfense
Associate Professor of Insurance, IFPASS, Paris

EMMANUEL BOUTRON
Associate Professor of Finance, University of Paris Ouest Nanterre La Dfense

Introduction
Private equity (PE) is a medium- to long-term financial investment provided in return
for equity in a company that is typically not listed on an exchange. However, the PE
fund itself is sometimes listed. The funds general partners (GPs) face a four-step task:
(1) screening investment opportunities, (2) investing the capital provided by limited
partners (LPs), (3) managing the portfolio of assets in order to create value, and (4)
harvesting the value in order to distribute it through exit channels.
Acting as an asset management company, a PE firm identifies and values potential in-
vestment opportunities on behalf of the financial partners. PE transactions are financed
using equity provided by LPs and in some cases debt raised from banks. GPs then ac-
tively manage the investment for the holding period, which is typically 10 years, seeking
to generate operational improvements in order to increase the companys value. Inves-
tors realize returns upon exiting the deal, which can be through a secondary buyout, the
sale of the portfolio company to another investor, or an initial public offering (IPO), in
which the portfolio company is listed on a public stock exchange for the first time.
Financial economics approaches to PE exit activities should consider information
asymmetry and risk management. Such analyses help explain the reasons behind choos-
ing certain exit routes and using insurance guarantees to avoid any liability of the GPs.
This chapter presents a comprehensive picture of PE exit strategies. The remainder
of the chapter is organized into five parts. The first part presents data on global exit ac-
tivities from institutional PE players. The second section describes available exit routes,
analyzing goals as well as advantages and disadvantages from the PE firms perspective.
The third section focuses on IPOs as an exit strategy and analyzes the practical factors
and theoretical arguments affecting this exit strategy. The final section offers a summary
and conclusions.

215
216 h o w p r i v a t e e q u i t y w o r k s

Global Exit Activities


The purpose of exit activities is to transfer value from the PE fund to its investors.
Excluding bankruptcies, PE firms typically engage in more than 1,000 exits annually.
Preqin (2014) reports that 1,348 PE-backed exits occurred in 2013 valued at $303 bil-
lion. Figure 13.1 shows the number of exit operations worldwide in 2013.
Figure 13.2 provides the number of global buyout-backed exits by region. North
America and Europe represent the most active areas in terms of PE exits and value. The
Asia-Pacific region has seen a growing share of exit activities and greater stability, mainly
due to a temporary fall in Asias emerging economies, primarily in China.

1,250 Figure 13.1 Number of Global Buyout


Exits, 2009 to 2013 This figure illustrates the
number of global buyout exits (vertical axis) by year
1,000
(horizontal axis). Source: Bain & Company (2014).

750
Buyout Exits

500

250

0
2009 2010 2011 2012 2013
Years

1,500

1,000
Buyout Exits

500
Rest of world
Asia-Pacific
Europe
North America
0
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013

Years

Figure 13.2 Number of Global Buyout-Backed Exits by Region, 1995 to


2013 This figure shows the number of PE fund exits (vertical axis) by year (horizontal
axis), represented by economic region, mainly North America, Europe, and Asia including
the Pacific. Source: Bain & Company (2014).
E xit S t rat e g ie s in P riv at e E qu it y 217

VA L U E G E N E R AT I O N B E F O R E E X I T
PE firms acquire businesses with the intent to exit at a higher equity value than what
they initially invested. PE investors expect a high internal rate of return (IRR) to
compensate them for the low liquidity of their investment. To increase the return
on equity GPs often use debt as leverage. For example, a leveraged buyout (LBO) is
where managers borrow money to buy a company. The PE fund manager makes crit-
ical assumptions on the potential exit multiple, which is typically earnings before
interest, taxes, depreciation, and amortization (EBITDA), the price-to-earnings
ratio (PER), and the maximum amount of debt the target company will assume.
Indeed, Kaiser and Westarp (2010, p. 35) point out that private equity firms often
direct management to focus on cash flow as the most important performance in-
dicator. Managers use three primary methods to increase the exit value realized
by the PE fund: free cash flow (FCF) generation, EBITDA/earnings growth, and
multiple expansion.

Free Cash Flow Generation


Managers can use FCF to first repay debt and then increase the companys equity
value. FCF represents the cash that a company can generate after investing the
money required to maintain both its asset base and production: the greater the
FCF, the higher is the theoretical value of the company. A PE firm also attempts to
increase the portfolio companys growth and margins, which in turn increases the
FCF generated by the respective portfolio company and results in a higher value
upon exit.

EBITDA/Earnings Growth
Increasing EBITDA, or earnings, consists of increasing sales, lowering overhead ex-
penses, or increasing gross margin. Assuming a fixed EBITDA multiple paid across time
(i.e., from investment entry to exit), a higher EBITDA increases the target companys
value, which accrues to the equity holders. For example, purchasing a target company
at 6x EBITDA and exiting at the same multiple, the EBITDA will directly affect the exit
price. If EBITDA grows from $40 million to $50 million, then the exit valuation is $300
million (6 x $50 million) compared with an entry valuation of $240 million (6 x $40
million), increasing the PE funds value by $60 million.

Multiple Expansion
Valuation multiples, such as the EBITDA/earnings ratio, are linked to the market envi-
ronment and a companys growth prospects, operating performance, and competitive
landscape. Therefore, a PE firm may try to time the macro-environment and the growth
trajectory of its portfolio company to sell at a higher multiple than it initially paid. Ide-
ally, the PE fund manager should purchase companies at a time when market multiples
are lower than usual and sell when multiples are higher than usual. The manager should
also enhance a companys exit multiple by shifting the company to a more attractive
mix of business lines during the investment period, typically focusing the company
on highly profitable business, allowing it to be potentially sold for a higher EBITDA/
earnings multiple than for which the PE firm acquired it. Whatever the chosen method,
218 h o w p r i v a t e e q u i t y w o r k s

exiting activities imply the sale of financial instruments by the PE firm to transfer the
generated value to the targets acquirer.

TYPES OF FINANCIAL INSTRUMENTS


PE fund managers invest in different financial instruments. When the time to exit ar-
rives, the financial instrument should reflect the value creation immediately before the
sale. Managers commonly use the following instruments.

Equity shares represent ownership interest carrying voting rights and are the pri-
mary financial instrument used by PE funds.
Preferred shares entitle investors to a fixed dividend, generally do not carry voting
rights, and have priority over common shares in the distribution of assets in the
event of liquidation.
Hybrid instruments include securities such as convertible bonds, exchangeable
bonds, and redeemable bonds.
Warrants are long-term call options that entitle the holder to buy the underlying
stock of the issuing company at a fixed strike price until or at an expiration date.
Depository receipts are securities such as American depository receipts (ADRs)
issued by a bank against underlying shares of a company incorporated in a foreign
country, but denominated in the investors currency.

Besides these instruments, the PE fund can use, directly or through its portfolio
companies, debt securities including bonds, notes, or bills, payments of fixed or floating
interest rates, or bank and shareholder loans as leverage to manage the investment rate
of return. The sale or ownership of the securities, mainly equities, requires different exit
routes.

Traditional Exit Routes


According to Povaly (2006), traditional exit routes for PE investments are trade sales,
secondary buyouts, and IPOs. An exit can be full or partial. Recent exit data suggest that
dividend recapitalization is not another exit method.
Figure 13.3 shows the evolution of the different exit strategies chosen by PE funds
between 1995 and 2013. Trade sales represent more than half of the exits, followed by
secondary buyouts (sponsor-to-sponsor), and then IPOs.
Practical reasons and theoretical arguments, which will be discussed in the next sec-
tion, can help to explain such an evolution. Also, data show that exit activity is highly
sensitive to economic conditions. For example, the financial crisis of 20072008 af-
fected both the number of exit deals and the chosen exit strategies. For example, very
few IPOs occurred during this period. In the case of a portfolio companys bankruptcy,
the subsequent exit strategy is to write off the equity, totally or partially, depending of
the capital reduction level. Capital reduction is a court-regulated process of distribut-
ing capital and retained earnings. A partial capital reduction can mean the redemption
of the shares value and their cancellation, or a reduction of the face value of the shares.
E xit S t rat e g ie s in P riv at e E qu it y 219

1,500

1,000
Buyout Exits

500

IPO
Sponsor-to-sponsor
Strategic
0
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
Years

Figure 13.3 Number of Global Buyout-Backed Exits by Channel, 1995 to


2013 This figure shows the number of PE funds exits (vertical axis) by year (horizontal
axis) by traditional channels: trade sale (strategic), secondary buyout (sponsor-to-
sponsor), and IPO. Source: Bain & Company (2014).

Figure 13.4 illustrates the recent global evolution of PE fund exit types, in terms of
the number of exits and their monetary amount. Trade sales and secondary buyouts
(sales to another GP) represent the most prominent exit strategies. During the financial
crisis of 20072008, both the number and global value of exits declined and restructur-
ing activities increased, Restructuring can involve making managerial changes to reduce

400 140
350 120
Aggregate Exit Value ($bn)

300
100
250
No. of Exits

80
200
60
150
40
100
50 20

0 0
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4

2006 2007 2008 2009 2010 2011 2012 2013


IPO Restructuring Sale to GP Trade Sale Aggregate Exit Value ($bn)

Figure 13.4 Private Equity-Backed Exits by Type, 2006 to 2013 This figure
shows the number (left scale) and amount (right scale) of PE fund exit types. Source:
Preqin (2014).
220 h o w p r i v a t e e q u i t y w o r k s

costs or to shut down unprofitable units, increasing the operating cash flow. Restructur-
ing often precedes a sale, merger, or an IPO.

TRADE SALE
A financial sponsor may realize gains in a portfolio company investment via a trade
sale to a strategic acquirer. The strategic buyer is usually a non-PE firm, and the acqui-
sition is in the buyers strategic interest whether the motive involves market growth,
patents, innovative products, or synergies. The acquirer can be the portfolio com-
pany itself, repurchasing its own shares from the PE firm or fund. The buyer expects
a greater competitive advantage and market share in its respective industry, intending
to hold the acquisition over the long term. That is, the buyer often agrees to pay the
value of strategic options embedded in the target price, hoping for a higher future op-
erating cash flow from the target, and thus paying a higher present value for it. There-
fore, the trade sale usually commands the highest sale price, allowing the PE vendor
immediate liquidity. Another advantage of a trade sale is that the negotiations take
place with a single buyer, allowing for a quicker and more efficient process that is not
subject to the regulatory restrictions applicable to public transactions such as an IPO.
For these reasons, selling to a strategic buyer is generally the preferred exit option for
a PE investor.
Yet, a trade sale implies potential difficulties. For example, the companys manage-
ment may resist the trade sale due to the risk of its own replacement. Additionally, a
trade sale exposes the portfolio company to the risk of confidential business informa-
tion being divulged during the negotiation process.
An illustrative case of a strategic trade sale involves Koch Industries and Goldman
Sachs acquiring Flint Group from CVC Capital Partners (Zephyr Database 2014).
CVC Capital Partners Ltd (CVC) is a PE firm that manages an equity portfolio. Part
of the portfolio involves an investment in Flint Group (Germany and Luxembourg),
a computer ink and image transfer product manufacturer. On June 11, 2012, a report
noted that CVC wanted to sell Flint Group. A report on April 8, 2014, indicated that
Koch Industries Inc. and the PE unit of Goldman Sachs Group Inc. were close to reach-
ing an agreement to acquire the Flint Group from CVC. According to an announcement
on April 10, 2014, Koch Equity Development LLC, a subsidiary of Koch Industries, and
Goldman Sachs Merchant Banking Division, through a newly formed entity, would ac-
quire a 100 percent ownership in Flint Group from funds advised by CVC. This $3bil-
lion deal is expected to close by the second half of 2014.
The rationale of such a strategic sale is the targets ability to deliver future value. On
April 10, 2014, Matthias Hieber, who heads Corporate Equity Investing at Goldman
Sachs Merchant Banking Division, said:

We believe Flint Group is uniquely positioned to capture growth in its attrac-


tive printed packaging markets while at the same time continuing to benefit
from strong and resilient performance of its print media business. With a sig-
nificantly improved capital structure, Flint Group is best positioned to pursue
its ambitious growth plans to further strengthen its market leading positions.
(Zephyr Database 2014)
E xit S t rat e g ie s in P riv at e E qu it y 221

INITIAL PUBLIC OFFERING


An IPO is the method in which a companys shares get listed on the stock market for the
first time. This strategy offers a PE firm a way to exit by selling the shares of a company
in its portfolio. IPOs are a popular exit route for PE providers. When the stock market is
bullish, this method is likely to enable the vendor to realize the highest return on its in-
vestment. This environment makes IPOs suitable for large portfolio companies or high-
performing companies. However, an IPO involves high transaction costs, notably due
to legal restrictions and the market supervisors rules. For example, regulatory require-
ments and restrictions come from the Securities and Exchange Commission (SEC) in
the United States and the Autorit des Marchs Financiers in France.
Additionally, if the PE fund wants to fully exit the portfolio company, potential
public investors might view this full exit as a lack of confidence in the future prospects
of the business. Furthermore, the terms of the IPO may prohibit the financial sponsor
from exiting some or all of its position for a period of time called a lock-up period. The
likelihood of a discounted price for the IPO represents a potential opportunity cost.
These factors make an IPO a lengthy and expensive process. A latter section of this chap-
ter focuses on PE-backed IPOs, presents data, and discusses advantages and disadvan-
tages in more detail.

SECONDARY BUYOUT
A financial sponsor sells a portfolio company to another financial sponsor in a buyout
transaction known as a secondary buyout or sponsor-to-sponsor buyout. The transaction
may or may not be leveraged. One possible rationale for this type of exit is that the finan-
cial sponsor and current management team believe that a larger financial sponsor can
add value to the portfolio company as it progresses to the next development stage. Alter-
natively, a financial sponsor may decide to sell the company to another financial sponsor
if it has reached its minimum investment time period and has already created a high rate
of return on its initial investment or if it is too close to the date on which it will lose the
ability to call uninvested capital known as dry powder. Other potential benefits of selling
to another PE firm include increased flexibility in the structure of the sale. For exam-
ple, the vendor could maintain partial ownership and enable the company to continue
conducting its business with the intent of long-term growth. Partial ownership can be a
means of shortening a transactions lifetime, which has become a priority for PE firms.
Sometimes, a PE firm is unwilling or unable to continue financing a business, even
though the company might not yet be ready for a trade sale or an IPO. In that case, sell-
ing the company to another PE firm that sees potential in further developing the com-
pany might be a cost-efficient solution. This method can also be used to solve a conflict
between the PE investor backing the company and the company management. A sec-
ondary buyout offers the advantages of an immediate and complete exit and is carried
out even faster than a trade sale or an IPO.
Ares Management and Ontario Teachers Pension Plan, which sold a majority own-
ership in AOT Bedding Super Holdings, serves as an illustration of a secondary buyout
(Zephyr Database 2014). The Ares Private Equity Group manages a portfolio of equity
investments through five equity funds. That portfolio includes ownership in the National
222 h o w p r i v a t e e q u i t y w o r k s

Bedding Company LLC, Hoffman Estates, an Illinois-based manufacturer of the Serta


brand of mattresses. On June 8, 2012, Ares Management LLC and The Ontario Teach-
ers Pension Plan Board hired Barclays PLC to look into selling National Bedding Com-
pany LLC. On August 5, 2012, Advent International Corporation, a PE firm, agreed to
acquire a majority ownership in AOT Bedding Super Holdings LLC, a Georgia-based
mattress manufacturer holding company and parent of National Bedding and Simmons
Bedding. Ares Management and Ontario Teachers Pension Plan retained ownership in
the company. The transaction, which was originally valued at approximately $3 billion,
was completed during October 2012. Advent Managing Director Steven Collins states,
Our strategy going forward is to continue to grow the businesses by ensuring that Serta
and Simmons remain focused on delivering high quality mattresses that help people
sleep better (Zephyr Database 2014).

L E V E R A G E D D I V I D E N D R E C A P I TA L I Z AT I O N
Dividend distribution to shareholders is the traditional way for a company to repatriate
profits. If a companys capital is divided into different classes of shares, the company
can distribute dividends only to a specific class of shareholders. Such a specific class of
shares can be issued from the beginning of the PE investment, offering investors a partial
exit through the dividend payment. Leveraged recapitalization is a partial exit method in
which a PE portfolio company issues new debt in order to pay a special dividend to
private investors or shareholders. The company raising money usually completes this
process by borrowing from a bank or issuing bonds. The amount raised is then used to
repurchase the companys own shares from the investor. This method is an alternative
to selling the companys equity.
For example, in 2010, Nordenia International AG, a German manufacturer of flexible
packaging, issued a 280 million bond that in addition to the cash on hand it used to
refinance an existing debt and fund a 195 billion dividend, paid to its sponsor Oaktree
Capital, a PE firm. This approach enabled the PE firm to receive more cash than the
amount it invested in the portfolio company since acquiring majority ownership in 2006.
The main advantage associated with leveraged dividend recapitalizations is that the
PE fund remains in partial control of the target. It still receives payment and possible
tax benefits compared to other types of exits. Yet, such an exit route may result in over-
leveraging, which can eventually lead to financial distress with associated agency costs.
Increased leverage limits the flexibility of the portfolio companys operations and can
affect its value: limited flexibility may reduce research and development (R&D) expen-
ditures (Long and Ravenscraft 1993), or can lead to missing new investment opportuni-
ties, also reducing a companys value.

Initial Public Offerings as an Exit Route for Private


Equity Firms
The market for PE-backed IPOs is well adapted to highly profitable portfolio compa-
nies with growth opportunities. However, the quality of the firm, assessed by its equity
story, is not the sole factor that determines a companys decision to exit through an IPO.
E xit S t rat e g ie s in P riv at e E qu it y 223

The economic and market environments are of great importance, and the IPO channel
seems to be the most profitable exit solution to reward LPs.

O U T L O O K O F T H E P R I VAT E - E Q U I T Y - B A C K E D I N I T I A L P U B L I C
OFFERING MARKET
According to the Zephyr Database (2014), 543 PE-backed IPOs (partial exits and full
exits) took place from January 2005 to June 2014 on the primary stock market amount-
ing to $197.5 billion worldwide. The decision to exit through an IPO is cyclical and
driven by market sentiment.
Table 13.1 and Figure 13.5 show the yearly number of PE-backed IPOs and the ag-
gregate deal value from January 2005 to June 2014. As Table 13.1 shows, PE-backed
IPO activity surged in 2005 before plunging and reaching a low in 2008 as a result of the
financial crisis of 20072008. The aggregate deal value of PE-backed IPOs fell sharply
from $27.8 billion in 2006 to $903.3 million in 2008. The value of PE-backed IPOs
rebounded gradually in 2010 but recovered its 2006 pre-crisis level only in 2012 with
a total value of $26.2 billion. The revival of the IPO channel as a PE firm exit strategy
soared in 2013 due to favorable stock market conditions: 91 IPOs in 2013 versus 50
IPOs in 2012. At the end of June 2014, the number of PE-backed IPOs was almost equal
to number of IPOs in 2013, suggesting the return of IPOs as an attractive way for PE
firms to liquidate assets.
Table 13.1 and Figure 13.6 report the S&P 500 index as a proxy for the market senti-
ment. Not surprisingly, the number and the aggregate deal value correlate with the stock
market performance: the more positive the market environment, the more successful is
the IPO process. Worldwide economic recovery and accommodative monetary poli-
cies, particularly in the United States and later in Europe, helped PE-backed IPOs to get
back on track.
Table 13.2 and Figure 13.7 report the breakdown of portfolio companies accord-
ing to their home country. They shed light on the main geographic areas where those
firms went public. Between January 2005 and June 2014, either the Americas region or
Europe, Middle East, and Africa (EMEA) region took the lead of the PE-backed IPOs
market. Western Europe and North America account for 44.9 and 34.3 percent, respect-
fully, of the value of all IPOs that occurred from January 2005 to June 2014. After the
financial crisis of 20072008, the North American market recovered more rapidly due
to a stronger economy and the quantitative easing steps taken by the U.S. Federal Re-
serve Bank in 2008, 2010, and 2012. European firms fueled the PE-backed IPOs market
between January 2013 and June 2014. Meanwhile, the number of American companies
that exited through an IPO slowed.
The Oceania, Far East, and Central Asia region has grown in importance from Janu-
ary 2005 to June 2014. A moratorium on IPOs imposed by the China Securities Regu-
latory Commission (CSRC) since October 2012 prevented Chinese companies from
going public. The end of the moratorium in December 2013 should lead to a sharp in-
crease in 2014 and the coming years in that region.
Table 13.3 and Figure 13.8 give a breakdown of PE-backed IPOs deal values accord-
ing to their industry sector. From January 2005 to June 2014, 63 percent of the total
value of PE-backed IPOs came from two industrial sectors: manufacturing and services.
Table 13.1 Private-Equity-Backed IPOs and Stock Market Evolution, 2005 to June 2014
Years Dec 2005 Dec 2006 Dec 2007 Dec 2008 Dec 2009 Dec 2010 Dec 2011 Dec 2012 Dec 2013 Jun 2014 Total
Volume of PE-backed IPOs 74 75 50 12 15 43 43 52 91 88 543
Aggregate deal value 15,784 27,841 18,237 1,039 4,845 12,346 20,160 26,212 29,884 41,176 197,525
(million USD)
S&P 500 index 1,248 1,418 1,468 903.25 1,115 1,272 1,258 1,426 1,848 1,925

Note: This table reports the yearly number of PE-backed IPOs and the aggregate deal value, which is assessed at the completion of the deal. It also shows the evolution of
the S&P 500 index between December 2005 and June 2014.
Source: Zephyr Database (2014).
E xit S t rat e g ie s in P riv at e E qu it y 225

45.000 100
40.000 90
35.000 80
30.000 70
60
25.000
50 Aggregate deal value
20.000
40 (mil USD)
15.000 30
10.000 Volume of PE-backed IPOs
20
5.000 10
0.000 0
De 005
De 006
De 007
De 008
De 009
De 010
De 011
De 012

4
Ju 13
01
0
c2
c2
c2
c2
c2
c2
c2
c2
c2
n2
De

Figure 13.5 Volume and Aggregate Value of PE-Backed IPOs: January 2005
to June 2014

45,000 2,500
40,000 2,000
35,000 1,800
30,000 1,600
25,000 1,400 Aggregate deal value
20,000 1,200 (mil USD)
15,000 1,000 S&P 500 INDEX
10,000 800
5,000 600
0 400
Dec Dec Dec Dec Dec Dec Dec Dec Dec Jun
2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

Figure 13.6 PE-Backed IPOs (Volume) and Market Conditions: January


2005 to June 2014

The biggest IPO of the decade is Facebook, Inc. whose proceeds were more than $16
billion. Among the top 10 deals of that period, five firms belong to the manufacturing
industry reaching a collective value of about $13 billion: Japan Display Inc., Tognum
AG, Evonik Industries AG, Petroplus Holdings AG, and Pandora A/S.
Of the PE funds that fully or partially sold their holding through an IPO, five are
American (Carlyle Group LP, Warburg Pincus LLC, Kohlberg Kravis Roberts & Com-
pany LP, Goldman Sachs Group Inc., and Blackstone Group LP), and three are Euro-
pean (3i Group plc, CVC Capital Partners Ltd, and Earlybird Venture Capital GmbH
& Co. KG).

THE INITIAL PUBLIC OFFERING PROCESS


PE firms prepare for an IPO far in advance since timing is critical. Usually an IPO re-
quires five steps and lasts from 6 to 12 months after the companys board decides to
initiate the IPO.
Table 13.2Breakdown of the Aggregate Deal Value of Private-Equity-Backed IPOs by Region of the Portfolio Companies,
2005 to June 2014
Country 2005 2006 2007 2008 2009 2010 2011 2012 2013 June 2014 Total
Western Europe 9,232 16,355 11,295 131 305 7,191 5,146 1,228 16,546 21,344 88,773
Eastern Europe 31 97 0 0 1 0 45 0 0 0 173
Middle East 0 0 8 0 0 0 0 0 138 0 146
Africa 0 0 47 791 0 79 0 0 0 0 917
EMEA 9,263 16,452 11,350 922 306 7,270 5,191 1,228 16,684 21,344 90,009
Far East and Central Asia 787 3,607 0 36 945 1,759 68 3,339 1,629 5,425 17,596
Oceania 61 10 0 0 1,820 0 217 0 1,690 1,291 5,089
PACA 848 3,617 0 36 2,765 1,759 285 3,339 3,319 6,716 22,685
North America 5,563 7,264 5,555 81 1,774 1,310 10,081 19,171 8,205 8,720 67,724
South and Central America 111 507 1,331 0 0 2,007 4,603 2,474 1,676 4,396 17,106
AMERICAS 5,674 7,771 6,886 81 1,774 3,317 14,684 21,645 9,881 13,116 84,830
Total 15,84 27,841 18,237 1,039 4,845 12,346 20,160 26,212 29,884 41,176 197,525

Note: This table shows the regions where PE-backed IPOs took place between 2005 and June 2014. EMEA stands for Europe, Middle East, and Africa; PACA stands for
Far East and Central Asia Oceania; and AMERICAS stands for North America and South and Central America.
Source: Zephyr Database (2014).
E xit S t rat e g ie s in P riv at e E qu it y 227

100%
90%
80%
70%
60%
50% PACA
40% EMEA
30% AMERICAS
20%
10%
0%
Dec Dec Dec Dec Dec Dec Dec Dec Dec June
2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

Figure 13.7 Breakdown of the Aggregate Value of PE-Backed IPOs by the


Region of Portfolio Company: January 2005 to June 2014

The first step is to choose external partners including accountants, independent au-
ditors, and legal advisors who will help the firm review its corporate governance and
structure to meet the listing criteria.
In the second step, these external partners exercise due diligence while the under-
writer conducts an initial share valuation and draws up documents such as the listing
prospectus and notice based on research and publicity guidelines, and possibly the
help of a financial communication agency.
During the third step, the IPO candidate begins the registration process with the fi-
nancial authority and submits the prospectus and the registration form. The pro-
spectus unveils business and financial information about the firm and its prospects.
It sheds light on the main risks that future investors could face. The prospectus also
informs potential investors about the operation itself such as the type of shares, offer
price, and timetable of the operation.
After financial authority approval, the fourth step involves the firm publicly an-
nouncing its IPO, holding pre-marketing analyst road shows, and establishing the
range of the share price.
In the fifth and final step, the IPO candidate holds group presentations and one-on-
one meetings to market the shares as the underwriter assesses the demand for
shares and determines their price through the book-building process. Upon selling
shares to the public and/or to institutional investors, the firm is now publicly
traded.

Firms initiate IPOs for three main reasons. First, through the IPO, a firm gains
attention from potential customers, suppliers, investors, and any other third parties.
Second, going public helps a firm fund its growth by providing long-term capital and
diversified financial resources (e.g., equity, convertible debt, and straight bonds).
Third, an IPO is a way for shareholders to sell their equity ownership and fully or
partially exit the company. Venture capitalists (VCs) or PE funds may be among these
existing shareholders.
Table 13.3Breakdown of the Yearly Aggregate Deal Value at the Completed Date per Portfolio Company by Industrial Sector,
2005 to June 2014
Breakdown of the Deal Yearly Value Dec Dec Dec Dec Dec Dec Dec Dec Dec June Total
(Completed Date) per SIC Code 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
(Target)
% % % % % % % % % % %
Agriculture, forestry, and fishing 0 0 0 0 0 0 0 0 0 1 0
Mining 8 3 4 0 0 5 3 1 6 7 5
Construction 0 1 0 0 0 1 0 3 2 0 1
Manufacturing 38 44 34 11 39 39 17 7 29 19 27
Transportation, communications, 10 1 17 76 10 3 14 5 13 5 9
electric, gas, and sanitary services
Wholesale trade 3 3 1 0 0 9 1 0 1 2 2
Retail trade 7 10 0 0 44 6 12 2 2 16 9
Finance, insurance, and real estate 12 21 2 0 0 10 4 10 25 11 12
Services 21 17 42 13 8 26 48 73 21 39 36

Source: Zephyr Database (2014).


E xit S t rat e g ie s in P riv at e E qu it y 229

120%
Services

100% Finance, Insurance, and Real Estate

80% Retail Trade

Wholesale Trade
60%
Transportation, Communications, Electric,
Gas, and Sanitary Services
40%
Manufacturing

20% Construction

Mining
0%
Dec 2005 Dec 2006 Dec 2007 Dec 2008 Dec 2009 Dec 2010 Dec 2011 Dec 2012 Dec 2013 Jun 2014

Figure 13.8 Breakdown of the Yearly Deal Value at the Completed Date
per Industry Sector This figure reports the industry sectors where the portfolio
companies exited through an IPO. IPOs most commonly occur in the services and
manufacturing sectors. Source: Zephyr Database (2014).

A R G U M E N T S F AV O R I N G A N I N I T I A L P U B L I C O F F E R I N G
AS AN EXIT ROUTE
PE refers to either investments in early stage companies or financing mature firms. The first
category is called venture capital (VC) investment and the latter is called buyout (BO) in-
vestment. These two types of PE funds provide both financing to the target company and
their operational and strategic knowledge (Metric and Yasuda 2011). The target company
may benefit from the professional network of the PE fund managers to develop its busi-
ness. PE funds try to acquire a partial ownership of the target company at the lowest price
and to sell with the highest premium. The shorter this cycle, the higher is the return.
Using a sample of North American VC investments from 1991 to 2004, Cumming
and Johan (2010) report an average time to IPO from the first investment of 2.45 to
2.95 years. This duration is a function of the projected marginal value added provided
by the venture capitalist to the portfolio companies and the related projected marginal
cost (Cumming and MacIntosh 2001).
Aghion, Bolton, and Tirole (2004) identify three motives for VCs to exit an invest-
ment. First, VCs consider the exit as a way to reward their LPs after a few years. Second,
they cannot afford to finance portfolio companies in their late stage because of their
limited financial resources. Third, VCs do not have the competencies to manage these
maturing firms (Gompers and Lerner 1999; Kaplan and Strmberg 2003, 2004).
Different exit options coexist, namely: IPO, trade sales, secondary sales, and write-
offs. Gao, Ritter, and Zhu (2013) quantify the percentage of IPOs and trades sales from
1990 to 2012. They show that IPOs were the most frequent exit strategy (70 percent
of VC exits) at the beginning of the 1990s in the U.S. market, but by the turn of the
century trade sales grew to become the most common exit strategy (80 percent for VC
exits). Since then, IPOs have become less common with fewer than 20 percent of VCs
using this strategy to exit. According to the authors, this trend is the consequence of the
increasing pace of technological change and the quest for economies of scope that firms
can better achieve through a trade sale than an IPO.
230 h o w p r i v a t e e q u i t y w o r k s

Nonetheless, the IPO channel appears to be the most profitable way to exit an invest-
ment. Smith, Pedace, and Sathe (2011) find that IPOs are a key factor in explaining the
performance of a U.S. VC fund even though trade sales are also important. Indeed, their
estimated contribution of trade sales to the funds internal rate of return (IRR) is around
three-fourths that of IPOs.
Schmidt, Steffen, and Szabo (2010) obtain the same results using a worldwide data-
set of more mature portfolio companies: management buyouts (MBOs), management
buyins (MBIs), and leveraged buyouts (LBOs). PE companies sold these buyouts in-
vestment through an IPO between 1990 and 2005. The authors show that IPOs are the
second most frequently used vehicle after a sale either to a strategic investor or to another
PE fund but before a write-off. An IPO is the most profitable exit channel with an IRR of
111 percent, compared with an IRR of 49 percent for sales and 100 percent for write-
offs. According to Schmidt et al., the probability of an exit through either an IPO or a
sale increases with the holding period and favorable economic and stock market envi-
ronments. The IPO channel boosts the achieved IRR during good economic conditions.
Cumming and MacIntosh (2003) address the question of a partial as opposed to a
full exit. A full exit implies the sale of all of the venture capitalists holdings within one
year of the IPO while a partial exit involves the sale of only a part of the venture capital-
ists holdings. The authors assume that the higher the degree of asymmetric information
between the venture capitalist and the public investors, the higher is the probability
of a partial exit. The ownership retention signals the quality of the underlying firm. By
partially selling its ownership, the venture capitalist reveals that the potential for growth
still exists. Their results support their asymmetry hypothesis in the Canadian market
but not in the U.S. market.
Schmidt et al. (2010) look at the factors affecting the decision to exit via an IPO.
They conclude that the main determinants are the holding period and the stock market
environment. The longer the holding period, the higher is the probability of an exit via
an IPO as opposed to write-offs.
Lets turn now to the venture-backed firm at the date of the IPO. According to Meg-
ginson and Weiss (1991), venture-backed IPOs experience lower underpricing than
non-venture-backed IPOs on the U.S. stock market. These results are consistent with the
monitoring and certification roles of VC that reduce the risk and the asymmetry between
the firm and potential investors. However, Barry, Muscarella, and Peavy (1990) report a
lower underpricing only for experienced VCs. Bradley and Jordan (2002) find a higher
initial underpricing for venture-backed IPOs than for non-venture-backed IPOs, particu-
larly during periods of hot issue. Young VC funds may seek to take ventures public early
to build their reputation, even if it reduces their return due to the undervaluation of the
venture in order to attract new financial resources. This grandstanding hypothesis seems
to be verified during periods of high activity on the primary market (Gompers 1996).

Factors Affecting the Exit Route Choice


In deciding which exit strategy to pursue, investors should consider several variables:
(1) the general economic perspective, (2) relative performance of PE investments
versus other asset classes, (3) bullish or bearish stock markets, (4) commitments of PE
E xit S t rat e g ie s in P riv at e E qu it y 231

investors (reimbursing a debt and LBO financing), and (5) quality of the PE investor
(e.g., investment fund, GP of the PE firm, and LPs).

MACROECONOMIC CONDITIONS
Macroeconomic conditions affect exiting activities and largely determine the market
timing of divestitures. Growing public equity markets facilitate IPOs due to investor
appetite for risky securities. Persistently low interest rates and accommodating debt
markets facilitate leverage dividend recapitalizations or secondary buyouts using lever-
age. In a period of tight credit, a bank might be unwilling to lend financing to a potential
acquirer, thereby reducing the selection of viable exits for the PE firm.
The legal, tax, and regulatory environment also play a role in the exit route deci-
sion. Typically, selling to a strategic buyer is the most desirable option because strategic
buyers generally pay higher multiples for a business than financial buyers. Some con-
sider exit by IPO as a desirable way to exit because of the possibility to sell high-priced
shares. However, the vendor has to wait for the public offering to be completed and
the subsequent lock-up period to expire before achieving liquidity. Information asym-
metry between both parties hinders the negotiation with a strategic buyer. That is, the
potential acquirer worries about the breach of a vendors obligations, especially in terms
of taxes or financial statements.

U S I N G R E P R E S E N TAT I O N A N D WA R R A N T Y I N S U R A N C E
As an asset seller, a PE firm can face costly liabilities due to inaccuracies in representa-
tions and warranties made by itself or the target company. The target company buyer
can be left without the ability to recover losses and the seller can be forced to return a
portion of the purchase price. One way to prevent this occurrence is by adding indem-
nity provisions to the deal contract; the vendor frequently pays a deposit to collateralize
its contractual obligations. Because this process is costly, an adapted third party war-
ranty is an efficient solution.
Representation and warranty (R&W) insurance helps protect both the buyer and
seller involved in a PE deal, especially in the case of exit transactions such as trade
sales, secondary buyouts, or IPOs. Insurance companies offer protection from finan-
cial loss in the event of inaccuracies in representations and warranties through insur-
ance brokers. In exchange for a fixed premium, the insurance policy may reduce the
need for seller accruals, reserves, or collateral for contingent liabilities. Consequently,
the buyer obtains protection from loss caused by a breach of the sellers representa-
tions and warranties such as taxes and financial statements without being exposed to
the risk of the sellers inability to pay an indemnity claim. In practice, the policy can
be arranged in terms of covered amount, premiums range from 2 to 7 percent of the
limit of liability purchased, and deductibles are typically between 1 and 4 percent of
the transaction value.
By purchasing R&W insurance, a PE firm acts as an asset vendor and can reduce in-
demnity obligations for both parties. Thus, the target company vendor and the acquirer
can close deals with ease and confidence. This is the main benefit of R&W insurance,
which is increasingly used in PE deals (Meshki and Vongsawad 2013). From the sellers
232 h o w p r i v a t e e q u i t y w o r k s

perspective, this clean exit can be particularly attractive for PE fund managers in their
holding periods and sellers whose investors focus on IRR.

FINANCIAL ECONOMICS APPROACH


A financial economics approach to corporate risk management has been the most pro-
lific so far in terms of both theoretical models and empirical research. This approach
analyzes the relevance of financial structure for corporate value, agency costs, informa-
tion asymmetry costs, or stakeholder behavior. It was later extended to the field of risk
management.

Information Asymmetry and Pecking Order Theory


Jensen and Meckling (1976) suggest that information asymmetry causes dilemmas
leading to the uninformed party incurring costs. In PE, Kaplan and Strmberg (2003)
identify an information asymmetry between the GP and the corporate portfolio com-
panies. Metrick and Yasuda (2010) recognize the asymmetry that exists between GPs
and LPs.
One important question is whether a specific exit route creates more value than an-
other. Using data on listed PE firms, Mller and Vasconcelos (2012) find that their exit
strategies seem to influence the market value of such firms. Exits in the PE industry
can be used to improve corporate governance. In line with the pecking order argument,
the authors find that IPOs trigger the highest excess return, compared to the normal
return, followed by trade sales and secondary buyouts.

Risk Management Theory


This risk management approach stipulates that hedging leads to a lower volatility of
cash flow and therefore a lower volatility of firm value. Rationales for corporate risk
management were deduced from the discussion of the conditions under which the
firms financial decisions affect or do not affect its value, and included higher debt
capacity (Miller and Modigliani 1963), progressive tax rates, lower expected costs of
bankruptcy (Smith and Stulz 1985), secured internal financing (Froot, Scharfstein,
and Stein 1993), information asymmetries, and comparative advantages in informa-
tion (Stulz 1996). The result of hedging activities should be a higher firm value, includ-
ing a hedging premium.
Whether the previous rationales apply to PE firms can be addressed through an
analysis of exit deals, comparing divestitures with and without the use of R&W insur-
ance. An important issue that needs further research is whether an exit route, including
the use of such an insurance contract, returns more money than an equivalent unin-
sured deal.
In order to provide an overview of the different exit strategies, Table 13.4 presents
the most popular strategies and the main expected advantages and favorable economic
conditions for each. For example, a trade sale is an easy way to exit because it implies a
business-to-business negotiation, and a small transaction cost if low information asym-
metry exists between the vendor and the acquirer. Additionally, the strategic buyer
expecting target synergies and growing cash flow will be able to pay a premium for the
resulting embedded value so that the vendors price will be higher.
E xit S t rat e g ie s in P riv at e E qu it y 233

Table 13.4 Exit Routes Overview


Exit Route Main Goal Favorable Conditions
Trade sale
PE fund sells the target to Easy exit: business to Buyers ability to pay a
a strategic buyer including business bargain high EBITDA/Earnings
the buyback by the target Low exit cost Low information
itself. Strategic option value asymmetry between the
embedded in the sale price vendor and the acquirer
Secondary buyout
PE fund sells the target Access to leverage Target cannot offer a
to another fund. Reduce the dry powder buyback
Avoid distressed selling Flexibility in the sale
structure
Initial public offering
PE fund sells the target High return expectation Bullish stock market
to investors through an Signals investors of the Pressure to repay a debt if
organized exchange. GPs ability to manage the PE fund uses leverage
investments successfully to acquire the target
Dividend recapitalization
Target issues new debt Profits repatriation toLPs Low interest rates
to pay stockholders, Recover more or less of Low taxes on dividends
including the PE fund the initial invested amount compared to taxes on
itself, as a special dividend. Remain in control of the capital gains
portfolio company through
partial exit

Summary and Conclusions


The ultimate objective of all PE investors is to realize the return on their investment after
a certain amount of time. Exiting is the process by which PE firms liquidate their assets
and is thus a natural part of the PE transaction life cycle. The number of successful exits
achieved by a PE firm strongly influences its ability to attract investors and raise funds.
Several methods are available to PE investors to exit their investments including a trade
sale, an IPO, and a secondary buyout. Each exit route has advantages and disadvantages.
The economic environment including macroeconomic conditions, laws, taxes, and reg-
ulatory requirements influence their use.
Theoretical arguments can help explain the behavior of the PE firm. For example,
information asymmetry or hedging activities seem to imply a sort of pecking order in
the choice among exit routes. The potential exit opportunities for a PE investment play
an important role in an institutional investors decision about whether to invest in a
PE fund.
234 h o w p r i v a t e e q u i t y w o r k s

Discussion Questions
1. Describe the traditional exit routes for a PE fund.
2. Discuss the advantages and disadvantages of a trade sale as an exit strategy.
3. List the three main reasons that a firm initiates an IPO.
4. Identify the main geographic areas where PE-backed IPO deals took place between
January 2005 and June 2014.
5. List the main reasons VCs want to exit an investment.

References
Aghion, Philippe, Patrick Bolton, and Jean Tirole. 2004. Exit Options in Corporate Finance: Li-
quidity versus Incentives. Review of Finance 8:3, 327353.
Bain & Company Inc. 2014. Global Private Equity Report 2014. Available at http://www.bain.com/
publications/business-insights/global-private-equity-report.aspx.
Barry, Christopher B., Chris J. Muscarella, and John W. Peavy. 1990. The Role of Venture Capital in
the Creation of Public Companies: Evidence from the Going-Public Process. Journal of Finan-
cial Economics 27:2, 447471.
Bradley, Daniel J., and Bradford D. Jordan. 2002. Partial Adjustment to Public Information and IPO
Underpricing. Journal of Financial and Quantitative Analysis 37:4, 595616.
Cumming, Douglas, and Sofia Johan. 2010. Venture Capital Investment Duration. Journal of Small
Business Management 48:2, 228257.
Cumming, Douglas, and Jeffrey G. MacIntosh. 2001. Venture Capital Investment Duration in
Canada and the United States. Journal of Multinational Financial Management 11:45, 445463.
Cumming, Douglas, and Jeffrey G. MacIntosh. 2003. A Cross-Country Comparison of Full and
Partial Venture Capital Exits. Journal of Banking & Finance 27:3, 511548.
Froot, Kenneth A., David S. Scharfstein, and Jeremy C. Stein. 1993. Risk Management: Coordinat-
ing Investment and Financing Policies. Journal of Finance 48:5, 16291658.
Gao, Xiaohui, Jay R. Ritter, and Zhongyan Zhu. 2013. Where Have All the IPOs Gone? Journal of
Financial and Quantitative Analysis 48:6, 663692.
Gompers, Paul A. 1996. Grandstanding in the Venture Capital Industry. Journal of Financial Eco-
nomics 42:1, 133156.
Gompers, Paul A., and Josh Lerner. 1999. Conflicts of Interests in the Issuance of Public Securities:
Evidence from Venture Capital. Journal of Law and Economics 42:1, 128.
Jensen, Michael C., and William H. Meckling. 1976. Theory of the Firm: Managerial Behavior,
Agency Costs, and Ownership Structure. Journal of Financial Economics 3:4, 305360.
Kaiser, Kevin, and Christian Westarp. 2010. Value Creation in the Private Equity and Venture Cap-
ital Industry. INSEAD Working Paper, 166.
Kaplan, Steven N., and Per Strmberg. 2003. Financial Contracting Theory Meets the Real World:
Evidence from Venture Capital Contracts. Review of Economic Studies 70:2, 281315.
Kaplan, Steven N., and Per Strmberg. 2004. Characteristics, Contracts, and Actions: Evidence
from Venture Capitalist Analyses. Journal of Finance 59:5, 21772210.
Long, William F., and David J. Ravenscraft. 1993. LBOs, Debt and R&D Intensity. Strategic Man-
agement Journal 14:1, 119408.
Megginson, William L., and Kathleen A. Weiss. 1991. Venture Capitalist Certification in Initial
Public Offering. Journal of Finance 46:3, 879903.
Meshki, Hamed, and Brandon Vongsawad. 2013. Why You Need M&A Reps and Warranties Insur-
ance. Law360, Kirkland & Ellis LLP.
Metrick, Andrew, and Ayako Yasuda. 2010. Economics of Private Equity Funds. Review of Finan-
cial Studies 23:6, 23032341.
E xit S t rat e g ie s in P riv at e E qu it y 235

Metrick, Andrew, and Ayako Yasuda. 2011. Venture Capital and Other Private Equity: A Survey.
European Financial Management 17:4, 619654.
Miller, Merton H., and Franco Modigliani. 1963. Corporate Income Taxes and the Cost of Capital:
A Correction. American Economic Review 53:3, 433443.
Mller, Gtz, and Manuel Vasconcelos. 2012. Listed Private Equity and the Case of Exits. In
Douglas Cumming, ed., The Oxford Handbook of Private Equity, 611635. Oxford: Oxford
University Press.
Povaly, Stefan. 2006. Private Equity Exits: An Analysis of Divestment Process Management in Rela-
tion to Leveraged Buyouts. PhD dissertation, University of St. Gallen, Graduate School of
Business Administration, Economics, Law and Social Sciences.
Preqin. 2014. The 2014 Preqin Global Private Equity Report. Available at https://www.preqin.com/
docs/samples/The_2014_Preqin_Global_Private_Equity_Report_Sample_Pages.pdf.
Schmidt, Daniel, Sascha Steffen, and Franziska Szabo. 2010. Exit Strategies of Buyout Investments:
An Empirical Analysis. Journal of Alternative Investments 12:4, 5884.
Smith, Clifford W., and Ren M. Stulz. 1985. The Determinants of Firms Hedging Policies. Journal
of Financial and Quantitative Analysis 20:4, 391405.
Smith, Richard, Robert Pedace, and Vijay Sathe. 2011. VC Fund Financial Performance: The Rela-
tive Importance of IPO and M&A Exits and Exercise of Abandonment Options. Financial
Management 40:4, 10291065.
Stulz, Ren M. 1996. Rethinking Risk Management. Journal of Applied Corporate Finance 9:3,
824.
Zephyr Database. 2014. Bureau van Dijk.
Part Four

PERFORMANCE AND MEASUREMENT


14
Private Equity Returns
Cross-Country Evidence
KENNETH SMALL
Marshall Butler Professor of Finance, Coastal Carolina University

JEFFREY S. SMITH
Assistant Professor of Economics and Finance, Virginia Military Institute

Introduction
For some, the term private equity (PE) brings to mind that scene from A Few Good Men
in which Kevin Bacon, a prosecutor for the Marine Corps is delivering his opening remarks
saying, Hes going to astonish you with stories of rituals and dazzle you with official sound-
ing terms like Code Red. Put in terms comparable to the SAT, the answer to the question
would be: private equity is to finance as a Code Red is to the Marine Corps. In reality, noth-
ing mystical exists about PE. The term typically refers to a class of investors that purchases a
public company with the intent of taking it private, improving its operating performance, and
then selling it for a profit. PE also refers to a group of investors that provides capital for young
companies to facilitate their growth, again so they can then sell those companies for a profit.
The standard PE fund operates much like a closed-end mutual fund, although with
a finite lifespan (10 years typically, with an option to extend a maximum of three more
years), with limited partners (LPs) playing the role of investors (Kaplan and Schoar
2005). The general partners (GPs) are responsible for fundraising and managing the
assets of the fund (i.e., selecting investments and choosing when to exit). According to
Kaplan and Strmberg (2008), the GPs typically supply at least 1 percent of the capital
raised, but are well compensated for their efforts. GPs are compensated in three ways:
(1) an annual management fee, which is a function of capital committed to deals; (2) a
share of the profits earned, more commonly referred to as carried interest, and (3) fees
associated with monitoring and selecting the deal.
The question then becomes: Are the returns generated by PE funds, net of fees and risks,
high enough to compensate LPs (investors) for the illiquidity associated with their invest-
ment? Even though PE LPs tend to be large, long-term liability managers such as insurance
companies, banks, pension funds, and more recently college endowments for which asset
matching is less of an issue, a theoretical cost still is associated with the long lock-up period.

239
240 p e r f o r m a n c e a n d m e a s u r e m e n t

This chapter examines the relative merits associated with PE investments as com-
pared to the costs in a global setting. The rest of the chapter is organized as follows. The
next section presents evidence on the growing global focus on PE. This section is fol-
lowed by a review of the academic literature as it applies to PE. The next two sections
present the general attractiveness of PE and highlight the specific attractiveness for cer-
tain foreign countries. Next, the chapter discusses the historical returns to listed PE. It
then explores the notion that while North America and Europe have been the focal point
for PE deals by dollar volume, other countries and regions started to attract an increas-
ing share of PE capital in the late 1990s and early twenty-first century. The chapter con-
cludes with a presentation and discussion of return data for several global PE indexes, as
well as the portfolio-diversifying ability of global PE.

The Growing Global Interest in Private Equity


Aizenman and Kendall (2008, p. 1) describe venture capital (VC) as a long-term,
hands-on equity investment in privately held, high-growth potential companies. Thus,
venture capitalists are professional investors who not only take a financial stake in trying
to establish a company but also actively participate in its management. VC financing
usually comes in rounds, including the earliest round (i.e., seed money viewed as most
risky and potentially most profitable) up through later stage financing, which corre-
sponds to lower risk and lower returns (Wright and Robbie 1998). Aizenman and Ken-
dall describe PE as deals that are not VC, specifically including deals such as buyouts,
buy-ins, mergers, and many other types of deals. The main difference between VC and
PE is that targets of PE funds are not simply trying to bring an initial product to market;
they are companies that need financial engineering or a change of ownership to increase
operating efficiencies. When discussing PE, non-VC deals are typically referred to as
buyouts, most of which are public-to-private (PTP) buyouts.
PE started in the United States in 1976 when Kohlberg, Kravis, and Roberts (KKR)
formed their eponymous company and purchased their first firm in 1979 (Seretakis
2012). Not surprisingly, the United States is the epicenter for PE investments. In fact,
The Companies Act of 1981 changed the law such that PE became legal in Europe
(Seretakis 2012). The rules differ for European companies undergoing buyouts, such
as the certain funds requirement and the general restriction banning target companies
from helping acquiring companies (Seretakis 2012). The restriction is such that acquir-
ing companies have been prevented, or greatly limited, in their ability to use assets of the
acquired company to secure funding for the acquisition. This severely constrained the
LBO market in Europe. Nonetheless, Europe including the United Kingdom has grown
into the second largest PE market. According to Kaplan and Strmberg (2008), West-
ern Europe represented a larger share of the PE transactions than did the United States
from 2000 to 2004 at 48.9 and 43.7 percent, respectively.
Within Europe, the United Kingdom, which serves as the headquarters to many
firms, is the largest market, exhibiting great capital depth. Sommer (2013) identi-
fies 40,682 completed deals in Europe from 1990 to 2009 with the United Kingdom,
France, Germany, and the Netherlands dominating this list. These completed deals rep-
resent between 60 and 70 percent of PE deals in any given year. Notably, according
P riv at e E qu it y R e t u rn s 241

to ImadEddine and Schwienbacher (2011), 25 percent of the capital in European PE


funds originated in North America, specifically by those investors who also maintain
local European facilities (e.g., a branch office for an insurance company). Aizenman and
Kendall (2008) explore this distribution by cataloguing deals with cross-border partic-
ipation. They find VC deals had a cross-border participation of only 15 percent in the
1990s, climbing to 40 percent by the late 2000s.
The North American or Pan-European focus is starting to shift. Klonowski (2013)
identifies $318.8 billion in PE funds in emerging markets from 2000 to 2011. Emerging
markets, especially the so-called BRIC (Brazil, Russia, India, and China) nations, rep-
resent a growing proportion of both the worlds population and gross domestic product
(GDP). At the end of 2012, the BRIC nations represented 21 percent of global GDP
and 43 percent of the worlds population (Herskovitz 2013). In terms of diffusion of
PE throughout the BRIC markets, when calculating the percentage of PE investment
to GDP, India ranks first at 0.32 percent; China is second at 0.13 percent; Brazil is third
with 0.07 percent; and Russia is the last of the four with 0.02 percent (Klonowski 2011).
Compared to the United States and the United Kingdom with 0.30 and 0.32 percent,
respectively, many institutional investors still focus on well-developed markets.
From 2001 through 2009, PE funds raised $31.8 billion in China, $31.1 billion in
Brazil, $23.3 billion in India, and $10.9 billion in Russia. Ippolito (2007) attributes dif-
ferences in PE growth rates between China and India, two countries that are otherwise
viewed as similar growth stories, to more targeted educational opportunities for Indian
workers and a more developed institutional and regulatory environment, which pro-
motes the growth of equity markets. The BRIC markets dominate the emerging market
arena but other regions have also witnessed growth in their PE market. According to
Klonowski (2011), PE funds based in emerging markets raised $22.6 billion in 2009,
which is a more than threefold increase over the $6.5 billion raised in 2001. In South
Africa, for instance, the amount of PE funds raised grew by almost 35 percent year-
over-year, from $1.29 billion in 2011 to $1.65 billion in 2012, with $14.5 billion in total
PE funds under management (KPMG and SAVCA 2013). The Middle East and North
Africa have averaged $1.2 billion in annual PE investments from 2002 to 2011.

Literature Review on Private Equity Returns


Discussions of reported returns for PE can get complicated. Cumming and Walz (2010)
split returns for PE into two categories: unrealized/partially realized and fully realized.
Why make this distinction? Fully realized returns are known as exits and are based on
the full PE cycle. The exiting strategy for a buyout (BO) fund is typically a PTP trans-
action, with subsequent operational and financial improvements, followed by a dispo-
sition of the company, usually by an initial public offering (IPO), but potentially the
sale of the company to other private investors. Unrealized or partially realized returns
show that a business is still attempting to increase value or has partially generated some
returns to investors, possibly through asset sales or other financial engineering. For un-
realized returns, some probability exists that the valuation identified by the GP may
not materialize. Wright, Renneboog, Simons, and Scholes (2006) note this trend in
the United Kingdom. Based on their analysis of 12,267 buyouts from 1985 to 2005,
242 p e r f o r m a n c e a n d m e a s u r e m e n t

12 percent entered bankruptcy protection. While secured creditors recovered about 60


percent of their investments, Wright et al. (2006) did not identify recovery rates for
equity investors, which are presumed to be negligible.
The sample period for the data set used by Cumming and Walz (2010) spans from
1976 through 2003. They report median internal rate of returns (IRRs) for fully real-
ized investments or exits as follows: United Kingdom (24.1 percent), United States
(13.8percent), France (12.4 percent), Germany (15.4 percent), Switzerland (35.5 per-
cent), and Sweden (21.4 percent). The average IRR across the same period is larger for
each country. However, some countries have few firms, which most likely skewed the
results. For example, Switzerland and Sweden had 12 and 27 firms, respectively, report-
ing results. These results are misleading because they likely suffer from survivorship
bias. Exits presumably show a successful investment, while the unrealized/partially re-
alized category is more likely to contain unsuccessful investments.
Recognizing this, Cochrane (2005) tries to correct the results for this systematic
bias, which he describes as selection bias. While his research looks at VC funds, his re-
sults also apply to PE. Cochrane finds a negative intercept in a market model estimated
in logs, suggesting management underperformance. Cochrane notes that these results
are similar to performance characteristics available in the smallest NASDAQ stocks.
Cochrane also calculates the returns for VC projects, not VC funds. Thus, returns would
be lower for those who invested in VC funds because Cochranes calculations are gross of
fees. Phalippou and Gottschalg (2009) correct for systematic reporting errors to include
selection bias and recalculate performance and alpha. Their results, net of management
fees, confirm Cochranes work, showing that PE alpha is about 3 percent annually.
Klonowski (2011) reports that combined five-year average annual returns for the
BRIC markets equal to 12.8 percent. This finding is presumably for the five years ending in
2010. The five-year average annual returns for PE in Latin America and the Caribbean Is-
lands are 11.9 percent (Klonowski 2013). This result compares to a five-year return in the
United States and Western Europe of 9.2 percent and 11.8 percent, respectively. Extending
to a longer time frame, he reports 10-year annual returns of 7.7, 7.4, and 14.8 percent for
the BRIC countries, the United States, and Western Europe, respectively. For the 10-year
period from 2001 to 2010, Russia experienced a 14 percent annualized return, with returns
for PE in Poland of 15.6 percent annually across the same time period. These returns suffer
the same criticism as noted by Cochrane (2005) and Phalippou and Gottschalg (2009)
among others, namely selection bias. All of these results are calculated based on exits.
Notwithstanding the selection bias issue, other problems are associated with accu-
rately assessing PE performance. Kaplan, Sensoy, and Strmberg (2002) examine the
accuracy of the two most widely used databases, VentureOne and VentureXpert. While
mainly focused on VC funds, the authors conclude that material issues stem from the
way information is collected on a self-reporting basis. Of more concern is the lack of
adjustments for risk. As Phalippou and Gottschalg (2009) note, BO funds typically use
large amounts of leverage giving them a risk profile similar to that of small growth com-
panies. Thus, assuming a market risk equal to one would be inappropriate, given the
high systematic risk inherent in this investment type. Once the authors correct for this
bias, they calculate a profitability index of 0.75 for BO funds. Any profitability index
less than one shows performance below the benchmark. Phalippou and Gottschalg use
the S&P 500 index as their benchmark. As with the other corrections, adjusting for risk
P riv at e E qu it y R e t u rn s 243

equates to an alpha of 6 percent a year. European Union (EU)-focused funds also ex-
hibit underperformance relative to the S&P 500 index, irrespective of the currency used
to value the gains (dollar or Euro).
A large body of literature discusses the benefits associated with diversification across
many asset classes. For this chapter, PE investing should not be viewed as an asset
class. PE and VC are not separate asset classes. PE should be considered a substitute
for equity in any traditional asset allocation plan. While PE funds are considered an
alternative investment, they represent ownership of a company in which value is cre-
ated or enhanced by improving either its operating performance or financial structure.
This classification suggests that investors considering an investment in a PE fund should
compare the expected risk and return for the PE fund against the same characteristics
for traditional equity investments.
Kaplan and Schoar (2005) use the Venture Economics data set to calculate PE fund re-
turns net of expenses. Their sample consists of 164 BO funds that are effectively liquidated
in that 10- to 13-year window. Assuming equal weighting when calculating returns and size,
the general results do not change: PE underperforms public equity after deducting fees.
Diller and Kaserer (2009) examine mature European PE funds to avoid problems
calculating illiquid investment returns but reach the same conclusion: net of fees perfor-
mance is less than the market benchmark. Specifically, the average excess IRR, in which
excess IRR is compared against the appropriate benchmark index, is 2.27 percent for
VC funds compared to 3.37 percent for BO funds from 1980 to 2003 using the Venture
Economics data set. However, the median excess IRR is negative for both VC funds and
BO funds (4.17 percent and 0.77 percent, respectively). Both Kaplan and Schoar
(2005) and Phalippou and Gottschalg (2009) note that gross of fees, PE slightly out-
performs public equity; however, only the net of fees performance matters to investors.
Driessen, Lin, and Phalippou (2012) examine the same data set as Diller and Kas-
erer (2009). Using a process that allows the IRR to be dynamic instead of static, they
find strong negative abnormal performance both gross and net of fees, with an alpha of
12 and 8.5 percent, respectively. As Driessen et al. (2012, p. 513) note, It is the risk
correction that makes alpha negative for VC funds, and some casual evidence indicates
that investors underestimated the risk.

Country-Specific Private Equity Attractiveness


No discussion of global returns to PE across countries would be complete without re-
viewing what drives PE investment. Examining the PE investment fund flow landscape
shows wide variation in investments across countries. The United States, Canada, and
the United Kingdom enjoy high rates of PE investments, while countries such as Italy,
Greece, and Spain attract relatively low levels of investing.
Focusing on six factors, including economic activity, well-functioning capital mar-
kets, taxes, corporate governance and investor protection, the social environment, and
the existence of entrepreneurial opportunities, Groh, Heinrich, and Lieser (2008, 2009,
2010, 2011, 2012, 2013, 2014), develop and report The Venture Capital & Private
Equity Country Attractiveness Index, which measures country attractiveness for PE
investors. The index includes 42 data items collected from various sources. To capture
244 p e r f o r m a n c e a n d m e a s u r e m e n t

economic activity, Groh et al. (2008) employ variables such as GDP, the general price
level, and net foreign direct investment. To measure the capital market strength, they
use the number of IPOs, stock market capitalization, merger and acquisition (M&A)
activity, and other variables. The taxation construct is measured by the highest marginal
tax rate and the difference between the income and capital gains tax rates. The authors
measure investor protection by using the World Bank data indexes of corporate gov-
ernance. The human and social environment construct is measured using educational
data, labor data, and crime statistics. Entrepreneurial opportunities are measured by
research and development (R&D) expenditures, the number of enterprises per capita,
and the burden of starting a business among other variables. The attractiveness index
ranges from 0 to 100 and the authors use factor analysis to calculate relative weighting
of the factors that drive the index value.
Groh et al. (2008) first tested the attractiveness index on a set of European countries.
Then they expanded the index to include non-European countries in 2010 (Groh et al.
2010) and finally expanded to 118 countries in 2013 (Groh et al. 2013). Table 14.1 lists
the 10 most attractive countries. The attractiveness index values are included in paren-
theses below the countrys name. The top 10 countries have remained relatively stable
over the five-year window. Countries can advance or decline within the top 10, but the
constituents are relatively stable over the analyzed time period.
Figure 14.1 illustrates the index values for the top and bottom three countries in
2014. The two most attractive countries, the United States and Canada, have held these
positions from 2010 to 2014. However, the third most attractive country has changed
over the five-year window. In 2014, Singapore held the position although the United
Kingdom held the position in 2010 and again in 2012 and 2013. The three least attrac-
tive countries have not changed since expansion of the index in 2012. Burundi, holding
the last position, had a score of 10.1 in 2014, 10.3 in 2013, and 11.0 in 2012. Chad
scored last in 2012 at 12.0, second to last in 2013 with a score of 12.8, and third from last

Table 14.1 The 10 Most Attractive Countries for Private Equity, 2010 to 2014

Rank 2010 2011 2012 2013 2014


1 United States United States United States United States United States
2 Canada United Canada Canada Canada
Kingdom
3 United Canada United United Singapore
Kingdom Kingdom Kingdom
4 Australia Singapore Japan Japan United Kingdom
5 Hong Kong Switzerland Singapore Singapore Hong Kong
6 Singapore Japan Hong Kong Australia Japan
7 Japan Australia Australia Germany Germany
8 Switzerland Sweden Sweden Hong Kong Australia
9 Netherlands Netherlands Germany Sweden Sweden
10 Germany Germany Switzerland Switzerland Switzerland
P riv at e E qu it y R e t u rn s 245

120

100

80

60

40

20

0
2010 2011 2012 2013 2014

USA Canada Singapore Burundi Angola Chad

Figure 14.1 Graph of the Top Three and Bottom Three Countries for PE
Attractiveness This figure shows the PE attractiveness index values for the top three
(USA, Canada, and Singapore) and bottom three countries (Burundi, Angola, and Chad).

in 2014 with a score of 13.7. Angola was second to last in 2014 with a score of 12.7, third
from last in 2013 with a score of 14.8, and third from last in 2012 with a score of 13.0.
Table 14.2 reports the attractiveness index values and rank of the BRIC markets. This
provides some insights into how emerging countries score on the ranking index. Exami-
nation of Table 14.2 shows that all BRIC countries increased in both rank and attractive-
ness score over the five-year analysis period. Brazils attractiveness score increased from
34.6 to 64.0an 84.9 percent increase, moving up 11 positions in rank. This was the
largest percentage increase in attractiveness score and largest rank increase of any BRIC
country over this time period. China increased the least of any BRIC country, in rank
and attractiveness score, over the five-year window. Its attractiveness score increased
from 48.5 to 78.4, an increase of 61.2 percent, and increased in rank by only six positions.
This change is mainly due to Chinas relatively high starting position. Russia increased
seven positions to 41 overall and India increased 10 positions to a rank of 28 overall.

Table 14.2 B
 razil, Russia, India, and China Country Attractiveness Index Values,
2010 to 2014

Country 2010 2011 2012 2013 2014


Brazil 34.6 (51) 48.7 (43) 63 (36) 66.1 (36) 64 (40)
Russia 38 (48) 51.1 (41) 60 (43) 61.3 (40) 63 (41)
India 40.9 (38) 61.4 (30) 66 (32) 69.9 (29) 70.7 (28)
China 48.5 (28) 72.3 (20) 79 (22) 77.4 (24) 78.4 (22)

Note: This table shows the PE attractiveness values, with their respective rank in parentheses, for
each of the BRIC countries.
246 p e r f o r m a n c e a n d m e a s u r e m e n t

Wide variations exist among the attractiveness that countries provide for PE inves-
tors. These factors have a direct impact on PE investment and economic growth. The
full attractiveness index is available at http://blog.iese.edu/vcpeindex/.

Data on Global Private Equity Activity


Where global PE deals commonly occur is important. Table 14.3 presents deal data
from 2006 through 2013 from Preqin Financial (Preqin 2014). When examining these
data, keep the previous discussion of attractiveness in mind. Recall that Canada and the
United States dominate the top two attractiveness positions. As expected, when exam-
ining the data from 2006 to 2010, in most years, close to 50 percent of all deals occur in
North America. In each year, Europe comes in second, Asia occupies the third position,
and the rest of the world follows. The deal data confirm the value of the attractiveness
indexs power to discern markets and countries where investors will actually invest.
Looking at the proportion and level of buyouts in each region from 2006 through
2013, the first trend is a substantial decline in BO activity in 2008, which represented
the beginning of the financial crisis. North American deals dropped from $450 billion in
2006 to $51 billion in 2009, representing an 88.6 percent decline. Over this same time,
Europe and the rest of the world, except Asia, experienced a similar percentage decline
in activity, while PE in Asia increased by $1 billion.
An informative way to examine the decline in activity over the entire analysis period
is to compare each geographical locations proportion to the starting value of PE BO
deals that occurred in 2006, by year. Figure 14.2 illustrates this comparison. Examining
the data in this fashion reveals that Asia has been surprisingly resilient to the financial
crisis of 20072008, while North America, Europe, and the Rest of the World all are
at about 40 percent of their 2006 value level. However, this only reveals how the regions
are doing relative to their starting values. It does not account for growth in PE buyouts
that may have occurred over the analysis period across the regions.

Table 14.3 A
 ggregate Value of Private Equity-Backed Buyout Deals by Region,
2006 to 2013

Year North America Europe Asia Rest of World


(USD billions) (USD billions) (USD billions) (USD billions)
2006 450 179 18 23
2007 421 190 28 23
2008 87 79 19 8
2009 51 29 22 4
2010 123 74 19 14
2011 128 96 25 17
2012 156 70 27 11
2013 171 74 19 9
P riv at e E qu it y R e t u rn s 247

1.6 1.6
1.5
1.4
1.4
1.2
1.2 1.1 1.1 1.1 1.1
1.0 1.0 1.01.01.0 0.9 1.0

0.8 0.7
0.6 0.5
0.6
0.4 0.3 0.4 0.30.4
0.5 0.40.4 0.4
0.4 0.3 0.3
0.2 0.2 0.1 0.2 0.2

0.0
2006 2007 2008 2009 2010 2011 2012 2013
North America Europe Asia Rest of World

Figure 14.2 Global Private Equity Buyouts as a Proportion of their 2006


Value This figure shows the proportion of BO activity across the world relative to their
values in 2006.

To account for inter-industry growth requires controlling for the proportion of each ge-
ographical area relative to the total deals for each year. Without this control, discerning the
changes in where the actual proportion of deal activity occurs is difficult. Figure14.3 shows
the proportion that each geographic area contributes to the total world value of PE BO
deals. Figure 14.4 shows the proportion of value that each geographical region contributes
to total world global equity BO value for 2006 and 2013 and presents the numerical values.
Examining the proportion, rather than the level, of activity across the geographical
regions over the period reveals that the proportions revert in 2013 to about where they

Figure 14.3 Proportion of Buyout Deals by Geographic Region,


2006 to 2013 This figure represents the amount of BO deals that occur in each
geographic region, as compared to the total for that year.
248 p e r f o r m a n c e a n d m e a s u r e m e n t

70% 67%
63%
60%

50%

40%

30% 27% 27%

20%

10% 7%
3% 3% 3%
0%
North America Europe Asia Rest of World
2006 2013

Figure 14.4 Global Buyout Deals by Geographic Regions, 2006 vs. 2013
This figure demonstrates the change in the percentage of BO deals by region from 2006
and 2013.

started in 2006 as shown in Figures 14.3 and 14.4. Of total global BO activity, 77.0 per-
cent occurred in North America in 2006 but by 2013 that value had declined to only 62.6
percent. In 2006, Europe created 26.7 percent of the total value of global PE BO value,
and by 2013 that value had increased slightly to 27.0 percent. As with the previous com-
parison using 2006 starting values, Asia ends 2013 with greater BO activity than in 2006.
Its total value contribution to global PE BO activity increased from 2.7 to 7.0 percent.

The Historical Performance of Global Listed


Private Equity
Raw self-reported PE returns in a global context are rife with self-reporting bias, in-
consistency, and lack of data in some countries and regions. As a result, this section
extends the previous data analysis by examining index data that are verifiable and read-
ily available and by employing two listed PE indexes to proxy for the returns to pure
PE investments. The two indexes are the Global Listed Private Equity (GLPE) Index
and the PowerShares Global Listed Private Equity (PSP) Portfolio ETF. Other global
PE indexes and exchange-traded funds (ETFs) are available that can serve as proxies,
although they all have drawbacks that outweigh the benefits of including them in the
studies of global listed PE. For example, the Proshares Global Listed Private Equity
ETF (PEX) has only been trading since February 26, 2013. The S&P I-shares Listed PE
Index (IPRV) has 68 percent of the index components with their assets invested in U.S.
firms. Even after excluding these globally recognized indexes, the return analysis that
follows must be prefaced by accepting certain limits and caveats.
First, the investment weights reported for each index are from March 31, 2014, and
may not represent the historical weights, by industry, within each index. However, they
represent historical weights within reasonable bounds. The historical weight of each
index investment across industries could be tracked, but that would not provide a useful
P riv at e E qu it y R e t u rn s 249

discussion of PE returns. Second, both indexes are heavily invested in the United States,
Canada, and Europe. For example, no widely recognized global PE ETF or index had
more than 15 percent of the fund invested outside the United States, Canada, and Europe.
Recall that according to the attractiveness index, the United States and Canada rank as the
most attractive locations for PE investment with the top 20 locations heavily dominated
by European countries. Third, the data are not widely available before 2008, so the time
period of the analysis spans the economic collapse that occurred in 2008 through the
recovery of the equity markets in 2009 and 2012. Thus, the analytical results should be
interpreted with this in mind. Fourth, listed PE serves as a rough estimate of the returns of
PE investments, but as previously stated, those data are not readily available or verifiable.

The Red Rocks Global Listed Private Equity Index


The Red Rocks Global Listed Private Equity Index (GLPE) is a benchmark index de-
signed to track the performance of the returns on investment in global PE firms. The
index began tracking global PE returns on September 28, 2007, and was constructed by
Red Rocks Capital, an asset management firm specializing in listed PE and other private
asset investments. The official descriptive language from the factsheet reads (Red Rocks
Capital 2014, p. 1):

The Index is designed to track the performance of private equity firms which are
publicly traded on any nationally recognized exchange worldwide. These com-
panies invest in, lend capital to, or provide services to privately held businesses.
The Index is comprised of 40 to 75 public companies representing a means of
diversified exposure to private equity firms. The securities of the Index are se-
lected and rebalanced quarterly per modified market capitalization weights.

As of March 31, 2014, the breakdown of PE activity by global region is 44.2 percent
in North America and Canada, 42.2 percent in Europe, 9.2 percent in the Far East Asia
and Japan, 0.5 percent in South America, and 3.4 percent elsewhere. The index is heavily
invested in developed PE markets that are strongly influenced by the returns in Europe
and the United States.
As of March 31, 2014, the index is also diversified across industries with 21.7 percent
industrials, 18.0 percent financial services, 15.6 percent consumer discretionary, 10.9
percent information technology, 9 percent healthcare, 5.9 percent energy, 4.4 percent
consumer staples, 2.9 percent invested in materials, 2.1 percent invested in telecom-
munications, 0.7 percent in utilities, and the balance diversified across other industries.
With respect to the stage of investment, 61.3 percent of investments are late-stage, 31.5
percent are mid-stage, and 7.1 percent are early-stage.

PowerShares Global Listed Private Equity Portfolio


The PowerShares Global Listed Private Equity Portfolio (PSP) is an ETF based on the
Red Rocks Global Listed Private Equity Index (GLPE). While the GLPE represents
250 p e r f o r m a n c e a n d m e a s u r e m e n t

the returns to a true index without consideration for market impact costs, other trading
costs, or taxes, the PSPs return accurately reflects the returns an investor would realize.
The PSP is based on the GLPE; PSP holds different proportions of the equities held
by the GLPE. It is included in this study because the PSP also captures the returns to
developed global listed PE.
On March 31, 2014, the PSPs fund country allocations were as follows: 39.70 per-
cent in the United States, 15.45 percent in the United Kingdom, 8.80 percent in France,
5.97 percent in Canada, 4.84 percent in Switzerland, and the remaining 20.00 percent is
invested in Sweden, Belgium, Hong Kong, Japan, and Malta. While the weightings are
similar to those of the GLPE Index, they are not identical. The weighting across indus-
tries also differs. The PSP holds 75.68 percent of its portfolio in financials and 11.58 per-
cent in consumer discretionary, with the remainder invested across the other industries.
Again, as with the GLPE Index, the PSP should be considered a proxy for the returns to
listed developed global market PE.

Comparison of Private Equity Investments Indexes


The analysis comparing the returns of listed global PE indexes is expanded to several
other investment asset class returns. Each comparison group is discussed in the follow-
ing sections.

The S&P 500 Index


The S&Ps 500 Index is one of the most widely recognized U.S. equity indexes in the
world, consisting of 500 large U.S.-based firms. The S&P 500 index is used in this analy-
sis as a proxy for the returns to the U.S. stock market as a whole.

MSCI World Index


The Morgan Stanley Capital International (MSCI) World Index is an index of 1,612
global corporations. It consists of stocks from developing markets but excludes stocks
from developing and emerging economies.

MSCI Europe Index


The Morgan Stanley Capital International (MSCI) Europe Index captures large and
mid-cap representation across 15 developed markets in Europe.

S&P GSCI Commodity Total Return


The S&P GSCI is a benchmark for investment in the commodity markets and a meas-
ure of commodity performance over time. It consists of investments across a wide range
of commodities including oil, which at 80 percent weighting is by far the most heavily
weighted commodity, metals, and agricultural products and livestock, which is the least
weighted at 3 percent.

FTSE EPRA/NAREIT Global Real Estate Index


The FTSE EPRA/NAREIT Global Real Estate Index is designed to represent general
trends in eligible listed real estate stocks worldwide. The index invests in companies
P riv at e E qu it y R e t u rn s 251

that are classified as real estate, heavy construction, and home construction in a limited
number of developed and emerging counties.

Global Listed Private Equity Return Analysis


For purposes of consistency, this chapter uses data from January 2, 2008, to December
31, 2013, which allows for calculating six full year returns. Table 14.4 shows that the
return series begins in a tumultuous time for the financial markets. The GLPE and PSP
both lost more than 60 percent of their value during 2008 and recovered some of that
loss in 2009 and 2010. The total return over the six-year time span was 6.40 percent
for the GLPE and 44.62 percent for the PSP. Out of the seven indexes evaluated, the
worst performing index was the PSP, while the third best was the GLPE. However, these
calculations do not consider the risk inherent within each index.
To take risk into account involves calculating daily returns as well as the standard
deviation of each index and its coefficient of variation. The coefficient of variation is the
standard deviation (a measure of risk) divided by the mean return, calculated over some
fixed period of time. This study, however, is interested in the inverse of this measure
because the inverse measures how much return an investor earns per unit of risk. Essen-
tially, each average daily return over the investigation period is divided by the average
daily standard deviation, which is estimated over the analysis window. Table 14.5 and
Figure 14.5 report these ratios. The higher the inverse coefficient of variation, the more
return is provided for each unit of risk taken.
Over the period studied, the S&P 500 index displays the highest daily risk-adjusted
return of 0.018 while the GLPE exhibits the second highest risk-adjusted daily return
of 0.011. The PSP, as expected from the returns discussed previously, came in second
to last with an inverse coefficient of variation of 0.004. The large overweighting in fi-
nancials may affect much of the seemingly sub-par performance of the PSP. A quick
check of the XLF, an ETF designed to track the performance of U.S. financial compa-
nies, shows that the XLF has not returned to its January 1, 2008, level. This is confirmed
by the IXG, an ETF designed to track the performance of financial companies globally,
which is also well below its January 1, 2008, level.

Table 14.4 Annual Returns by Year for Various Assets, 2008 to 2013

Annual Return 2008 2009 2010 2011 2012 2013 Total Return
% % % % % % 20082013%
Red Rocks GLPE 62.87 44.84 25.91 19.42 29.07 36.12 6.40
Proshares PSP 67.19 21.43 15.22 26.56 26.66 18.16 44.62
S&P 500 37.58 19.67 11.00 1.12 13.41 26.39 27.72
MSCI World 41.69 23.52 7.62 8.18 12.85 13.58 5.26
MSCI Europe 47.92 26.95 1.06 13.23 13.96 18.82 16.63
S&P Commodities 48.18 10.33 6.32 1.54 0.08 2.12 37.36
FTSE Real Estate 44.39 21.44 25.99 0.08 15.85 6.65 6.75
252 p e r f o r m a n c e a n d m e a s u r e m e n t

Table 14.5 Inverse Coefficient of Variation, 2008 to 2013

Coefficient of Variation 2008 2009 2010 2011 2012 2013 20082013


Red Rocks GLPE 0.109 0.080 0.082 0.037 0.107 0.179 0.011
Proshares PSP 0.095 0.043 0.051 0.046 0.089 0.076 0.004
S&P 500 0.059 0.056 0.047 0.007 0.065 0.149 0.018
MSCI World 0.091 0.071 0.039 0.016 0.064 0.137 0.009
MSCI Europe 0.081 0.063 0.010 0.022 0.049 0.088 0.002
S&P Commodities 0.088 0.034 0.032 0.004 0.006 0.003 0.010
FTSE Real Estate 0.064 0.042 0.080 0.010 0.087 0.025 0.006

1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
2008 2009 2010 2011 2012 2013

GLPE Proshares PSP S & P 500


MSCI World MSCI Europe S & P Commodities
FTSE Real Estate

Figure 14.5 Correlation Coefficient for Global Listed Private Equity


Compared with Other Assets This figure is a graphical presentation of the change in
correlations between the GLPE Index and other major asset class indexes.

Global Private Equity as a Diversifying Agent


Regarding the ability of PE to serve as a diversifying agent in a portfolio, this study
estimates the correlation coefficients among the daily returns of the GLPE, PSP, and
several other indexes. Tables 14.6 and 14.7 include the results. For further exposition,
Figure14.6 also illustrates the results.
Note the high overall correlation between the daily returns of GPE and PSP with the
S&P 500 index (0.695 and 0.870, respectively). A high overall correlation also exists be-
tween the daily returns of the GLPE with the daily returns of the MSCI World (0.826)
and European (0.765) indexes. However, the returns of the MSCI European Index are
not highly correlated with the PSP. The S&P Commodities Index displays moderate
correlations with both the GLPE and the PSP, while the returns of the FTSE Real Estate
P riv at e E qu it y R e t u rn s 253

Table 14.6 Global Listed Private Equity Correlation Coefficients, 2008 to 2013

Correlation 2008 2009 2010 2011 2012 2013 20082013


Proshares PSP 0.658 0.879 0.873 0.932 0.895 0.752 0.781
S&P 500 0.537 0.817 0.822 0.893 0.775 0.753 0.695
MSCI World 0.691 0.913 0.952 0.972 0.922 0.897 0.826
MSCI Europe 0.641 0.816 0.910 0.905 0.902 0.808 0.765
S&P Commodities 0.268 0.594 0.709 0.579 0.572 0.339 0.457
FTSE Real Estate 0.460 0.707 0.578 0.725 0.259 0.411 0.558

Table 14.7 P
 owerShares Global Listed Private Equity Correlation Coefficients,
2008 to 2013

2008 2009 2010 2011 2012 2013 20082013


Red Rocks (GLPE) 0.648 0.879 0.873 0.932 0.895 0.752 0.781
S&P 500 0.863 0.880 0.921 0.935 0.870 0.740 0.87
MSCI World 0.802 0.810 0.885 0.910 0.884 0.728 0.818
MSCI Europe 0.528 0.573 0.707 0.764 0.742 0.512 0.59
S&P Commodities 0.239 0.543 0.660 0.516 0.560 0.268 0.418
FTSE Real Estate 0.820 0.809 0.659 0.766 0.369 0.425 0.772

Index are more positively correlated with the PSP than with the GLPE. Interestingly,
the correlation between the GLPE and the PSP is 0.781, which is actually lower than the
correlation between the returns of the GLPE and the MSCI World Index (0.826) index.
In almost every case, the correlations in 2008 are lower than in the later years. Similarly,
in almost all cases, the highest correlations occur in 2011. This result is mainly due to
the economic collapse that occurred in 2008 and the recovery that started in 2009 and
continued into 2012.
Examining Figure 14.6, the correlation coefficients follow a typical pattern, except
for changes compared to the S&P Commodities Index and the FTSE Real Estate Index.
For both the GLPE and the PSP, greater variation in the changes to the correlation coef-
ficients occurs, as well as trends that move opposite the other asset classes.
Overall, because the correlations between the PSP and GLPE do not equal one, an
opportunity for diversification exists. Still, the correlations are much higher between
the equity indexes and the returns of the GLPE and the PSP. This could possibly be
because listed PE returns are used to generate these returns series. The returns to the
PE deals themselves may be much different than those of listed PE returns. As men-
tioned previously, PE deals and opportunities are highly concentrated in the United
States, Canada, and Europe. Additionally, most of the deal returns are not publically
available.
254 p e r f o r m a n c e a n d m e a s u r e m e n t

Red Rocks (GLPE) S & P 500


MSCI World MSCI Europe
S & P Commodities FTSE Real Estate
1.000

0.900

0.800

0.700

0.600

0.500

0.400

0.300

0.200

0.100

0.000
2008 2009 2010 2011 2012 2013

Figure 14.6 Correlation Coefficient for PowerShares Global Listed


Private Equity Compared with Other Assets This figure is a graphical
presentation of the change in correlations between the PSP Index and other major asset
class indexes.

Summary and Conclusions


This chapter reviewed the literature on global PE returns and discussed the reported
global returns. It also discussed the six primary factors that drive PE investments across
countries: (1) economic activity, (2) well-functioning capital markets, (3) taxes, (4)
corporate governance and investor protection, (5) the social environment, and (6) the
existence of entrepreneurial opportunities. The Country Attractiveness Index as it ap-
plies to PE investments is also reported. The most highly ranked countries tend to be in
North America and Western Europe. BO deal flow data show that most deals occur in
North America followed by Europe. Little BO activity occurs outside of North Amer-
ica, Eastern Europe, and Asia. PE is highly correlated with the returns of the S&P 500
index as well as the MSCI World and MSCI Europe equity indexes and may serve as a
poor diversifying agent if added to an all equity portfolio.

Discussion Questions
1. Explain why the PE return results reported in this chapter should be viewed with
some skepticism.
2. Explain whether investors should include PE in a diversified portfolio.
P riv at e E qu it y R e t u rn s 255

3. Identify the factors that affect cross-sectional PE BO activity and identify any global
trends that occurred between 2006 and 2013.
4. Identify the two largest PE markets in the world and factors contributing to the size
of these markets.

References
Aizenman, Joshua, and Jake Kendall. 2008. The Internationalization of Venture Capital and Private
Equity. NBER Working Paper #14344, National Bureau of Economic Research, Inc.
Cochrane, John H. 2005. The Risk and Return of Venture Capital. Journal of Financial Economics
75:1, 352.
Cumming, Douglas, and Uwe Walz. 2010. Private Equity Returns and Disclosure around the
World. Journal of International Business Studies 41:4, 727754.
Diller, Christian, and Christoph Kaserer. 2009. What Drives Private Equity Returns? Fund Inflows,
Skilled GPs, and/or Risk? European Financial Management 15:3, 643675.
Driessen, Joost, Tse-Chu Lin, and Ludovic Phalippou. 2012. A New Method to Estimate Risk and
Return of Nontraded Assets from Cash Flows: The Case of Private Equity Funds. Journal of
Financial and Quantitative Analysis 47:3, 511535.
Groh, Alexander, Liechtenstein Heinrich, and Karsten Lieser. 2008. Global Venture Capital and
Private Equity Attractiveness Index. IESE Business School Barcelona.
Groh, Alexander, Liechtenstein Heinrich, and Karsten Lieser. 2009. Global Venture Capital and
Private Equity Attractiveness Index. IESE Business School Barcelona.
Groh, Alexander, Liechtenstein Heinrich, and Karsten Lieser. 2010. Global Venture Capital and
Private Equity Attractiveness Index. IESE Business School Barcelona.
Groh, Alexander, Liechtenstein Heinrich, and Karsten Lieser 2011. Global Venture Capital and
Private Equity Attractiveness Index. IESE Business School Barcelona.
Groh, Alexander, Liechtenstein Heinrich, and Karsten Lieser. 2012. Global Venture Capital and
Private Equity Attractiveness Index. IESE Business School Barcelona.
Groh, Alexander, Liechtenstein Heinrich, and Karsten Lieser. 2013. Global Venture Capital and
Private Equity Attractiveness Index. IESE Business School Barcelona.
Groh, Alexander, Liechtenstein Heinrich, and Karsten Lieser. 2014. Global Venture Capital and
Private Equity Attractiveness Index. IESE Business School Barcelona.
Herskovitz, Jon. 2013. FACTBOX: BRICS Emerging Powers Grow in Global Strength. Reuters,
March 26. Available at http://www.reuters.com/article/2013/03/26/us-brics-summit-
factbox-idUSBRE92P0G920130326.
ImmadEddine, Gael, and Armin Schwienbacher. 2011. International Capital Flows into the Euro-
pean Private Equity Market. European Financial Management 17:1, 133.
Ippolito, Roberto. 2007. Private Equity in China and India. Journal of Private Equity 10:4, 3641.
Kaplan, Steven N., and Antoinette Schoar. 2005. Private Equity Performance: Returns, Persistence,
and Capital Flows. Journal of Finance 60:4, 17911823.
Kaplan, Steven N., Berk A. Sensoy, and Per Strmberg. 2002. How Well Do Venture Capital Data-
bases Reflect Actual Investments? Working Paper, University of Chicago, Graduate School of
Business.
Kaplan, Steven N., and Per Strmberg. 2008. Leveraged Buyouts and Private Equity. NBER Work-
ing Paper #14207.
Klonowski, Darek. 2011. Private Equity in Emerging Markets: Stacking Up the BRICs. Journal of
Private Equity 14:3, 2437.
Klonowski, Darek. 2013. Private Equity in Emerging Markets: The New Frontiers of International
Finance. Journal of Private Equity 16:2, 2036.
KPMG and SAVCA. 2013. Venture Capital and Private Equity Industry Performance Survey of
South Africa Covering the 2012 Calendar Year. Available at http://www.kpmg.com/ZA/en/
256 p e r f o r m a n c e a n d m e a s u r e m e n t

IssuesAndInsights/ArticlesPublications/General-Industries-Publications/Documents/
Private%20Equity%20survey%202012.pdf.
Phalippou, Ludovic, and Oliver Gottschalg. 2009. The Performance of Private Equity Funds.
Review of Financial Studies 22:4, 17471776.
Preqin. 2014. Private Equity-Backed Buyout Deals and Exits. Available at https://www.preqin.
com/format/private-equity-publications/1/1.
Red Rocks Capital. 2014. Red Rocks Capital Factsheet. March 31. Available at http://www.
redrockscapital.com/.
Seretakis, Alexandros. 2012. A Comparative Examination of Private Equity in the United States and
Europe: Accounting for the Past and Predicting the Future of European Private Equity. Ford-
ham Journal of Corporate & Financial Law 18:3, 613667.
Sommer, Claudia. 2013. Private Equity Investments: Drivers and Performance Implications of Invest-
ment Cycles. Mnchen, Deutschland: Springer Gabler.
Wright, Mike, Luc Renneboog, Tomas Simons, and Louise Scholes. 2006. Leverage Buyouts in the
U.K. and Continental Europe: Retrospect and Prospect. Finance Working Paper, European
Corporate Governance Institute.
Wright, Mike, and Ken Robbie. 1998. Venture Capital and Private Equity: A Review and Synthe-
sis. Journal of Business Finance & Accounting 25:56, 521570.
15
Benchmark Biases in Private
Equity Performance
ROBERT SPLIID
External Lecturer, Copenhagen Business School

Introduction
An investment manager who wants to attract investors often needs a track record showing
a higher return than that offered by most comparable investment opportunities. Some-
times, the risk-adjusted returns take expected stability of returns into consideration.
Mutual funds and hedge funds are usually organized as open-end funds, continu-
ously collecting cash for investments. Private equity (PE) firms typically use closed-end
funds with a maturity of 10 years and with the option of prolonging the fund for another
two years. Thus, PE firms are under pressure to have a good report to tell potential inves-
tors during the short fundraising period.
Raising follow-on funds is easier if the PE firm can refer to a previously successful
fund. This applies to both buyout (BO) funds and venture capital (VC) funds. The
PE firms success is more relevant to investors than the general performance of the
total asset class. When the PE firm raises its first fund, or if the performance of pre-
vious funds is unimpressive, referring to the performance of the whole asset class is
tempting.
In both cases, investors should be aware of potential benchmark biases. The PE firms
own historic performance might appear attractive on a total return basis or compared
with a benchmark, but the relevance of the benchmark to the asset class and invest-
ment style should be considered. The same question applies when comparing PE per-
formance as an asset class with the public stock market. Is the chosen stock market index
comparable from a risk and liquidity perspective? If not, can the index be adjusted to
provide a more appropriate benchmark?
Over the past 20 years, researchers have published studies (i.e., Ljungqvist and
Richardson 2003; Kaplan and Schoar 2005; Lerner, Schoar, and Wongsunwai 2005;
Acharya, Gottschalg, Hahn, and Kehoe 2012) on the overall performance of PE. Some
studies show that PE outperforms public markets while others show underperfor-
mance or find that the performance difference between public and private equity is
insignificant.

257
258 p e r f o r m a n c e a n d m e a s u r e m e n t

Choosing PE funds or any investment fund with a benchmark that points to the high-
est possible fund alpha is tempting. Thus, choosing the right benchmark with the relevant
adjustments is important in order to have a reliable performance analysis. Measuring the
performance of the relevant investment period is also important. Simply measuring per-
formance from the date when capital is called does not necessarily reflect the expectation
of the investor, who must keep money available once making the commitment. Ignoring
the opportunity cost between the commitment and call dates inflates PE performance.
The purpose of this chapter is to provide an overview of the challenges of measuring
performance of PE investments and the difficulties in benchmarking such investments
against public equity. The rest of the chapter is organized as follows. The first section dis-
cusses the importance of providing the appropriate data and benchmark. The data must
be representative of the measured asset class and the risk profile of the benchmark must
be similar to that of the measured assets. The next section outlines measurement meth-
ods, discusses the problems of using the internal rate of return (IRR), and describes
the advantages of total value to paid-in-capital (TVPI) and public market equivalent
(PME) when measuring PE performance. The third section focuses on the relevance of
adjusting for risk and liquidity as benchmarks often represent risk and liquidity profiles
that divert from the measured asset class. The fourth section discusses different types of
selection bias and the importance of adjusting data to create representative samples of
PE investments. The final section offers a summary and conclusions.

Finding the Appropriate Data and Benchmark


The first challenge in assessing performance is to find the appropriate data for the anal-
ysis. If the aim is to analyze the performance of an asset class, a distinction should be
made between BO and VC firms because their investment styles differ. Analysts may
collect data from managers, investors, or public data banks. If the aim is to measure rela-
tive performance, the next challenge is to find a relevant benchmark that reflects the risk
and return profile of the asset class.

B U Y O U T A N D V E N T U R E C A P I TA L
Private equity is the generic term for investment managers buying, controlling, or partly
controlling shares of companies and taking an active position in developing the com-
pany. Viewing buyouts and venture capital as one common asset class understates their
differences.
BO companies are usually mature companies that are financially leveraged by the PE
firm, whereas VC companies are newly established firms with a limited amount of debt.
VC companies consist of four groups depending on their development stage: seed,
start-up, early stage, and expansion.
VC fund managers typically advise their portfolio companies, while BO fund man-
agers are more directly involved in management. Additionally, while BO funds consist
of a stable financing structure for the entire expected holding period, VC funds typically
inject capital for a limited period because VC companies usually undergo more funding
rounds before exit.
B e n c h m a r k B i a s e s i n P E P e r f o r m a n c e 259

F I N D I N G T H E R E L E VA N T D ATA
Analysts can collect data from the investment manager, general partner (GP), limited
partner (LP), investors, or extensive PE databases. The advantage of using data from the
GP is that they come directly from the primary source. Such data are usually gross of
fees, but should be adjusted for all fees. GPs have no interest in reporting net-to-LP per-
formance figures; they may even restrict LPs from providing such information to third
parties (Harris, Jenkinson, and Stucke 2012). LP data are already net of fees. Many LPs
invest in several PE funds and can therefore provide researchers and analysts with data
from different funds. Ljungqvist and Richardson (2003) use data from only one LP to
analyze the performance of funds raised between 1981 and 2001. Although LPs spread
their investments among funds, the funds of the chosen LP are not necessarily represen-
tative of the PE industry.
Lerner et al. (2007) find dramatic differences in investment performance by differ-
ent institutions. For example, endowments have much higher returns on their PE in-
vestments than do investment advisors and banks. The main reason is that endowments
are apparently better in forecasting the performance of follow-on funds and therefore
reinvest less often in the same partnership. The funds in which they reinvest have much
higher performances than those in which they decide not to reinvest. These findings
suggest that endowments use the information they gain as inside investors much better
than other investors.
Using data from only one GP can also create bias problems. Kaplan and Schoar
(2005) find that performance improves with a GPs experience. Thus, generalizations
based on data from a well-experienced GP exaggerate PE performance. Also, collecting
data from GPs or LPs may be cumbersome not only because of their unwillingness to
share data with outsiders but also because their data might be structured differently. A
tempting shortcut is to collect data from one of the large databases containing data from
a wide range of PE firms. The largest and best known PE databases are Thomson Reu-
ters VentureXpert (TVE), Private Equity Intelligence (Preqin), Cambridge Associates,
CEPRES, and Capital IQ.

FINDING THE RIGHT BENCHMARK


PE performance is often measured relative to the public stock market. Most relative
performance analyses use the S&P 500 index as a benchmark because this index is the
most widely accepted stock performance index in the United States. By using the S&P
500 index, however, the capital asset pricing model (CAPM) is assumed to have a beta
of 1, which is rarely the case for any PE portfolio. Ljungqvist and Richardson (2003)
estimate excess IRRs relative to the S&P 500 index and the NASDAQ Composite
index. Based on 73 funds raised between 1981 and 1993, the mean excess IRR is 8.06
percent relative to the S&P 500 index and 6.28 percent relative to the NASDAQ Com-
posite index.
Driessen, Lin, and Phalippou (2012) estimate the relative performance of VCs to the
S&P 500 index and NASDAQ Composite index. Using the S&P 500 index as a bench-
mark results in a beta of 3.21 and an alpha of 1.24, suggesting a low correlation. When
measured against the NASDAQ Composite index, beta decreases to 1.55 while alpha
260 p e r f o r m a n c e a n d m e a s u r e m e n t

increases to 0.55. Thus, these studies suggest that the NASDAQ Composite index is a
more appropriate benchmark for VC funds than the S&P 500 index.
So far, researchers have focused on measuring the relative performance of private
equity to public equity, which is practical if measuring how PE performs as an asset
class. However, if the intent is to measure the performance of a specific PE fund relative
to the whole PE universe, using a PE performance index would be more appropriate.
Two examples of such indexes are the Private Equity Quarterly Index (PrEQIn) and the
Private Equity Performance Index (PEPI).

Measurement Methods
The most common performance measurement is the money-weighted IRR. Yet, when
GPs refer to their successful deals, they often talk about multiples. A multiple describes
how much the investment has increased during the investment period. For example, a
multiple of two indicates that the value of the investment has doubled. This concept is
called total value to paid-in-capital (TVPI). Both IRR and TVPI are difficult to compare
with the time-weighted measurement used for public equity. With the public market
equivalent (PME), money-weighted return metrics can be calculated for PE by mimick-
ing the timing and size of cash flows associated with a PE investment.

I N T E R N A L R AT E O F R E T U R N
The most applied method for measuring return on investment is the IRR. Using IRR
enables investors to compare performance across invested capital and investment hori-
zons because IRR expresses an annualized return. Early PE performance studies apply
both realized and unrealized gains in IRR calculations. More recent studies generally
ignore unrealized gains because they are only (manipulative) estimates (i.e., Ljungqvist
and Richardson 2003; Kaplan and Schoar 2005 Lerner et al. 2005; Groh and Gottschalg
2009; Higson and Stucke 2011; Driessen et al. 2012; Acharya et al. 2012; Korteweg and
Nagel 2013).
The main criticism for using IRR as a PE performance measurement is its assump-
tion that all cash flows are reinvested at the IRR. This assumption makes sense when
calculating the return of a public stock market index because all stocks are held during
the entire calculation period and because dividends may be reinvested in the stocks of
the index. This approach does not work for PE investments, as discussed in the follow-
ing paragraphs.
When investing in a mutual fund, the investor pays the full investment sum up front.
With a PE fund, the investor pays the investment sum in tranches as the GP calls in the
money for specific investments. The GP does not reinvest the proceeds when exiting
the investment but returns them to the investor. Thus, while the mutual fund investor is
fully invested during the whole commitment period, the LP steps up the investment as
commitments are called in for acquisitions and scales down the investment as proceeds
from divestments are returned.
The IRR of a PE investment is a return measured on a smaller amount and invested
for a shorter period than the amount and period for which the investor has made his
commitment. For example, an IRR of 20 percent for two years is better than an IRR of
B e n c h m a r k B i a s e s i n P E P e r f o r m a n c e 261

30 percent for one year, if the returned investment must be reinvested at a risk-free rate
of 2 percent in the second year. Likewise, $100 invested at 20 percent is better than $50
invested at 30 percent, if the remaining $50 is returning only a risk-free rate of 2 percent.
Ljungqvist and Richardson (2003) identify two potential problems of applying the
IRR approach. First, the IRR assumes that early distributions can be reinvested at the
funds IRR, which reduces the differences (both negative and positive) in relative per-
formance. Second, the authors contend that outflows should be discounted at a dif-
ferent and lower rate than inflows; otherwise the IRR will overstate the performance
relative to the true risk profile of the cash flows.
As an alternative, Ljungqvist and Richardson (2003) propose calculating the ex post
net present value (NPV), using the realized cash flows discounted at the risk-free rate
for outflows and at the cost of capital for inflows. They use the Treasury-bond rate with
the same maturity for the inflows and the expected return on the aggregate market for
the outflows.
Although the PE investor is only invested for a part of the commitment period and
only with a part of the committed capital, cash must be held available for coming calls.
Besides, it is highly improbable that the returned money can be reinvested in an equiva-
lent PE fund for the specific amounts and from the specific return dates. The return on the
total committed capital depends on the speed of investment, the amount invested, and
the holding period. In their analysis of 73 liquidated funds raised between 1981 and 1993,
Ljungqvist and Richardson (2003) find that the speed of investment is reflected in the
accumulated average drawdowns, which they estimate to be 16 percent, 36 percent, and
57 percent after the first, second, and third year, respectively. After six years, 90 percent
is invested. Thus, the challenge of measuring the performance of a continuing PE fund is
that investments are realized late in the funds life. The authors maintain that a comparison
between private and public equity requires an equivalent time profile of both investments,
meaning the investment in the stock market must be parallel to the PE funds drawdowns.

TOTA L VA L U E TO PA I D - I N - C A P I TA L
TVPI is a multiple measuring how many times an investment is paid back. All cash out-
flows (investments) as well as inflows (distributions) are discounted separately. TVPI
is the relationship between the values of these two cash flows. TVPI may be split into
two parts: (1) distribution to paid-in-capital (DPI), which reflects realized gains, and (2)
residual value to paid-in capital (RVPI), which measures unrealized gains. GPs often use
TVPI. This measurement ignores the time value of money, making it unsuitable as a rel-
ative performance measure.

P U B L I C M A R K E T E Q U I VA L E N T
Long and Nickles (1996) contend that because a private market investment fund has
achieved an IRR greater than the total return to the S&P 500 index over a particular
time period, this does not necessarily mean the investment has outperformed the stock
index. The problem is that a comparison of the IRRs does not consider the timing of the
cash flows. The authors conclude that a comparison of private market investments with
public market indexes requires an analytical method reflecting the timing and amount
of relevant cash flows.
262 p e r f o r m a n c e a n d m e a s u r e m e n t

Long and Nickles (1996) introduce the index return comparison (IRC), later referred
to as the PME, which measures the performance of a private investment relative to that
of a public stock index. Computing the IRC involves two steps: (1) determining the
IRR of the private investments cash flows and (2) calculating the end value or net asset
value (NAV), of the same cash flows invested in the stock index. Negative cash flows
(investments) are invested in the index, whereas positive cash flows (distributions) are
withdrawn from the index investment. The excess IRR (i.e., the IRR of the cash flows
of the private investments minus the IRR of the same cash flows invested in the stock
index) reflects the relative performance of the private investment. A drawback of this
method is that when stocks are falling, distributions may lead to a negative NAV of
the mimicking index portfolio, benchmarking the PE investment against a short stock
portfolio.
To avoid this effect, Rouvinez (2003) presents the PME + as an alternative perfor-
mance measure. Instead of redeeming the full amount bringing the NAV into negative
territory, he assumes that only such a proportion of the distribution is paid that NAV
does not turn negative. The downside is that cash flows of the two portfolios are not
perfectly matched.
Kaplan and Schoar (2005) present a PME based on the TVPI metric. All cash out-
flows and inflows are discounted separately to the S&P 500 index. The PME is then
defined as the relationship between the values of the two cash flows and reflects the
relative performance of the portfolio to the S&P 500 index. When PME is greater
than 1, the portfolio has outperformed the stock index; whereas, a PME less than 1
shows underperformance. Using the PME to estimate the relative performance of 746
funds, the authors conclude that both fund types underperform when compared to the
public stock market with a PME of 0.97 for BO funds and 0.96 for VC funds.
Mozes and Fiori (2012) use a slightly different method to measure the relative per-
formance of PE, called the relative compounded return (RCR). RCR also takes the com-
mitted but non-called capital into account and assumes that distributions are invested
in cash, not in the stock index. Thus, when calculating the RCR, the entire committed
capital is invested. The performance of the PE investment is estimated as though the
called amounts were invested in the specific underlying portfolio companies, but with
the non-called amount and the distributions invested in cash at the risk-free rate. The
performance of the stock index is based on an invested capital equivalent to the commit-
ted capital. The RCR is finally computed as the end value (NAV) of the PE investment
divided by the end value of the stock index. As with the PME, an RCR greater than 1
implies that PE performs better than public equity and vice versa.
The strength of the PME and RCR is that they both measure return on a fixed amount
for the whole investment horizon. A high IRR may result from a small investment over a
short time. Thus, PE funds that employ quick flips may rank lower on PME and RCR
than on IRR measures.

Risk and Liquidity Adjustments


Measuring an assets performance relative to a benchmark is only appropriate if both
have the same risk and liquidity profile. If the benchmarks risk is lower than the assets
B e n c h m a r k B i a s e s i n P E P e r f o r m a n c e 263

risk, the assets alpha is overestimated. This is also the case if the liquidity is higher for
the benchmark than for the asset. Therefore, the asset or the benchmark performance
should be corrected to adjust for these differences.
Calculating risk requires access to frequent valuations. This calculation is not a
problem if benchmark is a stock index. PE, however, does not generally deliver reliable
market values. The necessary risk and liquidity adjustments, as well as the valuation
problems of non-exited portfolio companies, are discussed next.

RISK ADJUSTMENTS
Performance analyses are often based on the assumption that the PE sample portfolio
has a beta of 1 (i.e., the risks of the PE portfolio and the benchmark public stock index
have the same profile). Ljungqvist and Richardson (2003) and Kaplan and Schoar
(2005) base their analyses on this assumption.
PE-owned companies generally have more leverage than public companies, resulting
in a higher PE investment risk. Kaplan and Schoar (2005) acknowledge that PE-owned
companies may understate the market risk if the betas of the funds are higher than 1.
Groh and Gottschalg (2009) estimate the average debt-to-equity ratio to be 2.94 for
BOs at the time of investment and 1.28 at the time of exit.
Ljungqvist and Richardson (2003) use industry betas to adjust for business risk.
They assign each portfolio company to one of 48 industry groups using industry equity
betas estimated from 1989 to 1994. The authors, however, cannot correct for leverage
due to a lack of leverage data on the PE-owned companies. Instead, they analyze how
much less equity the portfolio companies should have compared with the average of the
relative industries to have zero risk-adjusted over-performance. Ljungqvist and Rich-
ardson conclude that the BO funds create excess value as long as they do not use less
than 47.6 percent of equity used by the firms in the reference industries.
Gottschalg, Phalippou, and Zollo (2004) also consider business risk. They consider
leverage risk by leveraging the initial equity beta calculations by debt-to-equity ratios
of 3. In their analysis of 133 PE transactions completed by 41 different later stage PE
funds in the United States between 1984 and 2004, Groh and Gottschalg (2009) adjust
for both business risk and leverage risk. They measure business risk by comparing
public markets in 116 peer groups with at least three companies in each. The authors
determine leverage risk by the capital structure of the PE transaction, usually character-
ized by diminishing leverage due to debt redemption. They know the risk at entry and at
exit on each portfolio company and assume the leverage changes linearly. For write-off
transactions, Groh and Gottschalg assume a constant leverage over the total holding
period. To reach the relevant leverage level, they also assume that borrowing and lend-
ing are possible in unlimited amounts at the risk-free rate. Introducing a constant credit
spread of 3 percent does not significantly change the resulting alpha.
Groh and Gottschalg (2009) create a mimicking strategy by investing an equiva-
lent amount in the S&P 500 index portfolio and leveraging it with borrowed funds to
achieve an equal systematic risk exposure. They adjust the systematic risk of the mim-
icking strategy each year until exit to secure risk parity. Thus, the PE transaction and
mimicking strategy represent two cash flows with identical risk. The authors compare
the performances of these two strategies through regression analysis of their IRRs.
264 p e r f o r m a n c e a n d m e a s u r e m e n t

They obtain a Jensen alpha from the intercept of the regression, giving a risk-adjusted
measure of performance. Nevertheless, Groh and Gottschalg argue that PE should be
regarded as a separate asset class with its own risk/return profile because investments
cannot easily be tracked by liquid, publicly traded securities. In all of their scenarios, the
correlation coefficient between PE returns and the mimicking strategy is never higher
than 0.26.
Phalippou and Zollo (2005) also calculate beta similar to Ljungqvist and Richard-
son (2003). For each fund, they estimate beta of the investments based on the industry
of each portfolio company and its leverage. The authors assume that the leverage de-
creases from a debt-to-equity of 3 upon entry to the leverage that prevails in the industry
on exit. Although the average beta of each fund is not the same as the fund beta, it can be
regarded as a proxy. Beta is estimated at 1.7 for VCs and 1.6 for BOs; nevertheless, the
authors do not find systematic risk to be a driver of fund performance. Although beta
is positively related to performance, the relationship is not statistically significant. Thus,
findings based on a beta of 1 are optimistic.
Using the same industry-match method, Phalippou and Gottschalg (2009) find an
average beta of 1.3, which they use to adjust the discount rate. As a result, the average
TVPI falls from 0.99 to 0.84. The authors, however, do not adjust for differences in lev-
erage between public and private equity.
Robinson and Sensoy (2011) find a highly convex relationship between beta and
relative performance. Going from a beta of 0 to a beta of 1 cuts the estimate of excess
performance of BO funds over the funds life from 57 percent to 18 percent. Changing
the beta to 1.5 lowers the excess performance to 12 percent. Similar conclusions hold
for VC funds.
Driessen et al. (2012) define the mimicking fund as a leveraged position of the S&P
500 index. They ask which beta (i.e., leverage) of the mimicking fund best mimics a
PE fund. The mimicking fund should have the same cash flow as the PE fund. That is,
whenever the investor is called, the same amount is invested in the mimicking fund, and
whenever distributions are paid, the same amount comes out of the mimicking fund.

LIQUIDITY ADJUSTMENTS
While public equity is liquid because it is tradable on a stock exchange, PE is not. Thus,
PE should yield a higher expected return to compensate for the cost of illiquidity.
Ljungqvist and Richardson (2003), who find an excess return of PE compared with the
S&P 500 index, argue that a part of this excess return should be regarded as compensa-
tion for PEs extreme illiquidity.
Franzoni, Nowak, and Phalippou (2012) find that the seemingly high performance
of PE investments may be largely explained as compensation for the different risk fac-
tors to which returns are exposed, and liquidity risk is one important source of the risk
premium. Their main idea is that the relationship between market liquidity and PE
returns reflects the effect of funding liquidity on PE performance. They estimate the
unconditional liquidity risk premium to be about 3 percent a year, with a total risk pre-
mium of roughly about 18 percent.
Srensen, Wang, and Wang (2013) find a substantial cost of illiquidity in PE. Part
of the risk of PE is spanned by liquid, publicly traded assets and hence commands the
B e n c h m a r k B i a s e s i n P E P e r f o r m a n c e 265

standard risk premium for systematic risk. The remaining part of the risk is not spanned
by the market due to illiquidity, and thus the investor demands a liquidity risk premium.
The authors estimate the premium to be around 1.5 percent.

T H E S TA L E P R I C I N G P R O B L E M
As previously mentioned, an investments risk is generally measured by the volatility of
its expected return. Measuring risk for publicly traded stocks is relatively easy because
they are quoted on a stock exchange and therefore have a market price. The values of PE
investments are typically estimated quarterly by the GP and do not necessarily reflect
true market prices.
The stale pricing problem arises when a deviation occurs in the measured interval
between the returns of two investments used in regression analysis. This problem ap-
pears when conducting a regression analysis with thinly traded/non-traded stocks and
liquid stocks/stock indices. The result underestimates beta and overestimates alpha.
The problem may be addressed by looking at the lagged beta effect. As the valuations
of PE-owned companies are adjusted on a lagged basis to the stock market, the full cor-
relation between public- and private-owned companies should also include the lagged
correlation.
Emery (2003) analyzes the correlation among PE returns based on data from Ven-
ture Economics and the return of major U.S. stock indices between 1986 and 2001.
He finds that BO quarterly returns exhibit the greatest correlation with S&P 500 index
returns, while VC quarterly returns exhibit the greatest correlation with NASDAQ
Composite index returns. Nevertheless, public returns explain little of the movement in
private returns. The NASDAQ Composite index explains 35 percent of the movement
in VC returns, while the S&P 500 index does not explain any movement in BO returns.
However, when including lagged public returns, the VC/NASDAQ Composite index
explanatory power increases to 56 percent while that of BO/S&P 500 index explanatory
power increases to 10 percent.
Woodward (2004) addresses the stale pricing problem by including leading and lag-
ging market return values as well as the contemporaneous return, and sums the resulting
coefficients to get the correct beta. For individually traded stocks, she finds that two
days of leading returns and two days of lagging returns are usually enough to capture the
true correlation between a public stock and the overall market return.
The stale pricing problem is more pronounced for PE because GPs only evaluate
their portfolio companies quarterly. Woodward (2004) distinguishes between VC
funds in which valuation is typically based on the last funding round of the specific
company, and BO funds in which company valuation is based on future expected earn-
ings. This gives the GPs great freedom in reporting.
Anson (2013) finds significant serial correlation in up to four lagged quarters em-
bedded in PE returns, showing that up to one year of prior returns has some influence
on the current return on PE. He tests data from Cambridge Associates U.S. Private
Equity Index from 1986 to 2013 against the Russell 1000 index and the Russell 2000
index to exclude a capitalization range or size effect. Using a single-period model on
the Russell 1000 index results in a beta of 0.4 and an alpha of 2.6 percent, whereas the
multi-period model doubles the beta to 0.8 and reduces the alpha to 1.9 percent. Similar
266 p e r f o r m a n c e a n d m e a s u r e m e n t

results are reached with the Russell 2000 index. The results suggest that the stale pricing
effect underestimates beta and overestimates alpha.

VA L UAT I O N C H A L L E N G E S
While differences in accounting principles among countries decrease, the differences in
valuation methods persist. Ljungqvist and Richardson (2003) find that particular first-
time fund managers may be aggressive in their valuation by not writing down poorly
performing companies or even overstating the value of ongoing ones, especially in dif-
ficult times.
Woodward (2004) finds that GPs of VC have an incentive to raise a new round of
funding when values are rising so they can print a trade at a higher price. This presents
the opportunity to elevate the valuation of the whole portfolio, as the multiples of the
most recent trades carry a high weight in the overall valuation. When the market is fall-
ing, GPs might delay a round of funding to delay printing a trade at a lower value.
Kaplan and Schoar (2005) do not rely on interim IRRs measured before liquidating
a fund because they are based on the fund managers subjective valuations. Upon expira-
tion of the fund, however, the authors convert full residual values into a cash equivalent
before calculating total performance.
The accounting standards and the legal framework are the most important factors
of the reporting behavior of GPs. Cumming and Walz (2010) find that the average
unrealized IRRs are significantly higher in countries with weak legal environments
and less regulated accounting standards. Phalippou and Gottschalg (2009) find
that accounting values reported by mature funds mostly represent living-dead in-
vestments (i.e. self-sustaining investments that do not produce adequate multiples
of return). They therefore choose to write off all residual values before estimating
performance.
Previously, companies were valued at the purchase cost of investment and writ-
ten down if the recoverable value fell below costs, or adjusted if there had been a new
round of financing, recapitalization, or another injection of capital. Since 2006, industry
GAAP (FAS 157/ASC 820, IAS 39) requires portfolio companies to record companies
at fair value (i.e., written up or down to the current value). GPs still have a certain leeway
for finding fair value as they can use discounted cash-flow valuation, earning multiples,
and option pricing models. Also, the value can be adjusted with a liquidity discount as
well as a control premium.
As Anson (2013) reports, the new accounting regime has not resulted in any mate-
rial impact on the lagged beta effect. He tests whether the lagged beta before the change
(19862007) is different from the lagged beta after the change (20082013). His re-
sults show that GPs mark down their portfolios quickly in down markets (exhibiting a
beta of 1.05) and mark them up slowly in up-markets (exhibiting a beta of 0.86). Anson
attributes this behavior to the fact that the LPs monitor the GPs, wanting them to be
conservative in their valuations.
Driessen et al. (2012) construct an econometric model for estimating final market
value. For liquidated funds, the model finds market values to be close to the NAV. For
non-liquidated funds beyond the typical liquidation age (10 years), however, they esti-
mate the values to be only 30 percent of the self-reported NAV.
B e n c h m a r k B i a s e s i n P E P e r f o r m a n c e 267

This result explains that different ways are available to interpret final NAV calcu-
lations. Kaplan and Schoar (2005) assume that final NAVs reflect market values. By
contrast, Driessen et al. (2012) estimate the NAVs are 30 percent of reported value,
while Phalippou and Gottschalg (2009) assume the final NAVs of mature funds are
worthless.

MEASURING GROSS OR NET OF FEES


Performance can be measured gross or net of fees. Data from GPs are usually gross of
fees, while LPs deliver data net of fees. Which of the two measurements is relevant de-
pends on the aim of the analysis. If the goal is to measure how successful GPs are at cre-
ating value, measuring performance gross of fees is more appropriate because the return
distribution is unimportant. If the focus is on investors asset allocation, net return to
LPs is a more relevant measure.
GPs operate with two standard fees: a management fee and carried interest, which
is the accumulated performance fee. The management fee is usually 2 percent of the
committed capital until all capital is invested and subsequently 2 percent of the in-
vested capital. Usually, the carried interest is 20 percent of the return, once the return
exceeds a hurdle rate of 8 percent. Some PE firms charge a transaction fee and/or a
monitoring fee.
Kaplan and Schoar (2005) find that PE funds performed worse than the S&P 500
index net of fees with a PME of 0.97 for BO funds and 0.96 for VC funds. Thus, on a
gross-of-fees basis, the PME must be positive. According to Phalippou and Gottschalg
(2009), fees reduce alpha by 6.79 percent a year, from 2.96 percent to 3.83 percent.
They point to the possibility that investors have a biased view of performance since only
performance gross of fees is reported during fundraising.
Srensen et al. (2013) find that the fund manager needs to generate sufficient value
to cover the high costs of management fees and carried interest. Without leverage, the
break-even alpha is 1.5 percent each year. Assuming a debt-to-equity ratio of 3, the
break-even alpha reduces to around 1 percent.

Sample Selection Bias


If the sample affects the performance results because it is not representative, the sample
has selection bias. Several factors affect measured performance. If these factors are rep-
resented differently in the sample than in the total population, a correction should be
made for selection bias. In PE samples, the most frequent selection biases are due to
differences in survivorship, vintage year, fund size, update frequency, and skill of the
manager or investor. These biases are discussed in this section.

SURVIVORSHIP BIAS
Data providers must rely on voluntary reporting because GPs and LPs have no obli-
gation to report to them. According to Groh and Gottschalg (2009) and Harris et al.
(2012), this can lead to survivorship bias because poor performers stop reporting. It can
268 p e r f o r m a n c e a n d m e a s u r e m e n t

also lead to backfill bias, which is not reporting fund performance from inception, but
instead reporting after establishing a positive track record.
Using a sample of 746 funds from the TVE database, Kaplan and Schoar (2005)
find the net returns of PE funds are almost equal to the S&P 500 index, while the BO
funds among them perform slightly worse. The authors only sample funds that are no
longer active, defined as officially liquidated or with no cash-flow activity within the
last six quarters. Higson and Stucke (2012) criticize this definition. They find that
the approach oversamples incomplete funds by allowing all funds with truncated data
into the sample, while excluding mature funds with minor cash-flow activity. This ap-
proach places a downward bias on the result and underestimates the performance of
PE funds.

V I N TA G E B I A S
Although the performances of PE funds and the public stock markets are highly corre-
lated, the cyclicality of PE returns is substantial. Historically, much higher returns have
occurred for funds raised in the first half of the 1980s, 1990s, and 2000s (vintage years
or vintages) rather than toward the end of each decade (Higson and Stucke 2012). Vin-
tage year is defined as the year in which a PE firm begins to make an investment. This
year is important because typically a PE firms performance is measured relative to its
vintage year.
Two trends emerge over the past 30 years: (1) a downward movement in absolute
returns and (2) less variation among the performances based on data from TVE, Preqin,
and Cambridge Associates. The main reason for the reduced variation is that Preqin and
Cambridge Associates only have data from a relatively limited number of funds com-
pared with TVE until the mid-1990s.
Harris et al. (2012) estimate the median IRR for PE funds per vintage year based
on all three databases. Returns on VC increased steadily from 1981 to 1996 vintages
but dropped to around zero from 1998 and onward due to the burst of the dot-com
bubble. BO funds show better performance but also exhibit a downward trend over the
full period.
Performance depends on the vintage year. Funds launched in boom years tend to
underperform because GPs overpay for companies. Ljungqvist and Richardson (2003)
find fund inflows to be the only significant determinant of excess IRR in their specifica-
tions. The more money raised in the funds vintage year, the worse is the funds following
performance because more funds are chasing the same deals.
Kaplan and Schoar (2005) find that new partnerships are more likely to be started in
periods after the industry has performed well. However, funds raised in boom times and
partnerships started during booms are less likely to raise follow-on funds, suggesting
that these funds likely perform poorly.
Phalippou and Zollo (2005) and Jegadeesh, Krussl, and Pollet (2009) find that
PE performance is pro cyclical. Performance increases with the average gross domestic
product (GDP) growth rate and decreases with the average interest rates that prevailed
during the investments life. By contrast, Mozes and Fiore (2012) find investing in PE
is preferable when large amounts are allocated to BOs because these periods coincide
with stock market peaks. Because BO funds usually have higher returns during weak
B e n c h m a r k B i a s e s i n P E P e r f o r m a n c e 269

stock markets, they outperform stocks when stock markets have peaked. Mozes and
Fiore find a significant vintage effect on PE performance compared to stock market
performance. The strongest vintages for VC funds, both in absolute terms and relative
to the S&P 500 index, are the 19931997 vintages. The 19992006 vintages are the
weakest in absolute terms while the 20022006 vintages are the weakest compared to
the S&P 500 index. For BO funds, absolute performance is weakest for the 20042006
vintages. Relative to the S&P 500 index, performance is weakest for the 19921995
vintages and strongest for the 19992002 vintages.

SIZE BIAS
Various authors identify a size bias. For example, Kaplan and Schoar (2005) and Phalip-
pou and Gottschalg (2009) find that large funds produce higher total returns than small
funds. Mozes and Fiore (2012) compare the performance of equal-weighted and size-
weighted indices for BO and VC funds. For both investment styles, they find that the
size-weighted index outperforms the equal-weighted index, suggesting that large funds
perform better than small funds. Driessen et al. (2012) reach the same conclusion. They
do not find any relationship between alpha and size, but they find beta to be significantly
and positively related to size. Thus, the higher return of large funds is due to higher risk
exposure rather than higher abnormal performance.

U P D AT E B I A S
If a GP stops reporting distributions to investors, the residual value (RV) will remain
constant, even after the fund has been liquidated. The result can be an overvaluation if
the RV has been written off or its real value is less than reported. If the residual compa-
nies are worth more than reported or have exited at a higher value than the RV, perfor-
mance is underestimated.
Phalippou and Gottschalg (2009) report RVs for 462 of 852 funds; all of the funds
should have been liquidated given their age of more than 10 years. These funds report
a total RV of 43 percent of the overall invested amount. Writing off the RVs instead of
treating them as correct valuations reduces the TVPI by 0.07 from 1.01 to 0.94.
Stucke (2011) analyzes the TVE database to identify signs of corruption. He finds
that about 40 percent of the funds stopped being updated at a certain point during their
active lifetime. Their data records were truncated and the value of RVs were simply car-
ried forward and remained constant. Several reasons could explain why fund managers
stop reporting performance results including bad performance, a change in information-
sharing policy, or negligible last payments.

SKILL BIAS
Using reported returns from one GP or LP could also lead to selection bias due to the
learning effect. GPs might have above-average skills in finding and developing targets,
and LPs might be better than other investors at selecting the most successful PE funds.
These skills are learned through years of experience and their impact on investment se-
lection ability is therefore deemed the learning effect.
270 p e r f o r m a n c e a n d m e a s u r e m e n t

Ljungqvist and Richardson (2003) and Phalippou and Zollo (2005) find that small
and inexperienced funds have significantly lower performance. Nevertheless, such
funds manage to find investors for two reasons: (1) they are the only funds accessible to
less sophisticated investors and (2) some funds offer the highest potential commercial
side-benefits for investors (i.e., consulting work and underwriting debt or equity issues).
Kaplan and Schoar (2005) find that funds with higher sequence numbers (i.e., funds
with many preceding investment funds managed by the same GP), generate significantly
higher returns. GPs, whose funds outperform the industry in one fund, are likely to outper-
form the industry in the next. Underperforming GPs are likely to repeat. Thus, the authors
suggest a certain learning effect exists, as good managers tend to improve with experience.
Phalippou and Gottschalg (2009) analyze an extended version of Kaplan and Scho-
ars (2005) sample but correct for sample selection. Their evidence shows that funds be-
longing to a certain performance tercile have a greater than 40 percent probability that
the next fund will belong to the same performance tercile. A lack of consensus exists on
the GP learning effect. If such an effect exists, follow-on funds would perform better the
later they came in the sequence. Existing evidence provides conflicting results. For ex-
ample, Gompers and Lerner (2000), Kaplan and Schoar (2005), Phalippou and Zollo
(2005), and Aigner, Albrecht, Beyschlag, Friedrich, Kalepsky, and Zagst (2008) find
a strong correlation among the performances of a PE managers successive funds. By
contrast, Mozes and Fiore (2012) do not find any correlation among the performances
of successive funds sponsored by the same PE firm.

OUTLIERS AND WEIGHTING


The most common way of measuring performance is by calculating the average perfor-
mance based on value weighting. This method corresponds to the one used when calcu-
lating stock indices such as the S&P 500 index. A disadvantage of using averages is that
outliers can have an excessive influence. A few outliers can move the average (i.e., mean)
so that the measurement is no longer meaningful. This problem can be avoided by using
the median, which is a better proxy for the typical performance. Value-weighted meas-
ures can also lead to biased results if a few large funds dominate the total measurement.
In their sample of PE funds, Mozes and Fiore (2012) find an average annual IRR of 8.5
percent while the median is only 3.2 percent.
If the purpose is to measure the return of all PE investments, value weighting is the
preferred method because it shows the average return on total invested capital. Yet, if
the purpose is to measure the average managers ability to create performance, equal
weighting makes more sense because it shows the performance of the average manager.
Kaplan and Schoar (2005) estimate the PME during 19802001 to be 1.05 when value
weighting (over performance), but 0.96 when equal weighting (underperformance).

Summary and Conclusions


Researchers have analyzed how PE investments perform relative to stock investments
but reach different conclusions. Measuring relative PE performance is difficult because
of a lack of universally accepted measurements.
B e n c h m a r k B i a s e s i n P E P e r f o r m a n c e 271

When analyzing the PE industry, distinguishing between investments in new com-


panies (VCs) and mature companies (BOs) is important because the risk and return
patterns of these two investment styles differ. Performance analysis of the asset classes
requires representative data. Some data providers offer valid data going back 20 to
30 years at least for U.S. PE firms.
Relative performance involves measuring an asset class or asset performance against
a relevant benchmark. When measuring PE performance relative to stocks, researchers
often choose the S&P 500 index as a benchmark. However, a more appropriate index
would be one in which a risk and return profile matches the measured asset or asset
class. Thus, measuring VC against the NASDAQ Composite index rather than the S&P
500 index is more appropriate.
When measuring the relative performance of PE funds, the aim is to measure the
fund against other PE funds. A PE index could be used as a benchmark, although this
could understate the importance of a vintage year. As a result, using a weighted aver-
age of PE funds with the same vintage year as the measured fund provides the fairest
picture.
The consensus is that IRR does not give the full picture of relative performance be-
cause this measure fails to consider investment length and non-called commitments.
PME considers the investment length and RCR also includes the non-called commit-
ments. When measuring an assets performance relative to a benchmark, both measured
units performance must be adjusted to equalize risk and liquidity.
Risk can be measured by the volatility of the asset and the benchmark. This measure-
ment is easy for a public stock and for stock indices because of frequent pricing. How-
ever, PE-owned companies are valued by the GPs and usually only once a quarter. This
methodology creates a stale pricing problem because values of PE funds and the stock
index benchmark are not synchronized thereby underestimating beta and overestimat-
ing alpha. Including lagged beta effects provides a solution to this problem, but it does
not solve the problem of unclear rules for company valuations. Despite stricter interna-
tional rules, GPs still have much leeway to delay adjustments in the valuation. The only
safe performance measurement is to include only mature funds.
GPs typically report performance gross of fees, despite investors only receiving re-
turns net of fees. Although the performance gross of fees measures the GPs ability to
create value, the performance net of fees is what should be compared with public equity,
in which fees are negligible. Measuring PE performance includes selection bias effects.
For example, survivorship bias appears when PE firms stop reporting the performance
of less successful funds. Vintage year bias describes the effect by over- or underweight-
ing specific vintages, as the vintage year plays an important role in fund performance.
Thus, PE firms raising funds in boom years often pay too much for companies and invest
more rapidly compared to those raising funds in downturn periods. Although concen-
trating PE measurement on large funds is easier, this results in bias effects because large
funds create higher returns than small funds. Update bias appears when PE firms do
not update the NAV of their portfolio companies. Because non-updated NAVs are often
overvalued, they should be written down or omitted.
Skill selection bias may occur due to both GP and LP skills. Research suggests that
GPs who outperform with one fund tend to outperform with their follow-on fund.
Some LPs are apparently better at selecting well-performing PE funds than others.
272 p e r f o r m a n c e a n d m e a s u r e m e n t

Using only funds managed by one GP or only funds in the investment portfolio of one
LP could lead to skill selection bias.
Outliers could be a problem when measuring averages because a few outliers can
move the mean to a level that is unrepresentative of the sample. This can be avoided by
using the median instead of the mean as a measure of central tendency.
Finally, the analyst can use either value or equal weightings. Value weighting illus-
trates how one dollar invested in PE performs on average while equal weighting depicts
the average PE managers ability to create value.

Discussion Questions
1. Discuss the pros and cons of measuring PE performance using IRR, PME,
and RCR.
2. Discuss the factors to consider when choosing a benchmark, the drawbacks of using
a PE index as a benchmark, and how to adjust the index to compensate for these
shortcomings.
3. Explain the difference between measuring performance based on return gross of
fees and net of fees and why each measure is used.
4. Explain the stale pricing problem and how to address it.
5. Identify the most important types of selection bias and explain their effect if the
sample is not corrected for them.

References
Acharya, Viral V., Oliver F. Gottschalg, Moritz Hahn, and Conor Kehoe. 2012. Corporate Gover-
nance and Value Creation: Evidence from Private Equity. Review of Financial Studies 26:2,
368402.
Aigner, Philipp, Stefan Albrecht, Georg Beyschlag, Tim Friedrich, Markus Kalepsky, and Rudi
Zagst. 2008. What Drives PE? An Analysis of Success Factors for Private Equity Funds. Jour-
nal of Private Equity 11:4, 6385.
Anson, Mark. 2013. Performance Measurement in Private Equity: Another Look at the Lagged
Beta Effect. Journal of Private Equity 7:1, 2944.
Cumming, Douglas, and Uwe Walz. 2010. Private Equity Returns and Disclosure around the
World. Journal of International Business Studies 41:4, 727754.
Driessen, Jost, Tse-Chun Lin, and Ludovic Phalippou. 2012. A New Method to Estimate Risk and
Return of Nontraded Assets from Cash Flows: The Case of Private Equity Funds. Journal of
Financial and Quantitative Analysis 47:4, 511535.
Emery, Kenneth. 2003. Private Equity Risk and Reward: Assessing the Stale Pricing Problem. Jour-
nal of Private Equity 6:2, 4350.
Franzoni, Francesco, Eric Nowak, and Ludovic Phalippou. 2012. Private Equity Performance and
Liquidity Risk. Journal of Finance 67:6, 23412373.
Gompers, Paul, and Josh Lerner. 2000. Money Chasing Deals? The Impact of Fund Inflows on Pri-
vate Equity Evaluations. Journal of Financial Economics 55:2, 281325.
Gottschalg, Oliver, Ludovic Phalippou, and Maurizio Zollo. 2004. Performance of Private Equity
Funds: Another Puzzle? Working Paper, INSEAD.
Groh, Alexander, and Oliver Gottschalg. 2009. Risk-Adjusted Performance of Private Equity In-
vestments. Presented at the 10th Symposium of Finance, Banking and Insurance in Karlsruhe.
B e n c h m a r k B i a s e s i n P E P e r f o r m a n c e 273

Harris, Robert, Tim Jenkinson, and Rdiger Stucke. 2012. Are Too Many Private Equity Funds Top
Quartile? Journal of Applied Corporate Finance 24:4, 7789.
Higson, Chris, and Rdiger Stucke. 2012. The Performance of Private Equity. Working Paper,
London Business School and University of Oxford.
Jegadeesh, Narasimhan, Roman Krussl, and Joshua Pollet. 2009. Risk and Expected Returns of
Private Equity Investments: Evidence Based on Market Prices. Working Paper 15335, NBER.
Kaplan, Stephen, and Antoinette Schoar. 2005. Private Equity Performance: Returns, Persistence
and Capital Flow. Journal of Finance 60:4, 17911823.
Korteweg, Arthur, and Stefan Nagel. 2013. Risk-Adjusting the Returns to Venture Capital. Work-
ing Paper 19347, NBER.
Lerner, Josh, Antoinette Schoar, and Wan Wongsunwai. 2007. Smart Institutions, Foolish Choices:
The Limited Partner Performance Puzzle. Journal of Finance 62:2, 731764.
Ljungqvist, Alexander, and Matthew Richardson. 2003. The Cash Flow, Return and Risk Charac-
teristics of Private Equity. Working Paper 9454, NBER.
Long, Austin M., and Craig Nickels. 1996. A Private Investment Benchmark. University of Texas
Investments Management Company.
Mozes, Haim A., and Andrew Fiore. 2012. Private Equity Performance: Better Than Commonly
Believed. Journal of Private Equity 15:3, 1932.
Phalippou, Ludovic, and Oliver Gottschalg. 2009. The Performance of Private Equity Funds.
Review of Financial Studies 22: 4, 17471776.
Phalippou, Ludovic, and Maurizio Zollo. 2005. What Drives Private Equity Fund Performance.
Working Paper, University of Amsterdam.
Robinson, David T., and Berk A. Sensoy. 2011. Cyclicality, Performance Measurement, and Cash
Flow Liquidity in Private Equity. Working Paper 17428, NBER.
Rouvinez, Christophe. 2003. Private Equity Benchmarking with PME+. Venture Capital Journal
43:8, 3438.
Srensen, Morten, Neng Wang, and Jinqian Wang. 2013. Valuing Private Equity. Discussion Paper
No. 04/20122041. Netspar. Available at http://ssrn.com/abstract=2352129.
Stucke, Rdiger. 2011. Updating History. Working Paper, University of Oxford.
Woodward, Susan E. 2004. Measuring Risk and Performance for Private Equity. Working Paper,
Sand Hill Econometrics.
16
Return Persistence
Finding Top Quartile Managers
ADRIAN OBERLI
Research Fellow, Harvard Business School

Introduction
Although past superior performance does not guarantee future outstanding returns,
early research in the field by Kaplan and Schoar (2005) shows a surprisingly strong
correlation between predecessor and successor private equity (PE) fund returns. The
best fund managers of predecessor funds often appear in the top quartile again with
their successor funds for both buyout (BO) funds and venture capital (VC). Unlike PE
funds, mutual funds do not show any evidence of top-quartile return persistence (Car-
hart 1997; Berk and Tonks 2007).
Performance differences between top-quartile and bottom-quartile funds are large
for PE funds. Between 1984 and 2004, the spread between top-quartile and bottom-
quartile funds was close to 2,000 basis points based on the internal rate of return
(IRR) according to return data series obtained from Preqin, a database for alternative
assets (Boyd 2012). Therefore, fund manager selection can result in returns ranging
from exceeding the long-term stock market yield (top-quartile funds) to being unable
to even return capital (bottom-quartile funds). Whether an investor beats the long-
term stock market yield therefore depends widely on fund selection and access to the
top-performing general partners (GPs).
More recent empirical research questions the previous findings of top-quartile return
persistence. For example, using data after 2000, Harris, Jenkinson, Kaplan, and Stucke
(2013) show that persistence of BO fund performance has decreased. In practice, many
PE investors still pay attention to the top-quartile label proudly presented by a contin-
uously growing number of fund managers. However, does the past performance of top-
quartile managers justify recommitment from investors? Is past top-quartile performance
a sign of future outperformance? This chapter evaluates the empirical evidence on return
persistence for PE funds for the purpose of identifying future top-quartile PE managers.
The rest of the paper consists of four sections. The first section focuses on defining
various return and performance measures, while the following section reports empiri-
cal evidence on the persistence of these measures and discusses potential reasons for

274
F i n d i n g To p Q u a r t i l e M a n a g e r s 275

the phenomenon. The third section discusses the constraints of implementing a top-
quartile approach when selecting fund managers and suggests following a holistic and
thorough due diligence process. The final section summarizes and draws conclusions on
identifying future top-quartile managers.

Private Equity Return Measures


Evaluating the return persistence of a PE fund and classifying the partnership into quar-
tiles first requires an adequate measure of return. Evaluating the returns of PE funds
involves considering the specific characteristics of the asset class.

CHARACTERISTICS OF THE ASSET CLASS


Investors, called limited partners (LPs), commit a certain amount of capital to PE funds
at fund inception. GPs search for portfolio companies in which they invest and call
money from LPs when needed, up to the amount committed by the LPs (Phalippou and
Gottschalg 2009). When a divestment occurs, the GP distributes the proceeds to its LPs
less fees. The timing of these cash flows is by its nature unknown ex ante. A fund gener-
ally has a life of 10 years, but it can be extended by typically two or more years if both
GPs and LPs agree. Stakes in PE funds are non-tradable, although in the aftermath of the
financial crisis of 20072008, a remarkable amount of secondary transactions appeared.
To inform LPs on the value of their investments, GPs self-report a quarterly net asset
value (NAV) reflecting the value of continuing investments (i.e., the investment port-
folio companies). As portfolio companies do not trade publicly, measuring NAVs ob-
jectively is difficult and strategic valuation decisions of the fund managers can affect
valuations. Often, portfolio companies remain valued at cost for the first years after ac-
quisition and later are valued using public market comparisons in the form of multiples.
An ideal approach to objectively measuring return would not rely on NAVs. Several
studies use various methods to avoid this problem by inferring the BO funds reported
returns only from its cash flow history (Kaserer and Diller 2004; Artus, Teletche, Kas-
erer, Diller, Dupont, and Guennoc 2004).
The dominantly practiced methods used to measure returns include computing cash-
on-cash multiples and the IRR. Given the limits of cash-on-cash returns and IRR calcula-
tions, an approach has evolved from academia called the public market equivalent (PME),
which compares the result of PE investments to that of publicly traded companies.

C A S H - O N - C A S H M U LT I P L E S
A cash-on-cash multiple is the ratio of money returned by the total amount invested. One
common variant, known as distribution to paid-in capital (DPI), looks at the ratio of
the capital returned to LPs to the funds initially provided. Another approach is to use
the ratio of the capital returned to the LPs and the current value of the funds hold-
ings (NAVs) to the funds initially provided, also called the total value to paid-in capital
(TVPI) (Lerner, Leamon, and Hardymon 2012). Based on these measures, investors
can compare funds initiated in the same year, also called vintage year.
276 p e r f o r m a n c e a n d m e a s u r e m e n t

Multiples have both advantages and disadvantages. As a common performance meas-


ure, it is relatively easy to apply and intuitive to understand but can result in potentially
misleading results. Multiples are not annualized, as is common practice with most other
return measures. Also, the method does not consider the time value of money because
cash-on-cash multiples are based solely on absolute numbers. Thus, the stand-alone use
of multiples for performance evaluation of PE funds can be misleading.

I N T E R N A L R AT E O F R E T U R N
The investment process for BO funds differs substantially from those associated with
other asset classes such as closed-end funds. The capital committed to the fund is in-
vested over several years and the funds following distributions are irregular in both size
and frequency. Unlike open-end investment vehicles, PE funds are typically closed-end
funds in which the GP fully controls a fixed pool of capital along with the associated in-
vestment process. Thus, the timing of investments and divestments is a major part of the
value generation of GPs and should be considered when calculating returns.
Because cash-on-cash returns do not reflect the cash-flow timing, they represent
an inappropriate stand-alone return attribution method for evaluating the GPs per-
formance. Therefore, a value-weighted return measure that considers the timing of in-
vestments and divestments by the fund manager is preferred. Although the IRR has
drawbacks, it has emerged as the industry standard.
The IRR as a value-weighted return measure is computed using a funds cash inflows
and cash outflows and corresponds to an overall rate of return to investors considering
the various entry and exit points. As Equation 16.1 shows, the IRR is the interest rate
that forces the net present value (NPV) of all cash flows to be zero:
N
CFt
NPV = , (16.1)
t =0 (1+IRR )t
where N is the lifetime of the fund and CFt is the cash flow accrued over period t. Calcu-
lating the IRR involves several shortcomings and restrictions. GPs must estimate unre-
alized cash flows (NAVs) that are included in calculating the IRR. The NAV calculation
is unreliable during the first few years of a fund because estimates of future cash flows
may vary. Further, management fees, investment costs, and underperforming invest-
ments that are identified early and written down are all taken into account. IRR values
are therefore influenced in an inappropriate and negative way during the early years.
These concerns are why GPs used to value their investments at cost. Introducing fair
value accounting standards has led to some relief because GPs now report their invest-
ments at NAVs with the consequence that valuations have become more aligned with
publicly traded companies (Oberli 2012).
Figure 16.1 shows the top-to-bottom quartile of PE IRR ranges by vintage year from
1984 to 2004. Over that period, the gap between the top-quartile and the bottom-quartile
funds had an average spread of about 2,000 basis points (Boyd 2012). GP selection can
therefore result in returns ranging from exceeding the long-term stock market yield (the
top quartile) to being unable to return capital (the bottom quartile). Being able to iden-
tify the top-quartile PE managers can substantially enhance an investors returns.
F i n d i n g To p Q u a r t i l e M a n a g e r s 277

40
Median

30

20
IRR (%)

10

00
01
84
85
86
87
88
89
90
91
92
93
94
95
96
97

99

02
03
04
98
19
19
19
19
19
19
19
19
19
19
19
19
19
19

20
19

20
20
20
20
19
10

20

Figure 16.1 Top-to-Bottom Quartile Private Equity Internal Rate of


Return Ranges by Vintage Year This figure shows the top-to-bottom quartile of
PEIRR ranges and the corresponding median by vintage year from 1984 to 2004. Source:
Preqin (2014).

P U B L I C M A R K E T E Q U I VA L E N T
The public market equivalent (PME) concept appeared in the mid-1990s (Nickels and
Long, 1996) and is still widely referenced today (Sorensen and Jagannathan 2013).
PMEs offer a simple solution to the benchmarking problem related to time-weighted re-
turns. Nickels and Long propose to assess the opportunity cost of capital for buyouts by
creating a mimicking portfolio of PMEs (i.e., the S&P 500 index for the United States).
These investments are designed to replicate the risk profile of buyouts for timing and
systematic risk. Choosing a benchmark index requires care because only total return
indexes make sense for the analysis. Using the S&P 500 index as a benchmark while
ignoring dividend payments might reduce PME measures by several points and thus
distort the result of the comparison.
The PME approach starts with the following question: Given that an investor in-
vests one dollar in a PE fund in terms of present value, how many dollars would the
investor have to invest in a given public market index to produce a cash-flow equiv-
alent investment to end with the same terminal wealth (Kaserer and Diller 2004)?
The PME answers this question by using the ratio of the terminal wealth obtained
when investing in a PE fund and reinvesting intermediate cash flows in a given public
market benchmark and then comparing it to the terminal wealth obtained when in-
vesting the same amount of money in the benchmark. In this way, a performance
ranking of all available funds becomes possible. Equation 16.2 defines the PME as
follows:
N N
cf t i=t+1(1+ R Ii )
PME = t =1
N . (16.2)
(1+ R It )
t =1
278 p e r f o r m a n c e a n d m e a s u r e m e n t

In their seminal paper on PE returns and persistence, Kaplan and Schoar (2005, p.
1797) implement the PME calculation for performance calculation and underlie the
concept with an example.
A fund with a PME greater than 1 outperformed the S&P 500 (net of all fees). For
example, a private equity fund investing USD 50 million in March 1997 and realizing
USD 100 million in March 2000 would have generated an annualized IRR of 26 per-
cent. However, a limited partner would have been better off investing in the S&P 500
because USD 50 million in the S&P 500 would have grown to USD 103.5 million over
that period. The PME of 0.97 (or 100/103.5) for this investment reflects the fact that
the private equity investment would have underperformed the S&P 500. Alternatively,
a private equity fund investing USD 50 million in March 2000 and realizing USD 50
million in March 2003 would have generated an IRR of 0 percent. However, a limited
partner would have been better off investing in the private equity fund because USD
50 million invested in the S&P 500 would have declined to USD 29.5 million over that
period. The PME of 1.69 (or 50/29.5) for this investment reflects the fact that the pri-
vate equity investment would have outperformed the S&P 500.
Despite predominately being based on cash flows, the PME approach still depends to
some degree on NAVs when the fund is not fully liquidated and the NAV is different from
zero at the end of the time interval. In those cases, the benchmarking process actually re-
duces to the comparison of the end balance of the index-tracking fund to the NAV of the
PE fund, placing more emphasis on the latter number and relying indirectly on the assump-
tion that the PE investor can immediately exit the fund at that value. This feature is why an-
alysts generally use the PME to benchmark the performance of mature funds in which the
non-liquidated NAV represents a small fraction of the total distributions (Rouvinez 2003).
PMEs are simple and circumvent the problems associated with the IRR as a return
measure. As the IRR is calculated using cash flows rather than actually realized returns,
the IRR translates into returns only under extreme assumptions of constant and common
discount rates and reinvestment rates. A constant reinvestment rate is obviously not a re-
alistic assumption, making it problematic for investors to use the IRR as a return measure.
Specifically, as far as an investment in a PE fund is concerned, the LP may be interested
in knowing the terminal wealth of the investment relative to the terminal wealth of a risk-
equivalent public market investment. Obviously, the IRR cannot be used to answer this
question. The PME approach offers a way to compare and rank funds in this respect.
Yet, using the PME has shortcomings such as lack of representation of financial lev-
erage and systematic risk. Generally, studies use a public market benchmark unadjusted
for leverage and systematic risk because the data source lacks extensive deal-level data.
Extensive deal-level data would be needed to adequately adjust returns for financial lev-
erage and systematic risk (i.e., to assess adequate opportunity cost of capital for individ-
ual BO transactions).

Persistence in Private Equity Fund Returns


This section examines empirical evidence on the relationship between predecessor and
successor PE fund returns. Many LPs still pay close attention to the best performing
funds when deciding on PE fund investments. Funds are often classified according to
F i n d i n g To p Q u a r t i l e M a n a g e r s 279

the return quartile in which they are ranked. Thus, finding top-quartile managers and
getting access to their funds become important criteria for investors. But does manager
selection really come down to identifying past winners and is it profitable to recommit
to these funds? Is past performance a sign of future outperformance? A growing body
of academic research based on contingency table tests and cross-sectional regressions
of future fund performance on current performance shows mixed evidence on return
and performance persistence. While studies that include fund returns before 2000 find
statistically significant evidence of persistence, post-2000 returns show mixed results
(Harris et al. 2013).

W H AT I S A TO P - Q UA R T I L E F U N D ?
A top-quartile fund belongs to the 25 percent best performing funds in its peer group.
That is, top-quartile funds are the 25 percent of funds with the highest returns for a spe-
cific year when a fund is launched. Second-quartile funds are represented by the next 25
percent, down to the bottom-quartile funds consisting of the 25 percent of funds with
the lowest returns. If no persistence in returns exists, the expectation is that a succes-
sor fund has about a 25 percent chance of being in any given quartile, regardless of the
predecessor funds quartile. Academic research measures performance mainly based on
cash-on-cash multiples, IRRs, or PMEs.

E V I D E N C E O F Q UA R T I L E P E R S I S T E N C E
Kaplan and Schoar (2005) are the first to identify the persistence of PE and VC fund
returns. Several following studies confirmed their results using various data sources.
Later studies show mixed results. For example, Harris et al. (2013) show that post-2000,
persistence of BO fund performance falls considerably and is associated with poorly
performing funds, which tend to repeat their poor performance relative to other funds.
Empirical evidence on persistence is presented later in this chapter using both data
before and after the 2000s along with hypotheses on why post-2000 persistence falls.

Initial Empirical Evidence On Return Persistence


In a seminal paper on PE returns, persistence, and capital flows, Kaplan and Schoar
(2005) examine 764 PE funds between 1980 and 2001. They show that average PE
fund returns (net of fees) roughly equal the S&P 500 index and higher past returns are
associated with higher future returns. By contrast, Carhart (1997) and Berk and Tonks
(2007) find persistence for the worst performing mutual funds but not for the better
performing mutual funds.
Yet, for PE, top-quartile fund managers of predecessor funds appear in the top quar-
tile again with their successor funds for both BO and VC funds. Kaplan and Schoar
(2005) suggest that persistence results from the heterogeneity in GP skills and lim-
ited scalability of human capital. A puzzle remains, however, concerning why GPs
do not appropriate these returns to superior skills in the form of higher fees. Several
subsequent studies including Glode and Green (2008), Hochberg, Ljungqvist, and
Vissing-Jrgensen (2010), and Marquez, Nanda, and Yavuz (2010) try to rationalize
this phenomenon.
280 p e r f o r m a n c e a n d m e a s u r e m e n t

Why do investors not allocate capital to the point where persistence disappears?
Why do funds not adjust fees to the point where persistence disappears? Kaplan
and Schoar (2005) conclude that this result comes from the fact that the best funds
voluntarily restrict their size. This argument can be related to the Berk and Tonks
(2007) model of mutual funds. Investors learn about a managers skills. Following
good performance, they want to allocate more to the point where expected perfor-
mance of the top fund equals that of other funds. Persistence arises when funds limit
their size.
According to Hochberg et al. (2010), incumbent investors have soft information
about fund manager abilities on which they could hold-up the fund manager. The
idea is that if the investors do not reinvest in the follow-on fund, outside investors
will think that the soft information is negative and, as a result, will reduce their allo-
cation to the newly formed fund. The VC firm, therefore, needs to pay a rent to the
incumbent investors and they do so by limiting their size, hence offering high expected
returns. According to Glode and Green (2008), fund managers can be held-up not
with soft information about their abilities but with information about their investment
strategies. Fund managers need to pay a rent to incumbent investors, thus creating
persistence.
The findings on persistence by Kaplan and Schoar (2005) initiated further empirical
research on the relationship between predecessor and successor PE fund performance.
Table 16.1 gives an overview of the various studies and the main findings related to per-
formance and persistence.
Conner (2005) empirically examines the return persistence for VC firms. VC funds
following a top-quartile predecessor had a 44 percent probability of being top-quartile
again at a 99 percent confidence level. Successor funds to top-quartile predecessors fur-
ther had a 71 percent probability of performing above the median fund from the same
vintage year. Additionally, funds following bottom-quartile-ranked predecessor funds
had a 48 percent probability of being in the bottom quartile again (at a 99 percent con-
fidence level) and a 68 percent probability of being below the median. These results
suggest strong persistence in both good and poor performance in VC funds.
Rouvinez (2006) also shows significant top-quartile overall PE return persistence
and strong survivor bias with serial performers having no advantage over first-time
achievers. Figure 16.2 represents the transition probabilities of the Rouvinez data set.
The top line shows the probability of a fund ending in each quartile, given that the fund
is classified as top-quartile. The second line represents the second quartile and so forth.
Transition probabilities on the diagonal are higher than on the cross diagonal, which
can be interpreted as a sign of persistence. The largest probability is that lower quar-
tiles do not raise a successor fund. About a 40 percent probability exists that managers
with lower quartile funds will not come back to the market. The other cluster of high
transition probabilities is in the upper left corner of the matrix, a sign of top-quartile
persistence.
Rouvinez (2006) further finds that the probability of achieving a top-quartile per-
formance twice in a row does not increase with the managers experience. Both the
probabilities for top-quartile and top-half returns are much higher than the theoreti-
cal 25 and 50 percent for a random process, pointing to some long-term persistence in
returns. Are probabilities higher for managers with a track record of top-quartile funds
Table 16.1 Empirical Evidence on Private Equity Return Persistence

Authors Data Return Persistence


Measure
Kaplan and Thomson Venture IRR Strong persistence across subsequent
Schoar (2005) Economics (TVE), PME funds of a partnership.
19802001
Conner TVE, VC only IRR Significant top-quartile persistence
(2005) 19801995 TVPI (around 44 percent of the top-
quartile persistence and 48 percent
bottom-quartile persistence).
Rouvinez TVE, before 2000 IRR Significant top-quartile persistence
(2006) (around 40 percent of top quartile
persistence) coupled with strong
survivor bias.
Phalippou TVE, VC only IRR Strong persistence of Kaplan and
(2009) 19802003 Schoar (2005) confirmed on ex
post performance measures. No
significant persistence based on ex
post measures.
Paresys (2010) Preqin, before IRR Strong correlation between the
2010 performance of a fund and the
success of its predecessor. About 43
percent of GPs that manage a top-
quartile fund also have their next
fund ranked in the top quartile, and
a total of 68 percent beat the median
benchmark.
Chung (2012) Preqin, before IRR Performance persistence in PE
2005 industry is correlated with the
performance of the first follow-on
fund, but not with that of subsequent
funds. Common market conditions
largely explain persistence. Fund
growth based on past performance
erodes performance and persistence.
Harris et al. Burgiss, IRR Persistence confirmed pre-2000.
(2013) 19842008 PME Mixed evidence exists for Persistent
TVPI BO funds afterwards. Performance
persistence remains statistically and
economically persistent as in pre-2000.

Note: This table provides an overview on the academic research on the persistence of PE returns.
Although earlier studies find strong persistence across subsequent funds of a partnership, later studies
show mixed evidence.
282 p e r f o r m a n c e a n d m e a s u r e m e n t

Successor in Successor Successor in Successor in No successor


first in second third fourth fund
quartile quartile quartile quartile raised
First 33% 22% 15% 14% 16%
quartile
Second 20% 21% 15% 16% 27%
quartile
Third 12% 17% 17% 15% 39%
quartile
Fourth 11% 9% 16% 22% 43%
quartile

Figure 16.2 Transition Matrix: Probability of Transition from One


Quarter to Another This chart shows the transition matrix, which is the probability
of transition from one quartile to another when considering two successive funds by the
same manager. The most probable event for a top-quartile fund is to achieve top-quartile
again. The most probable event for lower quartile funds (Q3, Q4) is that no successor fund
gets raised. Source: Rouvinez (2006).

only? Given that all predecessor funds are ranked top-quartile, the probability increases
for the third fund to 51 percent but reverts to 40 percent for the fifth fund, showing no
clear advantage.
Further empirical studies find signs of return persistence. For example, Phalippou
(2009) confirms the results of Kaplan and Schoar (2005) of significant and robust per-
sistence for VC funds. Using a Preqin data set, Paresys (2010) also finds a strong correla-
tion between a funds performance and the success of its predecessor. About 39 percent
of managers with a top-quartile fund have their consecutive funds also ranked in the
top quartile, and nearly 70 percent of these managers beat the median benchmark with
their next fund. Persistence is also detected for the worst performing managers. About
38 percent of fund managers with bottom-quartile-ranked funds saw their next fund
ranked in the same quartile, while just 15 percent of managers with bottom-quartile
funds could later produce top-ranking funds.

Return Persistence Post-2000


More recent academic research that includes PE fund performance data post-2000
shows mixed evidence on persistence. Chung (2012) uses a data set of PE funds raised
before 2005 and finds that performance persistence in the PE industry is not long-lived.
Current fund performance is positively and significantly associated with the perfor-
mance of the first follow-on fund, but the persistence declines substantially afterwards.
Relatively underperforming funds (i.e., funds in the bottom or medium performance
tercile portfolios) largely keep underperforming. A one percentage point increase in the
IRR for a BO fund in the medium tercile portfolio leads to a 0.95 percent increase in
the IRR of its follow-on fund, whereas the change is 0.56 percent for a fund in the upper
tercile portfolio. In a regression analysis based on multiples, the difference is similar.
F i n d i n g To p Q u a r t i l e M a n a g e r s 283

A one percentage point increase in the multiple for a fund in the medium tercile port-
folio is associated with a 0.69 percent increase in the following fund multiple, while it is
only 0.11 percent for funds in the upper tercile portfolio, further underlining that largely
underperforming funds drive persistence.
Chung (2012) explains the short-lived performance persistence with the common-
ality of market conditions between two successive funds. Given that a PE funds life is
about 10 years, and a follow-on fund is usually raised three to five years after a preceding
funds raising, successive funds have an overlapping investment period of several years,
during which common economic conditions or shocks can simultaneously influence
the performance of preceding and following funds. Therefore, the similarity of market
conditions between the current and follow-on funds can affect persistence. The extent
to which common market conditions explain short-run persistence counters the view
that PE partnerships have proprietary skills.
Chung (2012) also shows that better performing funds raise larger follow-on funds
than their worse performing counterparts, but at the same time finds that funds with
greater growth subsequently underperform. The return-chasing capital cycle is more
pronounced for BO funds, and the diminishing returns to capital inflows are found
mainly among VC funds. The asymmetry between BO and VC funds on the effect of
capital flows on performance persistence is consistent with the view the VC industry is
labor-intensive, while the BO industry is capital-intensive. In other words, in managing
portfolio companies of a VC fund, fund managers provide not only capital but to a large
degree provide additional resources such as industry networks and management skills.
An increase in fund size, which increases either the target size or the number of invest-
ments, requires a greater amount of management care.
Harris et al. (2013) use a new data set from Burgiss, which is sourced from more
than 200 institutional investors to confirm significant persistence in performance, using
various measures, for pre-2000 funds, particularly for VC funds. Post-2000, persistence
of BO fund performance drops considerably and is associated with poorly performing
funds, which repeat their poor performance relative to other funds. After sorting funds
by quartile performance of their previous funds, performance of the current fund is
statistically indistinguishable regardless of quartile. However, the returns to BO funds
in all previous performance quartiles, including the bottom, exceed those of public mar-
kets as measured by the S&P 500 index.

W H Y R E T U R N S D O O R D O N OT P E R S I S T
Kaplan and Schoar (2005) view the heterogeneity in GP skills and limited scalability of
human capital as hypotheses for the persistence of PE fund returns. Indeed, one expla-
nation is that good performance puts talented GPs in touch with talented entrepreneurs
who create good outcomes. A virtuous cycle begins in which success begets success.
A successful GP could possibly choose more deals from better entrepreneurs. Further,
managers of top-tier funds can leverage their perceived expertise into investing on more
favorable terms, generating better returns for LPs.
Lerner, Schoar, and Wongsunwai (2007) examine LPs and show that the returns
realized by institutional investors from PE differ dramatically across institutions.
Endowments annual returns are nearly 21 percent greater than average. Analysis of
284 p e r f o r m a n c e a n d m e a s u r e m e n t

reinvestment decisions suggests that endowments and to a lesser extent public pensions
are better than other investors at predicting whether follow-on funds will have high re-
turns. The results are not primarily due to endowments greater access to established
funds because they also hold young or undersubscribed funds. These results suggest
that investors vary in their sophistication and potentially their investment objectives.
One caveat is that Lerner et al. (2007) conducted the study before the financial crisis of
20072008 initiated by the collapse of Lehman Brothers on September 15, 2008. This
crisis hit endowments particularly hard. Another caveat is that investment in VC funds
could drive the endowments superior selection skills.
More recent academic research that includes PE fund performance data after the
year 2000 shows mixed evidence on persistence. On the dynamic of top-quartile funds
attracting more investors and growing disproportionately, Lopez-de Silanes, Philippou,
and Gottschalg (2013) discover diseconomies of scale in managing PE funds. Since
performance is not scalable, this is a potential explanation for the mixed evidence on
the persistence of BO funds (Chung 2012; Harris et al. 2013). Investments held by PE
firms in periods with a high number of simultaneous investments underperform sub-
stantially. Harris et al. (2013) believe that one reason for underperformance is that firms
have difficulty keeping talent. Once they reach a certain point in their career, success-
ful GPs often leave to set up a partnership on their own. These spin-out groups then
go on to compete with their progenitors for deals, which can result in a less efficient,
more mature market for deals and higher prices. Partners at successful firms might also
become less eager to take the risks to get an outstanding return (i.e., they become more
risk averse), which can lead to more modest returns. Further, a refinement of areas of ex-
pertise and personnel additions/changes coupled with changing market dynamics can
alter the strategy pursued by the GP and the environment in which the GP will invest
the funds. GPs also tend to learn from each other, leading to a reduction in outstanding
individual fund returns and persistence (Harris et al. 2013).
Nonetheless, Chung (2012) and Harris et al. (2013) show continuing persistence of
VC returns and performance, supporting the industry rule of thumb to invest with GPs
that have previously performed well and to avoid those who have not. The stronger per-
formance persistence for VC as compared to BO suggests that GP skills and networks
for successful VC investing are harder to replicate than is the case for buyouts.

Finding Top-Quartile Managers


The conventional rationale that many PE investors still employ during GP selection is
to focus on prior fund returns as a means of predicting future top-quartile performance.
However, determining prior fund returns has its own limitations.

L I M I TAT I O N S I N I M P L E M E N T I N G A TO P - Q UA R T I L E A P P R O A C H
Implementing an investment approach based on the assumption of return persistence
for selected GPs has several practical limitations. First, due to ambiguous performance
criteria, some leeway exists in defining top-quartile performance. Second, PE groups
often raise subsequent funds before the previous funds performance can be accurately
F i n d i n g To p Q u a r t i l e M a n a g e r s 285

measured. Using a holistic due diligence process might offer the best insights into fac-
tors that lead to GP success and failure and give investors the best opportunity to find
top-quartile PE managers.

Top-Quartile Classification
As previously discussed, a top-quartile fund belongs to the 25 percent best funds in its
peer group. Yet, many more funds in the market claim top-quartile performance. One
reason is that except for the top-quartile classification itself, the definition is subject
to interpretation. For example, best performance is undefined and could refer to cash-
on-cash multiples, IRR, PME, or another performance measure. Other questionable
issues involve which funds to consider and what constitutes peers and the geographic
region. Therefore, investors need to be cautious when GPs claim top-quartile perfor-
mance because the potential for manipulation is high.
Wang and Conner (2004) show that based on two primary measures to determine
top-quartile performance (the IRR and cost multiples) only 84 percent of funds classi-
fied as top-quartile could claim to be top-quartile on both metrics. Thus, more than 25
percent of all funds claim to be top-quartile: those being top-quartile on both measures,
but also the ones based on the IRR or multiple only.
Furthermore, a fund may move across quartiles over its life. As Boyd (2012) points
out, a fund claiming top-quartile performance may be referring to a point in time
rather than to a persistent rating. Her results show that 64 percent of funds achieved
top-quartile performance at some point during their lives.
Hendershott (2008) examines the likelihood that GPs with top-quartile track re-
cords also have the corresponding top-quartile ability. Using a Bayesian approach, he
infers the ability to outperform from performance persistence. Varying manager ability
thereby increases return persistence over time because chance diminishes performance
persistence. The performance persistence level in any given asset class reveals the role
managerial ability plays in determining investment returns. Hendershott finds that top-
quartile performance in three or four funds is required in order to be in the top-ability-
quartile at an 80 percent confidence level.

Ex-ante Performance Evaluation


Performance persistence and a consistently large spread between better and worse per-
forming funds (Artus et al. 2004) lead many LPs to focus on investing in the follow-on
funds of past top-quartile performers (Hendershott 2008). Challenges are inherent in
implementing this strategy. Substantial time, typically years, is necessary before the re-
sults of PE investments become known. Meanwhile, fund managers carry their portfolio
at cost unless a subsequent financing event occurs. Thus, PE managers often raise their
next fund before the performance of the previous fund can be accurately measured.
As Phalippou (2009) reveals, a PE firm raises a follow-on fund after a median of
three years; 15 percent of the funds raise a succeeding fund after only one year, and 22
percent after two years. This finding means that for investors to act on this relationship,
they would need to know what the final performance of a fund is after only a few years
of operation, which is difficult.
Freidman (2011) shows that 50 percent of top-quartile performing BO funds ob-
served in the fourth year of investment life actually finish in the top quartile. For VC,
286 p e r f o r m a n c e a n d m e a s u r e m e n t

the value is higher with 60 percent ending in the top quartile. Of these top-quartile
funds, 43 percent are able to repeat their performance with their successor fund. Using
Bayes theorem, Freidman shows that 31 percent of the BO funds that are classified as
top-quartile (33 percent for VC) based on the observations during their fourth year
of investment life are likely to replicate top-quartile performance in their successor
fund. These results show the limits of manager assessment based solely on prior perfor-
mance, given that the ex-ante chance of selecting a top-quartile manager at random is
25 percent.
In cases in which the fundraising period for the successor fund occurs two or three
years after the top-quartile performing funds final close, the probability of maintaining
top-quartile performance decreases further. Therefore, the decision to reinvest based
on performance would need to be based on funds of an even earlier generation that
are more mature. However, Chung (2012) finds that the second or prior funds perfor-
mance can be a misleading signal of future performance.
Given that top-quartile performance cannot reliably be determined before investors
have to make their next commitment, Rouvinez (2006) concludes that due diligence
including proper understanding of the drivers of past returns is far more important than
the quartile itself. Understanding how and why a manager achieved superior perfor-
mance and whether the conditions for success are likely to be met again in the future is
crucial for the investment decision. Thus, thorough due diligence plays a central role in
LPs recommitment decisions.

THE ROLE OF DUE DILIGENCE


A successful due diligence review entails trying to understand the factors that influence
the success and failure of portfolio companies (Boyd 2012). Perhaps the best approach
is to talk to GPs, co-investors, and the chief executive officers (CEOs) of the portfolio
companies. Sensitivity analysis is also a critical tool for providing insight as to whether
performance is skewed up or down based on particular investments. The analysis en-
ables investors to understand how dependent the portfolios performance is on one
exceptional success or failure. Due diligence surveys of the investment strategy and phi-
losophy of the team add further depth beyond what is found in the private placement
memorandum (PPM) and other marketing materials. Direct meetings with GPs and
back-office personnel may reveal areas of expertise, as well as limits within the team and
the overall dynamics within the firm. Finally, prospective investors should make refer-
ence checks with current and past LPs, prior members of the firm, CEOs of portfolio
companies, lenders, and consultants to provide further verification of the firms ability
and identify any areas of concern. This in-depth screening process, which focuses heav-
ily on due diligence rather than past performance, is far more apt to capture both the
success of a particular strategy and potential warnings or red flags for investors.
That approach has implications on including new and emerging managers within
a PE portfolio. The most common rationale provided by investors for not committing
to these groups is their lack of a track record. Given the prior discussion of past perfor-
mance as a limited indicator of future success, new and emerging managers should not
be excluded solely on this basis. Through careful due diligence, selecting high-quality
new and emerging managers may prove effective in achieving top-quartile performance,
F i n d i n g To p Q u a r t i l e M a n a g e r s 287

while providing further portfolio diversification. Further, backing successful new man-
agers at an early stage may result in access to later oversubscribed fund generations
(Boyd 2012).
In sum, investors need to apply a holistic due diligence process to successfully eval-
uate PE investments. LPs that ignore this method and rely solely on investment per-
formance for selecting their GPs are prone to overlook the new top-quartile managers.

Summary and Conclusions


Empirical evidence on PE return persistence shows mixed results. Early academic
studies by Conner (2005), Kaplan and Schoar (2005), Rouvinez (2006), Phalippou
(2009), and Paresys (2010) show statistically significant return persistence for both BO
and VC funds. Later studies using data after the year 2000 such as Harris et al. (2013)
report mixed evidence on return persistence for BO funds.
Reasons for less persistence include that funds with strong performance attract in-
vestors and grow in size accordingly. Performance, however, is often not scalable, which
is a sign of diseconomies of scale in managing PE funds. Strong performing firms also
have difficulties in retaining talent. Successful managers often decide to set up their own
PE partnerships once they have reached a certain point in their careers. The newly cre-
ated partnerships then compete with their former firms for attractive deals, which can
result in a less efficient, more mature market for deals and higher prices. Partners at suc-
cessful firms might also become less eager to take the risks needed to get an outstand-
ing return. A refinement of areas of expertise and personnel additions/changes coupled
with changing market dynamics can further alter the strategy pursued by GPs and the
environment in which they will invest funds.
Even if persistence exists to some degree, implementing a strategy to invest in fund
managers who have an extended track record of superior returns is not easily imple-
mented. First, due to ambiguous performance criteria, some leeway exists in defining
top-quartile performance, which is also why much more than one-quarter of funds
claim to be top-quartile. Further, a fund may move between quartiles over its life. A
fund claiming top-quartile performance may be referring to a point in time rather than
to a persistent rating. PE groups often raise subsequent funds before the previous funds
performance can be accurately measured. Thus, in order for investors to act on any top-
quartile relationship between predecessor and follow-on funds, they would need to
know the funds final performance after only a few years of operation when the fund is
still in the investment phase. Today, observing that good vintage 2004 funds tended to
be followed by good vintage 2006 funds is possible. But exploiting this information for
investment purposes would require knowing in 2006 which of the vintage 2004 funds
were good. Interim performance, however, is an uncertain indication of future perfor-
mance of a fund. This compounds the challenge of identifying top PE managers (Hen-
dershott 2008).
Although empirical evidence shows that past performance can help to identify
top-quartile PE managers to some degree, it still cannot identify what makes the top-
ranking managers outperform. A thorough due diligence therefore takes a central role.
Perhaps the best approach is to talk to GPs, co-investors, and the CEOs of the portfolio
288 p e r f o r m a n c e a n d m e a s u r e m e n t

companies. Due diligence surveys of the investment strategy and philosophy of the
team add further depth beyond that provided in the PPM and other marketing materi-
als. Direct meetings with other GPs and back-office personnel can help to identify areas
of expertise, limits within the team, and the overall dynamics within the firm. Potential
investors should also make reference checks with current and past LPs, prior firm mem-
bers, CEOs of portfolio companies, lenders, and consultants to confirm the firms ability
or to identify additional areas of concern. Using a holistic due diligence process appears
to offer the best insights into the factors that lead to the success and failure of managers
and gives investors the most reliable indications toward finding future top-quartile PE
managers.

Discussion Questions
1. Discuss the factors to consider when evaluating a fund based on its cash-on-cash
returns (multiples) or the IRR.
2. Explain what the PME approach toward benchmarking PE returns seeks to
accomplish.
3. Explain the relationship between a PE firms fundraising and the valuation levels it
reports for its PE investments. How does that behavior influence the search for top-
quartile managers?
4. Identify the practical limitations of implementing an investment approach based on
top quartiles.
5. Explain why the worst performing BO funds over the long term are often the largest
funds by investment size.

References
Artus, Patrick, Jrme Teletche, Christoph Kaserer, Christian Diller, Olivier Dupont, and Didier
Guennoc. 2004. Performance Measurement and Asset Allocation for European Private
Equity Funds. European Venture Capital Association. Available at http://www.evca.eu/
uploadedFiles/Home/Knowledge_Center/External_Research/Academics/full_study.pdf
Berk, Jonathan B., and Ian Tonks. 2007. Return Persistence and Fund Flows in the Worst Perform-
ing Mutual Funds. NBER Working Paper Series 13042.
Boyd, Theresa. 2012. Predicting Private Equity Performance Based on Historical Returns. Invesco
Investment Insights. Available at https://www.invesco.com/portal/site/us/investors/insights.
Carhart, Mark M. 1997. On Persistence in Mutual Fund Performance. Journal of Finance 52:1,
5782.
Chung, Ji-Woong. 2012. Performance Persistence in Private Equity Funds. Available at http://
ssrn.com/abstract=1686112.
Conner, Andrew. 2005. Persistence in Venture Capital Returns. Private Equity International,
March, 6567.
Freidman, Tim. 2011. Preqin Private Equity Performance Report Early Warning System for LPs:
Using Year Four/Six Performance Metrics to Predict Final Outcomes. Available at https://
www.preqin.com/docs/reports/Early_Warning_System_for_LPs.pdf
Glode, Vincent, and Richard C. Green. 2008. Information Spillovers and Performance Persistence
in Private Equity Partnerships. Working Paper, Tepper School of Business, Carnegie Mellon
University.
F i n d i n g To p Q u a r t i l e M a n a g e r s 289

Harris, Robert S., Tim Jenkinson, Steven N. Kaplan, and Rdiger Stucke. 2013. Has Persistence
Persisted in Private Equity? Evidence from Buyout and Venture Capital Funds. Available at
http://ssrn.com/abstract=2304808.
Hendershott, Robert. 2008. Using Past Performance to Infer Investment Manager Ability. Work-
ing Paper, Leavey School of Business, Santa Clara University.
Hochberg, Yael V., Alexander Ljungqvist, and Annette Vissing-Jrgensen. 2010. Informational
Hold-up and Performance Persistence in Venture Capital. USC FBE Finance Seminar.
Kaplan, Steven N., and Antoinette Schoar. 2005. Private Equity Performance: Returns, Persistence,
and Capital Flows. Journal of Finance 60:4, 17911823.
Kaserer, Christoph, and Christian Diller. 2004. European Private Equity Funds: A Cash Flow
Based Performance Analysis. Center for Entrepreneurial and Financial Studies (CEFS) and
Department for Financial Management and Capital Markets Technische Universitt Mnchen,
147.
Lerner, Josh, Ann Leamon, and Felda Hardymon. 2012. Venture Capital, Private Equity, and the Fi-
nancing of Entrepreneurship. New York: John Wiley & Sons.
Lerner, Josh, Antoinette Schoar, and Wan Wongsunwai. 2007. Smart Institutions, Foolish Choices?
The Limited Partner Performance Puzzle. Journal of Finance 62:2, 731764.
Lopez-de Silanes, Florencio, Ludovic Phalippou, and Oliver Gottschalg. 2013. Giants at the Gate:
Investment Returns and Diseconomies of Scale in Private Equity. Available at http://ssrn.
com/abstract=1363883.
Marquez, Robert, Vikram Nanda, and M. Deniz Yavuz. 2010. Private Equity Fund Returns: Do
Managers Actually Leave Money on the Table? Working Paper, Washington University,
St.Louis.
Nickels, Craig J., and Austin M. Long. 1996. A Private Investment Benchmark. AIMR Conference
on Venture Capital Investing, San Francisco.
Oberli, Adrian. 2012. The Implications of Fair Value Accounting Standards on Private Equity
Buyout Returns. Journal of Private Equity 15:4, 5578.
Paresys, Etienne. 2010. Relationship between Predecessor and Successor Fund Quartile. Preqin
Research Center. Available at https://www.preqin.com/blog/101/3593/predecessor-
successor-funds.
Phalippou, Ludovic. 2009. Venture Capital Funds: Performance Persistence and Flow-Performance
Relation. Journal of Banking and Finance 34:3, 568577.
Phalippou, Ludovic, and Oliver Gottschalg. 2009. The Performance of Private Equity Funds.
Review of Financial Studies 22:4, 17471776.
Rouvinez, Christophe. 2003. Private Equity Benchmarking with PME+. Venture Capital Journal
43:8, 3438.
Rouvinez, Christophe. 2006. Top Quartile Persistence in Private Equity. Private Equity Interna-
tional, June, 7679.
Sorensen, Morten, and Ravi Jagannathan. 2013. The Public Market Equivalent and Private Equity
Performance. Netspar Discussion Paper No. 09/20132039. Available at http://ssrn.com/
abstract=2347972 or http://dx.doi.org/10.2139/ssrn.2347972.
Wang, Cheng, and Andrew Conner. 2004. Whats in a Quartile? Alignment Capital Group Re-
search. Available at http://www.alignmentcapital.com/pdfs/research/acg_top_quartile_
2004.pdf
17
Private Equity Due Diligence
MANU SHARMA
Professor of Finance, Panjab University

ESHA PRASHAR
Assistant Professor, Chitkara University

Introduction
Private equity (PE) is the equity capital of a company that is not quoted on any stock ex-
change. A PE firm, an angel investor, or a venture capital (VC) firm usually invest in PE.
Although each category of investors has its own investment strategies reflecting its set of
goals and preferences, each provides capital to the target company for its expansion or
for the targets restructuring of its ownership, operations, or management.
Due diligence refers to the responsibility, diligence, and prudence that a rational in-
vestor would expect and exercise in the process of investigating a potential investment.
It helps establish all the material facts about a company before entering into a contract,
thus preventing unnecessary harm to either party involved in the transaction. Due dil-
igence means ensuring that the party gets exactly what it expects out of a transaction
for which it is making a payment. It includes having a good understanding of liabilities,
any pending lawsuits, long-term customer agreements, leases, employment contracts,
or any such contracts that the company has already entered and is liable for paying in
the future. According to Scharfman (2012), due diligence is more an art than a science
and a detailed analysis allows investors to identify funds that may underperform or fail
in the event of unexpected pressure.
According to Loza (2006), due diligence is an essential process while developing
strategies for a company and structuring intellectual property and business transactions.
Gottschalg and Kreuter (2006) discuss the different selection criteria of PE fund man-
agers and their efficiency in the performance of PE portfolios. This process is particu-
larly relevant for due diligence when identifying potential PE firms for investment by
limited partners (LPs).
Ramsay and Sidhu (1995) discuss the importance of due diligence when valuing
stocks. They note that a proper due diligence can provide investors with more credible
information about the company before they invest. This action, in turn, might reduce
the underpricing of the shares.

290
Priv at e E qu it y Du e Dil ig e n ce 291

The purpose of this chapter is to examine the due diligence process for PE firms. Un-
derstanding both; the process of due diligence and the techniques used to carry out due
diligence is important. It provides basis for a future decision-making about whether an
investment should be made and at what price.
The remainder of the chapter has the following structure. The next section highlights
the meaning and importance of due diligence in PE and discusses the due diligence that
LPs conduct on the general partners (GPs), that GPs conduct on investments, and the
different parties involved in the buyout process. The following section examines the
structural framework of the due diligence process and the steps undertaken followed by a
detailed discussion of the different types of due diligence. This is followed by an examina-
tion of both the challenges of the due diligence process and those faced by a PE manager
when conducting due diligence. The final section provides a summary and conclusion.

The Meaning and Importance of Due Diligence


inPrivate Equity
Due diligence in PE has become increasingly important given the constantly changing
business environment. According to Ghai, Kehoe, and Pinkus (2014), PE has grown
from the equivalent of 1.5 percent of the global stock market capitalization in 2000 to
around 3.9 percent in 2012. Hence, the pace and the magnitude at which the PE market
is growing increases the importance of conducting a thorough due diligence before en-
tering into an investment contract.
PE has two types of partners: LPs and GPs. Most PE funds are structured as limited part-
nerships and are governed by the terms and conditions set forth in the limited partnership
agreement. GPs raise investment capital from the LPs, including high net worth individuals
and institutional investors such as pension funds, insurance companies, and foundations.
A well-structured due diligence process is an initial step of any investment decision.
From an LPs perspective, due diligence is important to get a complete analysis of GPs
under consideration for managing the investment as well as identifying the most prof-
itable fund available on the market. The main purpose of conducting due diligence
from the perspective of a GP is to identify the risk and return profile of the target com-
pany. Thus, in any PE contract, due diligence is carried out by LPs on GPs and by GPs
when investing capital in target companies. Each type of due diligence is discussed
subsequently.

D U E D I L I G E N C E B Y L I M I T E D PA R T N E R S O N G E N E R A L PA R T N E R S
In a PE fund, LPs are only passive investors who provide capital to GPs to help them
realize their investment strategy and derive value and returns from their investments.
LPs typically provide most of the money that the GPs get for investment. Hence, LPs
perform due diligence on the GPs before putting their own or clients money at stake
because the LPs need to ensure that the interests of GPs are aligned with their own. As
Elton, Halloran, MacArthur, and Varma (2011) note, the most important aspect that
LPs consider in GPs is their past behavior, which can serve as an indicator of what they
can expect in the future.
292 p e r f o r m a n c e a n d m e a s u r e m e n t

LPs should consider the following when conducting due diligence before selecting GPs.

Finding the management team. LPs should determine if the management team has
worked together in the past to determine its experience, sourcing opportunities, abil-
ity to generate returns, and networks. An LPs main focus lies with identifying how
many deals the GPs have done together. Although a PE firm may have strong histor-
ical returns, changes in team composition could alter its willingness to take risks
(Axial 2011).
Sourcing the deal. An important factor in evaluating a PE team is the predictability
and quality of deal flow. LPs need to determine if the GPs are sufficiently capable of
competing against other buyers in a competitive environment without overpaying.
Past failures. Before selecting the GPs, the LPs should conduct rigorous due dili-
gence and interviews of the GPs to inquire about any past failures and to understand
how the GPs analyzed their past mistakes and failures to avoid making similar mis-
takes moving forward. Learning the perspectives of different members of the same
team can help the LPs determine the strength of the team and ensure that the GPs
are not blaming each other, or any member of the team who has left, for previous
failures.
Others committed to the fund. Identifying the presence of other investors who are pro-
viding capital to the GPs is important because they validate the investment in a GP.
Operational skills. LPs should consider the value creating partnerships that the GPs
have formed with the management teams of their portfolio companies.
Environmental, social, and governance (ESG) issues. LPs should include questions on
ESG integration and/or portfolio company risk management in their formal due dil-
igence process. They should also determine if the GPs have an ESG policy, the con-
tents of the policy, and how it is implemented. Understanding how the GPs have
dealt with ESG-related opportunities and risks is important because their behavior
should align with the ESG policies of the LPs. The ESG policy also helps to influence
the LPs decision-making and ownership process. If the ESG issues are not addressed
before signing the deal agreement, doing so at a later stage may be difficult.

D U E D I L I G E N C E B Y G E N E R A L PA R T N E R S
Once the LPs secure the GPs, the major responsibility of the GPs is to conduct a thor-
ough due diligence before making an investment. Due diligence helps PE firms gain
better knowledge and understanding of the companies in which they are considering
investing. As the fund managers, GPs are fully accountable for sourcing, analyzing, and
monitoring the funds investments.

PA R T I E S I N V O LV E D I N T H E B U Y O U T P R O C E S S
Ashurst (2013) highlights the role of the following parties involved in a buyout process.

Management team. The management team consists of a few managers who make stra-
tegic decisions until the buyout is completed. Second tier managers can support the
core management team depending on the size of the deal.
Priv at e E qu it y Du e Dil ig e n ce 293

PE provider. Transactions usually allow one or more executives to represent the PE


equity provider who plays a central role during the deal negotiations.
Managements lawyers. Managements lawyers draft the contractual terms and advise
management on issues related to the investment. In a smaller management buyout
(MBO), their role may extend to conducting negotiations on the companys behalf.
In a larger buyout, these lawyers ensure that the contract for acquiring the target does
not become a tedious process for management.
PE providers lawyers. Lawyers for the PE provider are committed to verifying all the
details and aspects of the transaction. They prepare and negotiate the related docu-
ments for the equity part of the transaction and establish meetings with the manage-
ments lawyers to discuss the contract. Before the contract is signed, the PE provid-
ers lawyers make sure that the due diligence process is properly carried out.
Banks. One or more banks may fund the company in the form of senior debt, which is
debt that is generally ranked ahead of all other debts (e.g., a second lien or mezza-
nine) in terms of insolvency.
Bank lawyers. Bank lawyers review the due diligence report, terms of the acquisition
agreement, and the sellers disclosures against all warranties. If they find everything
appropriate, then the bank documents are drafted.
Accountants. The PE provider usually appoints an accountant to produce a detailed
report on the target company and to review the business plan and managements fi-
nancial projections. If the buyout process becomes a competitive auction, the seller
generally appoints a team of accountants to get a detailed vendor due diligence
(VDD) report on the target company. This report is then discussed with the success-
ful bidder and the concerned banks.
Investment bankers. At the beginning of the PE transaction, the parties include the
management team and their lawyers, the PE providers, including the PE firm, a ven-
ture capital, an angel investor or any other firm investing funds and their lawyers,
managements accountants, and the bankers who advise on tax and structuring
issues. In a large buyout, the PE provider typically appoints an investment banker to
manage the transaction. The investment banker also provides strategic advice on the
deal negotiations both to the seller for the sale proceeds and to the banks on financ-
ing the deal.
Environmental auditors. PE providers ask environmental auditors to carry out an en-
vironmental risk assessment of the target. The environmental risk assessment en-
sures that the target company complies with relevant environmental laws. The sell-
ers environmental auditor can also produce an environmental risk assessment to the
successful bidder and the banks that finance the deal.

FINANCING A BUYOUT
A PE firm acquires a target using both equity and debt. Regardless of the method
used, conducting due diligence is important due to the large investment required. No
prudent buyer should undertake such an investment without first investigating the
target company. All companies funding the transaction should conduct a thorough
due diligence about the transaction due to the risks involved with the large sum of
investments.
294 p e r f o r m a n c e a n d m e a s u r e m e n t

A deal can be financed in various ways but it is generally done through equity and
debt. Usually, debt financing is the largest part of financing a buyout. Several sources
such as banks, insurance companies, pension funds, government agencies, and individ-
uals provide the equity involved in financing a PE transaction and main source of debt
is senior-secured debt financed by a bank in the form of a secured loan. Using either
source of financing, conducting proper due diligence is important due to the risks in-
volved in financing the deal.

Steps and Structural Framework for Conducting Due


Diligence
STEPS IN DUE DILIGENCE
The process of due diligence is conducted so as to reduce the uncertainties and confirm
the assumptions that define and help prioritize the basic issues that companies would
face when entering into a PE contract. The process of due diligence should include an
understanding of the organization, its operations, financials, logistics, technology, cor-
porate strategy, and other aspects related to the target. Complex issues should be con-
verted into concise resolution oriented terminology that is easy to understand for the
various stakeholders involved in the process.
The due diligence process involves four phases.

Planning phase. In the planning phase, the due diligence team tries to understand the
projects technicalities and scope. The objectives are clarified and the availability of
resources is considered. This phase also considers the target companys financials,
competition, management team and organizational culture, infrastructure, and liabil-
ity structure as well as the sustainability of the business. The GP should ensure that
the team members have the background, skill, and interest needed to make a project
succeed. The process of due diligence requires that multiple partners integrate and
communicate. Thus, the team needs an understanding of its common goals.
Data collection phase. After understanding the primary requirements, the next stage
involves collecting the business data including the key products and factors that are
relevant to the company. Viable information sources are the Internet, competitors,
databases, vendors, customers, industry associations, stakeholders, or those associ-
ated with the company. Both primary and secondary searches can be conducted de-
pending on the information that is available and the type of information that is re-
quired.
Data analysis phase. This stage involves analyzing the data that has been collected and
drawing conclusions based on critical factors such as individual strategy of each busi-
ness function of the company, potential market growth, regulatory constraints, infra-
structure requirements, and technical complexities (e.g., a long lead time needed to
implement a particular project due to its technical complexity).
Report finalization. After the due diligence team collects data and visits the site, it
prepares the final presentations and reports, which become essential elements in the
decision-making and negotiation process.
Priv at e E qu it y Du e Dil ig e n ce 295

STRUCTURAL FRAMEWORK OF A DUE DILIGENCE PROCESS


According to Ender (2010), the first step of a due diligence framework consists of a top-
down review that assesses the attractiveness of the different types of PE sub-segments
and regions. Evaluation criteria should include factors such as the investment opportu-
nities in a particular segment, demand and supply for the set capital, type and quality of
the fund manager pool, entry and exit prices, and future developments. Understanding
the investment strategy of the fund manager is also important in determining the attrac-
tiveness of the PE investment to that of the investors total portfolio.
The basis of any successful due diligence process is a structured process, a proven
evaluation framework, and an experienced team. Effective PE due diligence also re-
quires (1) in-depth knowledge of the past fund investments, the performance of these
funds, business and investment performance, and the fund managers value creation
record; (2) strength and unity among the team that is working on an investment proj-
ect; and (3) an experienced senior professionals consultation on the investments risk.
Through due diligence, a firm can gain insights into the credibility of the targets
owners and managers and eliminate less profitable ones. After the deal is completed,
identifying the appropriate strategic direction helps to increase performance and build
firm value. Perhaps the best and easiest way to identify a target is by:

Identifying the right sector. Finding the right sector involves weighing its attractiveness
in the market as well as identifying the companys size, growth rate, and ease of entry
and exit.
Finding the right sub-sector. This process involves concentrating on the most promis-
ing segment and the geography that the sector has to offer.
Identifying a potential target. Attention should focus on identifying factors such as the
companys relative market share, new industry trends, earnings volatility, and emerg-
ing new profits. After identifying a target firm, the final step is to conduct the process
of due diligence for the target company. Due diligence provides the necessary inputs
to decide whether or not to move forward with the deal.

CONDUCTING THOROUGH DUE DILIGENCE


According to Scharfman (2012), a particular firm may look attractive, but ignoring
due diligence red flags can result in investment problems that far outweigh any possi-
ble future benefits. Hence, conducting a thorough due diligence process can minimize
the risks and maximize the opportunities for investment by thoroughly analyzing the
targets past, present, and the projected performance. Thorough due diligence involves
understanding every aspect related to the target that can provide useful information
such as the targets financial statements, strategies, information technology, human re-
sources, and operations.

Types of Due Diligence


Conducting PE due diligence can be difficult because the relevant data and mate-
rial for investor decision-making is often not readily available. Hence, with whatever
296 p e r f o r m a n c e a n d m e a s u r e m e n t

Operational
Special Due Commercial
Due
Diligence Due Diligence
Diligence

Tax Due
Diligence Legal Due
Types of Due Diligence Diligence

Financial
IT Due Due
Diligence Diligence

Strategic
Cultural Due Human Due
Due
Diligence Diligence
Diligence

Figure 17.1 Types of Due Diligence This figure shows the different types of due
diligence that every company should conduct for a target company before finalizing a
deal.

information that is openly available and the best possible information that can be ex-
tracted from secondary sources, a prudent investor has to make wise decisions about
investments. The following are some important types of due diligence that should be
considered before finalizing the contract. Figure 17.1 outlines the various types of due
diligence needed in PE.

COMMERCIAL DUE DILIGENCE


Commercial due diligence is the process of obtaining information about buyers and sup-
pliers of the respective company. It also involves understanding the factors that influ-
ence a specific firms performance. The three main factors that could affect the business
are market performance, driving forces of the industry in which the company operates,
and regulatory reforms. Investors select PE to gain profits. By ignoring relevant factors,
they may incur a loss. For example, if the industry in which the company operates is
governed by the bargaining power of the suppliers, production costs can rise depend-
ing upon market conditions, industry dynamics, and product demand. In such a case,
profits may decline when suppliers use their bargaining power. Also, when the cost of
switching companies is less for the customers, then they can use their power to drive
down product prices. In this situation, the companys revenues can decrease.
PE investors need to have confidence in the numbers underlying the targets ex-
pected and current performance and an understanding about the environment and
market conditions. Commercial due diligence helps determine how changes in the
market or competitive conditions affect the value of a particular company. Changes in
Priv at e E qu it y Du e Dil ig e n ce 297

the market may result from new technologies, trends, new buyers or customers, new
geographical markets, and laws and legislation. Commercial due diligence also helps
in understanding earnings before interest and taxes (EBIT) projections that may be
aggressive compared to actual results. Additionally, PE firms can verify such assump-
tions about the success of new products, customers, or markets. The key factors to assess
when considering commercial due diligence are:

Industry attractiveness. This factor involves evaluating the companys market size, its
growth projections, basic profitability factors that drive the market, and the compa-
nys future projections.
Positioning of the target company. This factor entails identifying the factors that help to
create value in the industry and the probability of a sustainable market.
Opportunities for value creation. This factor involves identifying opportunities that
help reduce the costs and enhance revenues of the company and lead to consolida-
tion in operational activities.
Exit strategies. This factor concerns identifying exit strategies and the expected hold-
ing period.

To conduct a commercial due diligence requires comparing the companys external


analysis with the targets competitive position. This process involves evaluating the tar-
gets strategies and examining the companys future growth plans. Customers can some-
times be interviewed to gain their insights. Commercial due diligence helps in making
an informed decision about any PE transaction by analyzing the market trends, size and
growth potential, and their impact on the targets ability to achieve the forecasted results.

O P E R AT I O N A L D U E D I L I G E N C E
Operational due diligence refers to the process of gathering information about a particular
targets operations. It encompasses an in-depth analysis and assessment of selected key
business areas such as the top and bottom line potential of the target company, thereby
evaluating its short-term value. It also helps to detect fraud. According to a report by
KPMG (2013), operational due diligence of investment managers has become increas-
ingly critical due to the recent scandals. Notable examples involve Bernard Madoff, the
admitted operator of a Ponzi scheme that is considered to be the largest financial fraud
in U.S. history, and Weavering Macro Fixed Income Fund Limited, involving grossly
negligent conduct by fund directors.
Operational due diligence focuses on reviewing a companys operational risks, which
are non-investment-related risks. The most common operational risks appearing areas
include: (1) cash management, (2) regulatory and legal compliance, (3) counterparty
management, (4) back office operations, (5) trade operations, (6) valuation policies,
(7) disaster recovery, and (8) information technology.
Effective operational due diligence enhances decision-making and helps reduce risk
by presenting a better understanding of the processes and functioning of the business.
The PE firm may also seek the guidance of specialized consultants who can provide
inputs about a companys effectiveness and efficiency and can also review manage-
ments performance targets with respect to industry benchmarks.
298 p e r f o r m a n c e a n d m e a s u r e m e n t

LEGAL DUE DILIGENCE


Legal due diligence involves assessing possible legal risks that may affect the company
in the future. Such legal risks may relate to the corporate status, contracts, assets, intel-
lectual property, securities, or even employment of personnel of the target company.
Legal considerations are often as important as the financial and tax considerations for
the deals success. Legal strategy of due diligence requires various investment decisions
with respect to tax exemptions and the approval mechanism of the various committees.
It also involves a thorough review of the agreements terms with respect to various part-
ners involved in the fund and its portfolio. The negotiation process with the GPs and
role of LPs in PE funds are also a part of the legal strategy of due diligence.
Legal Due diligence also entails identifying the status of investment managers with
respect to other investment managers. Ascertaining the various laws involved with the
respect to that fund and its portfolio in addition to the fund managers qualifications and
governance structures also falls under this category. Return due diligence encompasses
the fund managers track record and references. Risk due diligence concerns various
aspects involved in risk mitigation at different levels and investments at different stages,
especially upon exit. Correspondingly, a legal analysis of existing employment condi-
tions should be clearly known in order to define the legal impact of the proposed deal
and anticipate possible burdens. Having a clear understanding of the context and timing
of the information and knowledge of the obligations toward the employees and their
representatives is also important.
Legal due diligence can help to avoid legal and regulatory pitfalls by identifying and
mitigating them during the initial stages of the deal. This form of due diligence often
occurs during negotiations with the assistance of a legal expert, who can help draft the
transaction contracts and related documents essential for both parties.
Reasons for conducting legal due diligence include:

Gaining a better understanding of the business. Legal due diligence provides informa-
tion about the target company that enables the buyer to structure the purchase and
to facilitate a smooth communication process between both parties at the time of
negotiation.
Valuing the target company. The buyer can use the information obtained in the legal
due diligence process to determine how much to pay for the target company. The
buyer should thoroughly examine the indicators of value provided in the targets fi-
nancial statements. The PE firm should be able to identify warning signs of under-
stated potential liabilities, lawsuits to which the company is a party, employee bene-
fits and labor arrangements, any insurance policy that may benefit the company, or
rights or obligations under the indemnification provisions. Also, any past criminal or
administrative proceedings or investigations against the firm, its affiliated entities,
and/or its current and former team members should be identified.
Drafting the relevant documentation. The information that is collected during legal due
diligence would be helpful when drafting and negotiating the agreements. If so, any
pre-closing promises and post-closing indemnifications can then be amicably set-
tled. The agreement also highlights the exceptions to the representations and the
expectations from the other party during the time period the contract exists.
Priv at e E qu it y Du e Dil ig e n ce 299

Key issues in legal due diligence are: (1) to discover any major legal issues of the
company, (2) to understand the essential legal drivers that affect the targets profita-
bility, (3) to identify how to create value from the due diligence process, (4) to help
adequately address the due diligence findings that affect the transaction structure and
contracts, and (5) to recognize the hurdles that may lie ahead at the time of transaction
completion whether contractual, regulatory, or other.
While conducting legal due diligence, the PE firm should consider the following fac-
tors: (1) the major legal risks that may be visible from the target, (2) the quality of legal
and compliance management, (3) any hidden contingencies or commitments that the
target may face, (4) the essentials or the scope of price negotiation options, and (5)
trade secrets, intellectual capital, and any contractual or regulatory obstructions that
need to be resolved before finalizing the transaction. Drafting the corporate legal docu-
ments in an accurate manner while implementing the transaction and thereby complet-
ing the deal is also important.

S T R AT E G I C D U E D I L I G E N C E
Strategic due diligence helps explain the variability of a respective companys future cash
flows which are used to determine the target companys value. It helps explain a compa-
nys position in the market. In this sense, position refers to the size of the company (e.g.,
whether the company is a large cap, mid cap, or small cap), closest possible competitors
that are working on selling products and services to same target market, and driving
forces that affect the company by considering its size. After evaluating the companys
position, the PE investor can evaluate its future cash flows, which can help determine
both the present value of the company and its equity.
PE firms should exhibit deal discipline to help them decide whether to proceed with
a new deal or to disinvest from an existing deal. PE firms need to focus on both growth
and diversification in new sectors, new markets, or even new type of deals. According
to Harding and MacArthur (2010), formulating a strong, well-articulated deal thesis in
advance and concentrating on a bottom-up analysis are critical. Ultimately, the aim of
every PE transaction is value creation.
Cross-border transactions occur whenever companies seek out opportunities for
global expansion. However, such deals may not only face regulatory and legal issues
but also complex cultural differences. The PE firm should develop relevant strategies
so that people from different cultures and countries can work together effectively.
Thus, ensuring that the countries involved in a cross-border deal can achieve strategic
alignment necessitates a strategic due diligence. Strategic due diligence begins with
a companys corporate or strategic planning. This process requires understanding
the target firms strategic planning process to get a better picture of the strategic due
diligence.
The PE firm should engage in a transaction only if it improves the strategic position
of its existing business or contributes to its core competencies. Understanding the feasi-
bility of the transaction requires getting answers to the following questions:

Strategy formulation. Is a transaction required to fulfill the corporate or business unit


strategy?
300 p e r f o r m a n c e a n d m e a s u r e m e n t

Finding the target. Who is the best candidate for the PE firm in terms of industry at-
tractiveness and availability?
Conducting due diligence. Can the selected target company meet the investors strate-
gic objectives?
Target valuation: Can the deal be conducted at the right price?
Incorporation: Can the PE firm reconstruct its plans after the deal to extract the full
value of the deal?

Strategic planning can be viewed from a corporate or a business unit level. From a
corporate perspective, strategic planning refers to being in the right portfolio of busi-
nesses. From a business unit level perspective, this process means making the target the
best in its industry. Strategic planning requires assessing the strategic position of the
target in order to extract the maximum benefit. This process requires analyzing the four
Cscosts, customers, competitors, and capabilitiesto assess the full potential of a
business. Even if the deal closes smoothly, the PE firm may not create full shareholders
value because it fails to line up its strategic goals with that of the process of generating
and executing transactions.
In a cross-border deal, an extra layer of complexity exists due to the potential lack of
knowledge about the foreign markets. A complex combination of legal, cultural, eco-
nomic, technological, and political factors drives the potential market penetration levels
for any product of a given country.

FINANCIAL DUE DILIGENCE


Financial due diligence primarily focuses on an in-depth study of accessing the quality
of both earnings and business of assets and finding out the potential undisclosed lia-
bilities and matters that might increase the risk of an investment. The following tasks
should be considered when conducting financial due diligence.

Assess the targets quality of earnings by identifying whether accounting policies are
aggressive or conservative and review company accounts, generally accepted ac-
counting principles, and reporting compliance with the regulators.
Identify the key business drivers including profitability trends.
Identify the concentrations of risk.
Review the assets and the liabilities both on and off the balance sheet.
Review the cash flows, changes in working capital, and capital income and
expenditures.

Following an integrated approach to financial due diligence involves a multidisci-


plinary team consisting of industry experts, market and environment analysts, tax
specialists, employee benefit consultants, actuaries, and experts in technology busi-
ness intelligence to investigate the company and its financials. These experts also help
to identify the investment opportunities that are available in the market and to dis-
cover new ways to attain greater profitability. Although specific value generators differ
among investments, important factors include management expertise, positive cash
flows, market dominance, technical and product leadership, and the customer base.
Priv at e E qu it y Du e Dil ig e n ce 301

Once the due diligence team identifies these factors, it can then provide advice on how
to best carry out the deal.

HUMAN DUE DILIGENCE


Companies spend much time collecting financial, operational, and commercial infor-
mation about the target company, but sometimes they ignore people-related issues in
PE transactions. Human due diligence involves understanding organizational culture and
the roles, responsibilities, capabilities, and the attitudes of people in the company. Ac-
cording to Rimmer and SanAndres (2012), human resource refers to the target compa-
nys management team and broader employee talent pool.
Both parties entering a contract should understand the needs and expectations of
employees to ensure smooth workflow. Making a deal without conducting human due
diligence can lead to the loss of talented employees. Unfortunately, identifying and
qualifying human resourcerelated risks is challenging due to the difficulty of predicting
human behavior and psychology.

C U LT U R A L D U E D I L I G E N C E
Cultural due diligence is the process of assessing, defining, understanding, and mapping
the cultures of different companies during the earliest possible phase of a deal. This type
of due diligence helps assess the quality and sustainability of the companys health and
contributes to a better understanding of the companys future. Avoiding cultural pitfalls
requires a proper due diligence and corporate cultural scan. If the cultural due diligence
reports identify numerous cultural differences, then a common path should be devel-
oped to reduce them.
The recent growth of the international PE industry has garnered much interest due
to larger sizes of capital flowing into the industry. One major area of focus for PE firm
managers should be to identify the similarities in culture to harness them and to explore
differences in culture to mitigate the associated risks. PE firms should be sensitive to the
fact that deals not only have financial implications but also involve the confrontation of
two different organizational cultures. For cross-border deals, different national cultures
and their implications need to be understood. In many cases, deals fail due to associ-
ated cultural issues between different organizational entities that are not handled well
or given the necessary attention.
Perceptions about the work environment vary from one individual, organization, or
country to another. These perceptions influence the way people respond to different
situations and influence their decision-making processes. When PE firms undertake
investments without studying the differences and similarities in respective cultures, fi-
nancial benefits cannot be assured. The advisors and regulators in PE firms need to un-
derstand cultural issues before making investments, and advisors should continue to
work on any cultural issues even after investments are made.
Organizations generally exhibit two layers of culture: organizational culture and
national culture. Organizational culture refers to an informal set of values, norms, and
beliefs that control the way people and groups in an organization interact with each
other and with people outside the organization. Most members of the organization
302 p e r f o r m a n c e a n d m e a s u r e m e n t

have a common set of values and attitudes that shape the work behavior of employ-
ees, their perception of situations, and their responses to them. Hence, the decisions
that people make in organizations are largely dependent on their values and attitudes.
These values and attitudes are passed on from the older members of the organization
to the younger members, making these even more permanent in the organization.
Personal values can become one of the biggest challenges at the time the deal is ex-
ecuted, if the two organizations do not have complimentary cultures. Organizational
culture has a deep impact on important processes such as decision-making, commu-
nication, conflict management, performance management, and teamwork. Organiza-
tions also have a strong history that is deeply etched in the minds of employees and
which also plays a very important role in shaping the organizations culture. Thus, or-
ganizational culture is not easily separable from the organization. Unless the comple-
mentary nature of the two organizations cultures is well researched and established,
the deal may fail.
National culture is another major factor influencing the success or failure of deals
struck by PE firms. According to Hofstede, Hofstede, and Minkov (2010), national cul-
tures can be understood through six dimensions:

Power distance. A national culture attribute describing the extent to which a society
accepts that power in institutions and organizations is distributed unequally.
Individualism vs. collectivism. The degree to which a country prefers to act as individu-
als rather than as members of a group.
Masculinity vs. femininity. The extent to which assertiveness and materialism charac-
terize societal values.
Uncertainty avoidance. The extent to which a society feels threatened by uncertain
and ambiguous situations and tries to avoid them.
Long-term orientation. The extent to which a society emphasizes the future, thrift, and
persistence.
Indulgence vs. self-restraint. The extent to which members of a society attempt to con-
trol their impulses and desires.

Hofstede et al. (2010) cluster countries around these six dimensions. Concerning
international deals, the national cultures of both the countries to which the organiza-
tions belong need to be understood in order to guide performance. Companies often
fail to assess the similarities and differences in their national cultures, which makes
coping with diverse cultures difficult post investment. National culture attributes are
rarely compared and contrasted, resulting in higher chances of failure. A prior under-
standing of the dual layers of organizational and national culture goes a long way for PE
firms to be more cautious, systematic, and equipped to ensure smooth functioning and
higher returns on investments. Cultural due diligence can help gain insight about the
targets cultural strengths and weaknesses. Basic values and cultural factors that drive
a company include loyalty, involvement of key employees, and mutual respect and un-
derstanding among employees. While conducting cultural due diligence, the PE firm
should also identify and prioritize potential improvements or changes to such factors
in the future.
Priv at e E qu it y Du e Dil ig e n ce 303

I N F O R M AT I O N T E C H N O L O G Y D U E D I L I G E N C E
The targets information technology (IT) environment including its business ap-
plication and organizational infrastructure can be critical to the success of a PE deal.
However, management often understates its impact on a companys risk profile. While
conducting IT due diligence, the team should review IT operating and capital budgets
as well as earnings before interest, taxes depreciation and amortization (EBITDA) pro-
jections and the business requirements, functionalities, and services required from the
IT department. This helps in recommending the appropriate amount of IT investment
for the company.
The next step is to review the current and planned IT projects and investments. This
step requires prioritizing projects, estimating the amount of expenditures, and analyz-
ing the IT project pipelines and their impact on capital expenditures (CAPEX).
Part of IT due diligence involves reviewing the technology environment and archi-
tecture, which should focus on a companys IT related operational and financial risks
and opportunities. This process involves thoroughly evaluating whether the IT assets
would scale to meet business growth projections. Another aspect of IT due diligence is
to assess the IT Infrastructure and the network technologies, which requires assessing
the stability of the core IT assets to determine whether upgrades are needed and the
sufficiency of the disaster recovery capability. The last steps are to review the IT organi-
zation, evaluate the capability of the IT leaders and the key resources, and then provide
recommendations on developing the transition plans.

TA X D U E D I L I G E N C E
Tax due diligence is the investigation of the current and the future tax liabilities of a com-
pany. A tax due diligence team should consist of tax professionals who are experienced
in providing tax advice to corporate and PE buyers throughout the life cycle of a trans-
action. Tax due diligence should be conducted in order to identify and understand the
potential deal breakers and to value the drivers and the other areas in which the buyers
have common interests. The team should review the targets historic tax profile, tax plan-
ning, and tax filing history to determine the potential risks and opportunities involved.
Other areas to be reviewed include the amount of tax attributions in case a taxpayer is
insolvent or goes bankrupt, credits and incentives, executive compensation, purchase
agreements, and disclosure schedules. In case of cross-border deals, extra care must be
exercised in understanding the cross-border tax rates that are applicable and the targets
tax profile, so as to avoid any future pitfalls.

SPECIAL DUE DILIGENCE

Companies may also want to find out information that may not be easily available about
the target company. This is known as special due diligence. For example, any existing or
potential newsworthy event involving the target including undue political influence
should be examined during the process of special due diligence.
Due diligence is a technical and time-consuming process that needs to be conducted
with precision. Every bit of information is vital and helpful in analyzing the target.
304 p e r f o r m a n c e a n d m e a s u r e m e n t

Discussion with Key Management Figure 17.2 Summary of the Process of


Due DiligenceThis figure summarizes
Provides the best way to judge the due diligence process and highlights its
promoters and managements ethics. key elements.

Financial and Legal Due Diligence

Uses data provided by the target


company.

Reputation Checks

Includes discussions with the


companys suppliers, customers, and
bankers.

Operational, Technological and


Other Due Diligence

Is performed by technical experts to


oversee all processes.

Ongoing Monitoring

Involves checking ongoing


performance.

Due diligence processes rely on different stand-alone pieces of analysis. Figure 17.2
summarizes the complete due diligence process.

Challenges of Due Diligence


CHALLENGES OF THE CURRENT DUE DILIGENCE PROCESS
Due to the increasing competition for deals, due diligence becomes a concession game
in the following scenarios:

When the value being derived from the deal is eroded,


When the cost from the deal is substantially increased,
Priv at e E qu it y Du e Dil ig e n ce 305

When the PE firm has to agree to terms and conditions that it would not consider in
ideal situations such as limited say in the management and operations of the target
firm.

The teams for conducting due diligence vary. For example, the team conducting fi-
nancial due diligence may not be the same team conducting the technological or the
legal due diligence. If the teams do not have an effective communication system among
themselves, the deal managers may have difficulty taking necessary action on problems
that are highlighted by a particular team. PE firms that have limited access to the target
companies pose a challenge during the due diligence process. Therefore, in an environ-
ment where PE investors have poor access to information, they need to redefine the due
diligence process to get a realistic view of the targets business.
A poor corporate governance environment combined with increasing competition
for investments in the market enables promoters to manipulate the financial position of
the target. When portfolio companies in the due diligence process receive false or mis-
leading information, this issue can lead to a failure in detecting fraud and uncovering
errors. Thus, PE investors could often lack a true understanding of the target company.

CHALLENGES FACED BY LPs IN CONDUCTING DUE DILIGENCE


For an LP, one of the biggest challenges is the lack of publicly available data. The LP typ-
ically has to build a database and rely on outside sources for information. The problem
with outside sources is that they are generally limited. The LP should ask the GP for de-
tailed information for each fund on which they have worked especially when no rating
agencies, standard documentation, or comprehensive published measures are available
for individual funds. Nonetheless, conducting proper due diligence affects all parties
involved in the transaction.

Summary and Conclusions


Conducting due diligence is vital for every PE transaction because it helps to establish
the material facts about the companies in which the investment is being made and to
prevent harm or loss to any party involved with the transaction. Due diligence is impor-
tant from the perspective of both the LPs and GPs. Due diligence by the LPs helps them
find the right GPs who in return can help create value for the LPs by selecting the right
targets. Thus, every PE firm should conduct thorough due diligence before entering into
a transaction. Identifying the skeletons in the closet before potentially entering into
a deal is much better than discovering them after consummating it . Thorough due dil-
igence can be conducted by identifying a target in the right manner, which is decided
based on profit expectations, risk-taking ability, or any other parameter that has been set
forth. Once the target has been selected, different types of due diligence are conducted
including finding in-depth information on the operational, financial, legal, strategic,
human, cultural, IT, and tax sectors of the target. Relevant information can be obtained
from many sources including the Internet, annual reports, newspapers, clients, employ-
ees, or any other reliable source associated with the company that can provide first hand,
306 p e r f o r m a n c e a n d m e a s u r e m e n t

unbiased information about the workings of the company. Due diligence enables the
negotiating team to identify the critical areas of concern involving the transaction. Most
importantly, due diligence allows the acquirer to determine if the transactions proposed
approach fits within the original strategic objectives of the acquirer.
PE fund due diligence is a rigorous undertaking. It requires a thorough top-down
assessment of the different geographies and investment strategies that appear lucrative
for investment. Those conducting due diligence in PE should (1) understand the expec-
tations of their client, (2) be mindful of the objectives of the transaction, (3) maintain
cordial relationships with the target, its advisors, and representatives, (4) understand
the importance of timelines, (5) maintain professional etiquette, and (6) provide timely
inputs to the negotiating team. Although many promising opportunities may be avail-
able, the most important element for due diligence is to identify the risk and returns
involved with each investment opportunity.

Discussion Questions
1. Define due diligence and discuss its importance in PE.
2. Explain the importance of commercial due diligence in PE.
3. Explain why the law is important in conducting due diligence in PE.
4. Discuss strategic due diligence.
5. Discuss how both corporate and national culture can influence a deal.
6. Identify the challenges faced in the due diligence process.

References
Ashurst. 2013. Private Equity Transactions: Overview of a Buy-Out. International Investor Series
No. 9, October, 117. Available at http://www.ashurst.com/doc.aspx?id_Resource=6172.
Axial. 2011. 5 Questions LPs Ask GPs before Investing. Forum, November, 1. Available at http://
www.axial.net/forum/5-questions-lps-ask-gps-investing/.
Elton, Graham, Bill Halloran, Hugh MacArthur, and Suvir Varma. 2011. Limited Partners Set up
Due Diligence. Forbes, August 3. Available at http://www.forbes.com/sites/baininsights/
2011/08/03/limited-partners-step-up-due-diligence/.
Ender, Rainer. 2010. Carrying out Due Diligence on Private Equity Funds. Adveq, November 15.
Available at http://www.adveq.com/nc/media/press-releases/press-details/article/carrying-
out-due-diligence-on-private-equity-funds-iquantum-financei.html.
Ghai, Sacha, Conor Kehoe, and Gary Pinkus. 2014. Private Equity: Changing Perceptions and New
Realities. Insights & Publications, McKinsey & Company, April. Available at http://www.
mckinsey.com/insights/financial_services/private_equity_changing_perceptions_and_
new_realities.
Gottschalg, Oliver, and Bernd Kreuter. 2006. Quantitative Private Equity Fund Due Diligence:
Possible Selection Criteria and their Efficiency. Working Paper, HEC Paris and Feri Institu-
tional Advisors GmbH. Available at https://www.researchgate.net/publication/228202788.
Harding, David, and Hugh MacArthur. 2010. Deal Making: Using Strategic Due Diligence to
Beat the Odds. Insights, Bain & Company, April 21. Available at http://www.bain.com/
publications/articles/deal-making-using-strategic-due-diligence-to-beat-the-odds.aspx.
Hofstede, Greet, Gert Jan Hofstede, and Michael Minkov. 2010. Cultures and Organizations: Soft-
ware of the Mind, 3d Edition. New York: McGraw Hill Professional.
Priv at e E qu it y Du e Dil ig e n ce 307

KPMG. 2013. Operational Due Diligence. Investment Management and Funds. Available at http://
www.kpmg.com/Global/en/industry/Investment-Management/Documents/operational-
due-dilligence-v2.pdf.
Loza, Emile. 2006. Due Diligence in Business Transactions. The Advocate 49:9, 1819.
Ramsay, Ian M., and Baljit K. Sidhu. 1995. Underpricing of Initial Public Offerings and Due Dili-
gence Costs: An Empirical Investigation. Company and Securities Law Journal 13:3, 186201.
Rimmer, Steve, and Aaron SanAndres. 2012. Human Resource Due Diligence. PwC, 112. Avail-
able at http://www.pwc.com/us/en/hr-management/publications/hr-due-diligence.jhtml.
Scharfman, Jason A. 2012. Private Equity Operational Due Diligence: Tools to Evaluate Liquidity, Valu-
ation and Documentation. Hoboken, NJ: Wiley Finance.
Part Five

PRIVATE EQUITY
Uses and Structure
18
Institutional Investors and Private Equity
PARVEZ AHMED
Associate Professor of Finance and Director for the Center for Sustainable
Business Practices, University of North Florida

Introduction
Institutional investors such as endowments, foundations, pension funds, and insurance
firms include private equity (PE) as part of their portfolio mix. Institutional investors
typically allocate a specific amount to PE as part of their overall investments (Sharpe
2010). Most institutional investors fulfill this allocation indirectly through PE funds
rather than through direct PE investments (de Zwart, Frieser, and van Dijk 2012). Suc-
cessful management of PE investments requires properly selecting the PE firm targeted
for investment, managing the investment, and successfully exiting the direct PE invest-
ment. This process is generally recognized as complex and risky, requiring a high level
of expertise and experience that some institutional investors lack. Search costs and the
investors institutional structure affect the selection of the proper PE allocation.
Many consider American Research and Development (ARD), which was formed in
1946, the first modern PE firm. ARD was initially structured as a publicly traded closed-
end fund, marketed mostly to individuals (Liles 1977). This configuration was partly
due to the fact that investors were not initially enamored with this type of fund. The
venture capital (VC) industry, which is a subset of the broader PE world, raised no more
than a few hundred million dollars annually during the 1960s and 1970s. The reduction
in capital gains tax rates in 1978 and the U.S. Department of Labors reinterpretation
of pension investment rules in 1979 fueled the rise of PE. In less than a decade after
these actions, spurred by investments from institutional investors such as pension plans,
annual commitments to PE increased dramatically. Committed capital is the amount of
money committed by investors to PE, but has not necessarily been fully applied to the
investment. It is drawn down over time. Drawdowns, or capital calls, are issued when
the general partner (GP) identifies a new investment and uses a portion of the commit-
ted capital to pay for that investment (Investopedia 2014).
The PE market has experienced tremendous growth over the last two decades. From
1991 to 2013, the number of deals rose from 200 to 2,410 with total value increasing
from $7.5 to $455 billion (PitchBook 2013). This change roughly translates to a 20 per-
cent annual growth rate. The annualized total value of the PE market is equivalent to 2.4

311
312 p r i v a t e e q u i t y : u s e s a n d s t r u c t u r e

$600 3000
$533
2544
2532 2410
$500 2500
2417 $451
Capital Invested in Billions of Dollar

$455
2232
2137
$397
$400 $375 2000

Number of Deals
1728 1637
1583
$300 $280 1500
1377

$200 $174
1000
$168
$152
$114
$100 500

$0 0
2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
Year

Figure 18.1 Private Equity Deal Flows, 2004 to 2013 This figure shows the
number of PE deals and amount of capital invested in these deals from 2004 to 2013.
Source: PitchBook (2013).

percent of the U.S. stock market capitalization. Figure 18.1 shows the total number of
deals and the total capital invested in billions of dollars each year since 2004. PE deals
peaked in 2007 at $533 billion but declined dramatically as a result of the global finan-
cial crisis of 20072008. However, since 2010, the industry has rebounded and by the
end of 2013, it approached its 2007 record. In 2013, institutional investors flocked back
to PE with the highest level of contributions since 2008 (Kreutzer and Canada 2014).
In 2013, investors placed $162.19 billion in U.S. PE funds including buyout (BO),
industry-focused, diversified PE, restructuring, distressed debt, and real asset funds. BO
funds stood at $74.17 billon, with 45 percent accounted for by five firms: Apollo Global
Management, Bain Capital, Carlyle Group, Kohlberg Kravis Roberts & Co., and Silver
Lake. Much of the source capital in PE comes from wealthy private investors called ac-
credited investors as well as endowments, foundations, pension funds, and insurance
companies. Together they comprise the institutional investors in PE.
Market observers attribute the rise of the PE market to the superior performance of
the PE market relative to publicly traded corporations. Harris, Jenkinson, and Kaplan
(2014) find that U.S. BO and VC funds outperform the S&P 500 index by 20 to 27
percent over a funds life, which translates to about 3 percent difference annually. Bailey
(2013) also reports the annualized investment returns in PE for 5-, 10-, and 25-year
periods. As of September 30, 2012, the annualized investment returns were 6.55, 13.71,
and 13.10 percent, respectively. In contrast, the S&P 500 index returned 1.05, 8.10, and
8.61 percent, respectively, in annualized returns. Institutional investors seem to be chas-
ing these returns.
I n s t i t u t i o n a l I n v e s t o r s a n d P E 313

PE is equity capital that is not traded on a public exchange. Table 18.1 shows a com-
parison between private and public equity. Stowell (2013) cites that PE includes the
following:

Leveraged buyout (LBO). A mature company is purchased using equity from a small
group of investors with substantial debt.
Venture capital (VC). Equity is invested in less mature private companies to fund,
launch, and initially develop businesses.
Mezzanine capital. Investments are made using debt or preferred stock without
taking voting control of the firm.
Growth capital. Investments are made in mature companies needing capital infusion
for expansion.

Table 18.1 Comparing Private and Public Equity

Characteristics Private Equity Public Equity


Availability of The information about the The information about the
information investee companies is publicly investee company is readily
unavailable. available to investors as
required by the SEC.
Relationship with The relationship with The relationship with
management management is typically active. management is typically
passive, except for a few
instances of investor
activism.
Return on capital ROC is generated over the ROC is usually short term,
(ROC) long-term as investments may but may be long-term if
take years to recover due to investors take a buy-and-hold
operational changes. approach.
Core advantage PE firms are not answerable Public firms can tap a wide
to shareholders and are not range of investors to invest in
required to file disclosure new and expansion projects.
statements with SEC.
Role of board The board is the source of The board is mostly limited
strategic initiatives and ideas to providing oversight for the
and is more micro than macro. executive team and is more
macro than micro.
Managerial Managerial compensation Managerial compensation
compensation results from the sale of the is the return on assets under
company to the public or management.
another strategic buyer.
Deployment Capital is deployed over time as Capital is usually fully
of capital new opportunities are identified. invested at all times.
314 p r i v a t e e q u i t y : u s e s a n d s t r u c t u r e

GPs in PE provide capital and management expertise. They earn most of their fixed
revenue from management fees. In contrast, limited partners (LPs) provide capital via
funds and have their investment contracts often locked in for 10 to 12 years.
A financial institution is an enterprise such as a bank, trust company, insurance com-
pany, savings and loan association, or investment company specializing in the handling
and investing of funds. Financial institutions stand between the security issuer (firm)
and the ultimate owners of the security (investors). Financial institutions are also finan-
cial intermediaries, including banks, investment companies (pensions, mutual funds,
hedge funds, and PE), insurance companies, and credit unions.
Most institutional investors gain access to PE through funds rather than direct invest-
ments. Entering, managing, and exiting direct PE investments generally requires expe-
rience and expertise that many institutional investors lack. A PE advocacy group, the
Private Equity Growth Capital Council (PEGCC), contends that since the 1980s, pen-
sions investments in PE funds greatly increased in number and size, becoming the larg-
est contributor of capital to PE investments between 2001 and 2011. Nearly 43 percent
of capital invested in PE came from pension funds of which 30 percent came from public
pension funds. Additionally, endowments and foundations contributed 19 percent of
capital invested in PE (Bailey 2013). A report by Bonafede, Foresti, and Walker (2012)
of Wilshire Consulting shows that the percentage of assets allocated to PE by state pen-
sion plans more than doubled from 3.9 percent in 2001 to 8.2 percent in 2011. PEGCC
asserts that the largest 151 U.S.-based public pension funds have $277 billion invested
in PE, which represents about 10 percent of their investment portfolios (Bailey 2013).
Sahlman (1990) asserts that funds sharing a common organizational structure gen-
erally make PE investments. According to Axelson, Strmberg, and Weisbach (2009),
these funds raise equity at inception and then raise additional capital when new invest-
ments are necessary. This additional capital is usually debt, particularly when the invest-
ment can be secured with collateral such as in buyouts. The funds are usually organized
as limited partnerships, with LPs providing most of the capital and GPs making invest-
ment decisions and receiving a substantial share of the profits typically 20 percent. This
structure is instituted because GPs are skillful in identifying and managing investments
that have the potential to be profitable. However, GPs have to rely on LPs to provide the
external capital needed to finance such profitable opportunities.

Private Equity Structure


Investment managers also known as sponsors create private funds to raise capital for
more risky investment opportunities usually in a specific industry, sector, or geographic
region. PE funds can be viewed as blind pools under which passive investors make a
commitment to invest capital over time. These passive investors trust the funds sponsor
to source, acquire, manage, and divest the funds investments. Fund sponsors are moti-
vated to undertake this task because they can collect high management fees in addition
to participating in profits generated by the funds investments.
In contrast, investors in PE want to earn a high rate of return on their invested capital
through access to a portfolio of investments that is unavailable to the general public. PE
funds come in a variety of types. For example, VC funds invest in companies that are
I n s t i t u t i o n a l I n v e s t o r s a n d P E 315

in early development stages. Growth equity funds invest in firms that are in later stages
of development but are pre-IPO. BO funds acquire controlling interests in companies,
usually planning to take them public at a future time. Distressed funds invest in debt
securities of firms, typically at a large discount, for companies that are experiencing fi-
nancial hardships. Figure 18.2 shows the structure of a PE fund. The structure involves
the following:

The fund either by itself or as fund-of-funds is a pool of capital with direct opera-
tions. Institutional or private investors acquire interests in the fund and in turn make
investments in a portfolio of companies.
The GP has the legal power to act on behalf of the investment.
The management company or investment advisor plays an advisory role to the fund.

Institutional investors have two main ways to channel their investments. First, they
can channel their PE funds to firms that specialize in identifying, investing, managing,
and harvesting investments in private companies. Second, institutional investors can
channel their investments to a fund-of-funds that invests in a portfolio of PE funds.
In both cases, investors make a commitment to invest and the PE fund calls on the
capital when new investment opportunities materialize. Thus, the amount committed
by investors and the amount invested by the fund may differ over time. Several years
may pass before investors achieve their desired level of investment in private compa-
nies (Takahashi and Alexander 2002). Ljungqvist and Richardson (2003) show that
about three to six years are required to fully invest the funds and eight to 10 years may
pass before the internal rate of return (IRR) on these investments becomes positive.
PE firms rationally respond to attractive investment opportunities signaled by public
market shifts (Gompers, Kovner, Lerner, and Scharfstein 2008). Ljungqvist and
Richardson also show that the rate of investment depends on the underlying market
conditions (i.e., available investment opportunities) and competition among firms
(i.e., the origination levels of PE funds). The cost of financing also affects buyouts

Sponsor

Institutional
Investor General
Investment via Fund of Funds Investment via Funds Investment
Advisor Partner
Management
Private Equity Fees
Fund-of-Funds
Fund
Carry

Portfolio Portfolio Portfolio Portfolio Portfolio Portfolio


Company Company Company Company Company Company

Figure 18.2 Model Investment Structure of Private Equity This figure shows
a stylized model indicating the investment structure of PE. The model shows how in-
stitutional investors access PE through direct investment in funds and investment in
fund-of-funds.
316 p r i v a t e e q u i t y : u s e s a n d s t r u c t u r e

(Gompers and Lerner 2000). As Takashi and Alexander show, expectations about
drawdown and distribution rates also influence investors decisions to commit funds
to the PE firm.
PE funds are typically formed as limited partnerships or limited liability corpo-
rations (LLCs) (Cumming and Johan 2009). The U.S. government considers these
limited partnerships and LLCs as pass through securities that are not subject to
corporate income tax. The funds earnings are passed to the partners and taxed only
once at the investor level. This limited liability feature allows investors to benefit
from this arrangement in the same way that shareholders benefit from public corpo-
rations. LPs and the members of the LLCs are limited to making capital investments
only (Practical Law 2014). The structure of the partnership agreement between in-
stitutional investors (LPs) and the fund manager (GP) contains covenants that spec-
ify the rules of engagement. Cumming (2010) outlines the terms of the covenants
as follows:

Description of the investment mandate including the target companys stage of de-
velopment.
Restrictions on investment activities such as limits of investment horizon, amount of
equity to be invested, and prudential limits on geographical exposure.
Limits on financing such as maximum borrowings, cross-fund investing between
two managers, and reinvestment of capital.
Investor rights such as the ability to remove managers, reporting and accounting,
and investor representations on advisory committees.
Economics of the fund such as a management fee (typically 1 to 3 percent of capital
committed to the fund) and a performance fee (typically 20 to 30 percent of the
profits after return on capital but only after the fund has crossed a minimum thresh-
old rate of return, typically 8 to 10 percent annually).

Gompers and Lerner (1996) provide a more in-depth account of these covenants
and the economic motivations for such covenants. Restrictions on how much to
invest in any one firm is intended to ensure that GPs do not reinvest in opportuni-
ties that have not yielded good returns. The restrictions are designed to avoid what
is often described in behavioral finance as the break even effect, in which investors
are prone to accepting a risky double or nothing proposition after losing money in
a bet (Bruce 2010). They prevent follow-on funding from chasing poorly perform-
ing firms. GPs do not receive their share of the profits until LPs have recovered their
investments. This arrangement can be thought of as a call option for GPs who may
gain disproportionately by increasing the risky bets in the portfolio. This limitation
is frequently expressed as a maximum percentage of capital invested in the fund that
can be invested in any one firm or as a percent of the current value of the funds assets.
Additionally, given that GPs hold call options, they may be tempted to increase the
riskiness of their portfolio by using debt. Increasing the riskiness as measured by the
variance or standard deviation of the portfolio increases the value of the call option.
Partnership agreements often limit borrowed funds and may limit debt to a set per-
centage of committed capital or restrict the maturity of the debt to ensure that all
borrowing is short term.
I n s t i t u t i o n a l I n v e s t o r s a n d P E 317

GPs in PE may manage several funds concurrently. This can tempt opportun-
istic behavior such as directing money from a second fund to rescue bad invest-
ments made in a prior fund. Consequently, partnership agreements for later funds
frequently require that the funds advisory board reviews such investments or that a
majority or supermajority of LPs approve these investments. Finally, reinvestment
of profits may also require LP and advisory board approval. Such reinvestments
may be prohibited after a certain date or proportion of committed capital has been
invested.
Restrictive covenants also prevent GPs from investing their personal funds in the
firms in which the PE is invested. GPs are also restricted from selling their partner-
ship interests for fear that any reduction in the GPs stakes may reduce their incentive
to monitor LPs investments. Covenants may prohibit the sale of general partnership
interests outright or require that a majority or supermajority of LPs approve such sales.
Covenants may also prohibit GPs from fundraising until the PE fund has invested a cer-
tain percentage of the committed funds or until a specific set date has passed. Finally,
any new GPs are restricted in order to prevent dilution in management quality. Such
additions may require advisory board and LP approval. Covenants also restrict invest-
ments to a set percent of capital or asset value in a given investment class. The concern
that GPs may be receiving inappropriately large compensation drives this restriction.
For instance, the average money manager who specializes in investing in public secu-
rities receives an annual fee of about 0.5 percent of assets under management (AUM),
while venture capitalists receive 20 percent of profits in addition to an annual fee of
about 2.5 percent of capital. LPs seek to limit the amount of PE invested in public
securities.
Covenants can reduce the inherent principalagent conflict of interests in PE. Fund
managers ensure quality by staying consistent with the stated style of the portfolio
(Cumming, Fleming, and Schwienbacher 2009). The PE compensation structure may
mitigate some of the inherent principalagent conflicts. Gompers and Lerner (1999)
describe the two primary modelsthe learning and signaling models. In the learning
model, neither the venture capitalist nor the investor has any initial certainty about the
PEs potential. In its early days, PE did not come without explicit pay-for-performance
incentives. By establishing precedence and a good track record, for either selecting at-
tractive investments or adding value to the portfolio, the PE manager preserved the
ability to raise more money later. These reputational concerns lead to lower pay-for-
performance provisions for smaller and younger organizations. Once a GP establishes
a reputation, explicit incentive compensation is needed to reduce the divergence of ob-
jectives between the principals and agent.
In contrast, the signaling model suggests that PE managers know their abilities while
investors do not. Thus, funds have higher pay-for-performance sensitivities and a lower
base compensation to signal to the investors that the fund is mitigating principalagent
conflicts. PE managers with strong abilities try to separate from the pack by accepting
a riskier pay-for-performance fee structure. Results from Gompers and Lerner (1999)
show that using reputation as a proxy for the age and size of a PE firm, the compensation
of established funds is significantly more sensitive to performance and more variable
than that of other funds. Older and larger firms also have lower base compensation. Per-
formance and pay sensitivity do not appear to be related.
318 p r i v a t e e q u i t y : u s e s a n d s t r u c t u r e

Institutional Investor Behavior in Private Equity


The literature on institutional investor behavior in PE is sparse. One way to proxy such
behavior is to examine other equity asset classes such as hedge funds and mutual funds.
This section also contrasts institutional investor behavior with that of individual inves-
tors. Barber and Odean (2008) show that institutional investors, unlike individual in-
vestors, do not display attention-driven buying (i.e., act as net buyers on high-volume
trading days). While individual investors face both economic and behavioral hurdles to
selling, often eschewing short sales, institutional investors such as hedge funds routinely
sell short. Compared to individual investors, institutional investors also hold larger num-
bers of stocks in their portfolios, which allow them more choices when selling, even if
they avoid using short sales. Furthermore, attention is not as scarce a resource for institu-
tional investors as it is for individual investors. Institutional investors, who are less con-
strained by time limits, devote more time to searching for stocks than do most individual
investors. Institutions may limit their search to stocks within a particular sector or based
on specific criteria, thus reducing attention demands. Institutional investors in PE also
face similar circumstances such as not being encumbered by time limits or resources.
Grinblatt, Titman, and Wermers (1995) show that mutual funds, which are by defini-
tion institutional investors, have a propensity for herding behavior, in which they buy and
sell the same stocks at the same time. They are also momentum investors, buying past win-
ners (i.e., stocks that have performed well in the recent past) although not systematically
selling past losers (i.e., stocks that have underperformed in the recent past). On average,
funds following momentum strategies realize significant excess performance. The relation-
ship between the tendency to buy past winners and performance is positive. However,
funds exhibiting herding behavior do not have spectacular performance and are subsumed
when controlling for any tendency of the fund to buy past winners. Thus, momentum is
more impactful in determining future success than herding. Still, studies show that herding
by institutional investors is increasing (Lakonishok, Shleifer, and Vishny 1992; Grinblatt
et al. 1995; Wermers 2002; Sias 2004; Patterson and Sharma 2006; Gutierrez and Pirinsky
2007). The institutional inertia that leads to herding in mutual funds is likely to persist in
PE because investing in PE equity is far easier than divesting from past investments.
Institutional investors differ from individual investors because they carry larger portfo-
lios and have greater resources available to them. Systematic differences exist across institu-
tions. Institutional investors have more sophisticated organizational structures and more
defined investment objectives. These differences can have a profound impact on portfolio
allocation decisions and thus influence investment returns. Shleifer and Vishny (1997)
suggest that information asymmetries between investors and intermediaries limit arbi-
trage and thus affect strategies for asset allocation and eventual returns on the portfolio.
As Lerner, Schoar, and Wongsunwai (2007) note, accounts by both observers and
practitioners suggest that substantial variation exists in the investment criteria and so-
phistication of PE investors. Lerner et al. (2007, p. 732) summarize the statement of a
large endowment manager:

[Endowments] on the cutting edge choose from a broader opportunity set.


By considering alternatives outside the mainstream, investors increase the
I n s t i t u t i o n a l I n v e s t o r s a n d P E 319

likelihood of discovering the next big winner well before it becomes the next
big bust. By evaluating managers without the requisite institutional character-
istics, investors might uncover a highly motivated, attractive group of partners.
Operating on the periphery of standard institutional norms increases oppor-
tunity for success.

Unlike endowments, institutional investors in other organizations may lack a clear


understanding of asset classes. Inappropriate incentive structure, poor human resource
practices, and conflicting objectives contribute to failure in choosing the proper asset
classes in which to invest.
Cumming (2010) contends that institutional investors prefer to syndicate invest-
ments in order to diversify risk. Institutional investors may also meddle in corporate
governance in order to reduce agency costs. Institutional investors in PE often have sim-
ilar effects on a firm as active block-holder investors of publicly held stocks. Just as inves-
tors holding large blocks of stocks in a company wield considerable clout, institutional
investors can also have similar influence. Institutional investors have primarily invested
in PE through PE funds. This preference is related to the desire to earn the potentially
lucrative returns that PE offers and yet be able to mitigate possible agency costs. The
governing structure of a limited partnership helps fund managers focus on long-term
performance.

Determinants of Institutional Investment


in Private Equity
The ownership of any privately held corporation is concentrated (i.e., the principals
are not atomistic). This concentration gives owners sufficient control to monitor and
discipline management. The nature of agency problems in such a situation shifts from
the classical model of divergence in objectives between management and sharehold-
ers ( Jensen and Meckling 1976) to conflicts between controlling and non-controlling
owners (Shleifer and Vishny 1997).
Easterbrook and Fischel (1986) suggest that investors in privately held corpora-
tions are primarily concerned about the expropriation of their interests at the hands
of the controlling shareholders. The fear of minority expropriation is important for
institutional investors because the law prohibits them from being controlling owners
and participating in making decisions. Institutional investors fear expropriation even
in publicly held firms (Giannetti and Simonov 2006). The problem is exasperated
in privately held firms. Unlike publicly held corporations, investors in PE do not
have the option to liquidate their investments if they are dissatisfied with the con-
duct of management. The lack of regulatory oversight for PE further complicates the
situation.
Institutional investors can use various tools to mitigate the inherent agency problems
in PE. Pagano and Roell (1998) and Bennedsen and Wolfenzon (2000) suggest that in
order to avoid the perception that minority investors can get shut out, PE managers
should dilute control in a way that no one group of investors can assert unilateral control
320 p r i v a t e e q u i t y : u s e s a n d s t r u c t u r e

over decision-making. This dilution of control would make PE investors somewhat at-
omistic like their public equity counterparts. Nielsen (2008) asserts that another way to
mitigate this potential risk is to invest in firms with governance mechanisms that make
the companys operations more transparent to minority shareholders. Institutional in-
vestors should avoid investing in PE that does not take proactive steps to reduce agency
costs in ways to assure minority shareholders that they need not fear expropriation by
controlling shareholders.
Another factor that may mitigate the problems from agency cost is social ties (i.e.,
the degree to which an LP is socially and personally related to the GP). Social ties may
affect decision-making. Cornelli and Goldreich (2001) contend that investment banks
use their strong relationships with institutional investors when pricing and distribut-
ing corporate securities. Ljungqvist, Marston, and Wilhelm (2009) show that banks
tend to co-underwrite securities with other banks with which they have long-standing
relationships. PE firms often syndicate their investments with other PE firms, rather
than investing alone (Lerner 1994). Once making the investments, the PE firm uses
its professional network of head hunters, patent lawyers, and investment bankers to
help their target company succeed (Sahlman 1990). Hochberg, Ljungqvist, and Lu
(2007) assert that PE generally exhibits a propensity for professional networks and
business relationships and is less reliant on arms-length transactions. That is, PE firms
enter into deals based on who they know in their professional circles, not necessarily
added shareholder value. Syndication relationships affect the ability of the PE firm
both to select promising companies and to nurture investments, thus adding value
to them. Information exchange via professional networks and business relationships
often divulges more private information than information exchange in an arms-length
transaction. This action can increase a firms transactional efficiency (Uzzi 1996). In-
stitutional investors are more likely to become LPs in PE if they have direct social ties
with the GP.
Agency costs are further reduced when professional networks build trust. As trust
improves, uncertainty and risk decreases. Freiburg and Grichnik (2012) show that
direct and indirect professional networking influences investment decisions for PE.
Social ties not only transfer information between LPs and GPs but also engender trust,
which then further reduces agency costs by removing information asymmetry. Social
ties also affect investment decisions for PE by preserving capital reputation. GPs should
consider a firms professional network when choosing LPs. Even indirect social ties can
be beneficial as they can transmit information about the GP to the LPs who otherwise
would suffer from information asymmetry. Indirect ties also reduce the cost of obtain-
ing information because they reduce the required time and investment to collect it (Na-
hapiet and Ghoshal 1998).
Diversification is generally the motive for institutional investors to invest in PE.
Benefits from diversification accrue if PE has a low correlation with public equity. Ph-
alippou and Zollo (2005) find that the performance of PE funds is positively correlated
with both business cycles and stock market returns. Moskowitz and Vissing-Jrgensen
(2002) find a correlation of 0.7 between the book equity return of public equity and
PE from 1963 to 1999. Thus, the potential benefits from diversification, which require
a low correlation between PE and the rest of the institutional portfolio, are not all
they have been touted to be. Nonetheless, Nielsen (2008) asserts that institutional
I n s t i t u t i o n a l I n v e s t o r s a n d P E 321

investors should still give PE serious consideration if this market offers investment op-
portunities (e.g., new technologies) that may be unavailable in public equity markets.
In other words, institutional investors should be careful when investing in PE. They
should do so only if sufficient transparency exists to mitigate agency costs and if the
investment opportunity brings diversification benefits that public equity markets do
not offer.
Another factor affecting institutional investor strategy in PE concerns commitment
and recommitment. Typically, institutional investors participate in PE funds by com-
mitting a certain amount of capital to them. Fund managers then call these investments
over a period of time at their discretion to be invested based on opportunities that arise.
Leaving some committed capital untouched is common. After a few years of investing,
payouts from disinvestments or distributions may start, often before all of the commit-
ted capital has been invested. Typically, these disinvestments or distributions are re-
committed to new PE funds. De Zwart et al. (2012) contend that cash flows to PE are
uncertain and uncontrollable by institutional investors, which can create problems with
portfolio misallocations. Unpredictability of cash flows and illiquidity of the markets
may affect an institutional investors decision to invest. If committed capital is not fully
deployed, it can create a cash drag on the portfolio. However, if institutional investors
do not carefully monitor their asset allocation, they can easily become over invested by
making excessive commitments. Over investing can also undercut the performance of
portfolios as the institutional investors may not have the cash available when capital is
called. Institutional investors need an efficient recommitment strategy to stave off both
under investing (thus losing economic value) and over investing (thus creating cash
shortages and problems with liquidity).
The final factor that may affect institutional investment in PE is the institutional af-
filiation of the PE fund. Five major categories of PE funds are available depending upon
their institutional affiliation: (1) independent funds, (2) funds affiliated with banks, (3)
funds affiliated with insurance companies, (4) subsidiaries of industrial corporations,
and (5) funds sponsored by public institutions. Huyghebaert (2010) finds that the in-
stitutional affiliation of the PE fund bears only a weak relationship with the industries
selected by the PE fund. However, a managers choice of a funds investment stage and
geographical area depends largely on its institutional affiliation. Funds affiliated with
banks commonly prefer investments in life sciences, bio-tech, and health-related firms.
Funds by banks also focus on domestic investments. Funds sponsored by life insurance
companies and pension funds show a lack of interest in start-up ventures. Further, they
do not have any sector preference or specific geographical focus. Corporate-backed
funds favor start-up ventures and are less inclined to invest in late-stage companies.
Corporate-affiliated funds operate in a wide spectrum of industries and tend to focus on
global investments. Finally, funds sponsored by governments and public organizations
are less likely to invest in late-stage ventures. Also, they have a propensity for invest-
ments in high-tech electronics. Their geographical focus tends to be limited to domestic
investments.
Mayer, Schoors, and Yafeh (2005) examine the preferences of LPs when deter-
mining the compositions of PE funds. Funds from financial institutions such as
banks, pension funds, and insurance companies usually focus on late-stage invest-
ments. Banks, in particular, seem to invest more in later stage VC funds. In contrast,
322 p r i v a t e e q u i t y : u s e s a n d s t r u c t u r e

corporate and individual investors focus more on early stage VC funds. Banks and
governmental investors also seem to invest more often in domestically active funds
while insurances, corporations, and individuals invest in broader, global funds. The
differences in PE across the world seem somewhat muted, but institutional backing
matters.
According to Lerner et al. (2007), endowments earn an annual return that is about
21 percent higher than that earned by other institutions such as banks, insurance compa-
nies, and pension funds. Additionally, follow-on funds chosen by endowments achieve
significantly higher mean IRRs, while banks are the worst reinvestors. Institutional in-
vestors acting as LPs differ in their ability to evaluate the quality of funds and make in-
vestments based on the information available. Major differences in sophistication exist
among investors. Although overall funds selected by endowments perform better than
those selected by other types of institutional investors, whether this situation is due to a
general preference of certain types of investments or to the sophistication of better asset
allocation is unclear. Thus, institutional investors face two distinct issues when selecting
PE: how they define their preferences for PE and how performance differs based on the
type of PE selected.
Institutional investors determine the performance of PE by calculating the IRR of
the cash flows invested. The investment round-trip determines the managers ability to
add value and thus affects the performance of the funds. Investors can compare their
returns to industry averages by using an index of fund averages. For example, the S&P
Listed PE Index consists of leading listed PE companies that meet specific size, liquid-
ity, exposure, and activity requirements. The PE Quarterly Index (PrEQIn) captures the
average return earned by investors on average in their PE portfolios, based on the actual
amount of money invested in PE partnerships. These indexes are often the benchmarks
to which PE is compared.
Currently, analysts favor a market-adjusted PE return model. Cumming (2010) ex-
plains that PE capital has an opportunity cost, given that the return generated could
be invested elsewhere, particularly in public markets. The market-adjusted PE return
model incorporates opportunity cost measures into return measurements. Ljungqvist
and Richardson (2003) create a profitability index that discounts outflows, inflows, and
the market index rate at the risk-free rate.
The performance of PE funds shows wide variance. Ewens, Jones, and Rhodes-Kropf
(2013) calculate that the average BO fund has a value-weighted IRR of 14 percent, a
beta of 0.72, and an alpha of 4 percent. The average VC fund has a value-weighted
IRR of 15 percent, a beta of 1.24, and an alpha of zero. Lerner et al. (2007) report a
mean value-weighed IRR of 14 percent for VC funds and 0 percent for BO funds. In
contrast, Phalippou and Gottschalg (2009) find that the mean dollar-weighted IRR
for VC funds is higher than that of BO funds at 13.23 and 16.79 percent, respectively.
They also find that in most cases PE does not outperform public equity. Only buyouts
outperform on a dollar-weighted basis. However, the findings of Kaplan and Schoar
(2005) indicate that BO funds underperform public equity although VC funds outper-
form public markets. All researchers report a wide variation in the returns to PE. One
caveat to all these studies is that the sample period tends to be small. Lack of externally
provided data over very long periods of time is a function of the relative youth of the
PE industry.
I n s t i t u t i o n a l I n v e s t o r s a n d P E 323

Types of Institutional Investors in Private Equity


An institutional investor is an individual or organization that trades securities in large
enough share quantities or dollar amounts that it qualifies for preferential treatment and
lower commissions. Institutional investors face fewer protective regulations because
regulatory agencies view them as being more knowledgeable and better able to protect
themselves when investing. This observation is because institutional investors have the
resources to hire professionals while individual investors typically do not. The role and
influence of institutional investors have grown over time. Investment companies in the
United States managed $13 trillion in assets for more than 92 million U.S. investors
at year-end 2012 according to the Investment Company Institute (2012). Blume and
Keim (2012) show that institutional management of public equities rose from about
8 percent of market capitalization in 1950 to about 67 percent in 2010. According to
The Conference Board (2010), institutional investors owned 73 percent of outstanding
equity in the 1,000 largest U.S. corporations.
Pension funds account for more than half of investments in PE (Cumming, Fleming,
and Johan 2011). Financial institutions and endowments together account for another
one-fifth of investors. Thus, institutional investors account for nearly $7 out of every $10
invested in PE. Institutional investors typically invest indirectly in private companies
using PE funds. Occasionally, banks and other financial institutions invest directly into
the companies via captive divisions or business units (Cumming et al. 2009).
Table 18.2 shows the financial assets of institutional investors. Nearly one-third of
institutional investors in the United States are public and corporate pension funds. Pen-
sion funds act as investment fiduciaries for the contributing employers. Defined benefit
pension plans emphasize investing in bonds or other inflation-neutral assets. Defined

Table 18.2 Financial Assets of Institutional Investors by Country

Values in USD in billions

Country Insurers Pension Mutual Total Assets GDP Assets as


Funds Funds % of GDP
Australia 241 556 700 1,497 754 199
Canada 391 550 491 1,432 1,269 112
Euro Zone 4,665 1,149 4,308 10,122 9,304 109
Japan 3,243 997 470 4,710 4,367 107
United Kingdom 1,979 1,487 547 4,013 2,373 169
United States 5,601 7,306 8,905 21,812 13,244 164
Total 16,120 12,045 15,421 43,586 31,311 139

Note: This table shows the market value of the financial assets held by institutional investors in
various countries and regions. The Euro Zone includes Belgium, France, Germany, Italy, Luxembourg,
the Netherlands, and Spain.
Source: Adapted from Hobohm (2010) with data from 2007.
324 p r i v a t e e q u i t y : u s e s a n d s t r u c t u r e

contribution plans often include equities and alternative assets. Understanding the
general landscape of institutional investors is important as industry insiders claim that
institutional investors are pouring money into private equity at an astounding rate
(Mergenthaler and Moten 2008). Additionally, 62 percent of current institutional in-
vestors in PE expect to increase their allocations in the near future. Thus, understanding
the general behavior of institutional investors can help explain their role in PE.
Table 18.3 shows the largest 20 pension plans in the world. Life insurance policies
often provide retirement benefits, which enable them to act more like defined contribu-
tion pension plans. U.S. insurance companies hold nearly 40 percent of their assets in
equities (Committee on the Global Financial System 2011). Investment corporations
such as hedge funds are also a type of institutional investor. Investment corporations are
distinguished from pensions insofar as they fulfill a specific saving function on behalf of

Table 18.3 Largest Retirement Funds

Rank Fund Country Total Assets


(USD in millions)
1 Government Pension Investment Japan 1,292,003
2 Government Pension Fund Norway 712,606
3 ABP Netherlands 372,860
4 National Pension South Korea 368,450
5 Federal Retirement Thrift United States 325,682
6 California Public Employees United States 244,754
7 Local Government Officials 1 Japan 201,443
8 Central Provident Fund Singapore 188,430
9 Canada Pension 1 Canada 184,425
10 National Social Security China 177,486
11 PFZW1 Netherlands 177,311
12 Employees Provident Fund Malaysia 175,720
13 California State Teachers United States 155,739
14 New York State Common United States 150,110
15 Florida State Board United States 134,345
16 New York City Retirement United States 132,071
17 Ontario Teachers Canada 130,198
18 ATP Denmark 129,009
19 GEPF2 South Africa 122,225
20 Pension Fund Association Japan 119,199

Note: This table shows the largest retirement funds ranked in millions of U.S. dollars. U.S. fund
data are from the Pension & Investments 1,000 published February 4, 2013; non-U.S. fund data are as
of December 31, 2012.
I n s t i t u t i o n a l I n v e s t o r s a n d P E 325

clients and also earn returns for their shareholders. P&I/Towers Watson (2013) esti-
mates that the worlds top 500 asset managers managed a total of $68 trillion by end of
2012. Asset managers in North America accounted for nearly 52 percent of the market.
According to Braun (2011), the average pension fund increased its allocation to PE to
8.8 percent in 2010 from 3 percent in 2000. The 10 largest public pension funds paid PE
$17 billion in fees from 2000 to 2010. Understanding the pension plan industry is im-
portant because pension plans pay billions in fees to PE, at times making up more than
50 percent of PEs financing (Solomon 2011). Underfunding of pension plans is forcing
pension plans to look toward the lucrative returns in PE.
In the United States, legislation permits banks and bank holding companies to di-
rectly invest in PE. The Small Business Act of 1953 allows banks to set up subsidiaries
as investment corporations. The Bank Holding Company Act of 1956 permits banks to
be shareholders in companies as long as their ownership does not exceed 5 percent of
outstanding voting shares. The Gramm-Leach-Bliley Act of 1999 allows banks to set up
financial holding companies that can make direct investments in PE, with the stipula-
tion that the holding period does not exceed 15 years. By investing in PE, banks hope
to take part in any future LBOs. This may prompt banks to invest in less lucrative PE for
the opportunity to take part in the companys future banking needs (Lerner et al. 2007).
Endowments and foundations also invest in PE but are considerably smaller than
other types of institutional investors. Foundations typically have a 5 percent minimum
payout requirement, whereby they must make eligible charitable expenditures that
equal or exceed roughly 5 percent of their value to retain their favorable tax status. Even
without such restrictions, endowments have payout patterns similar to foundations.
Thus, endowments and foundations have to generate a 5 percent return per year, net of
all fees, to preserve their asset base. With the notable exception of large endowments
such as at Harvard University and Yale University, most foundations are small and tend
to invest in a fund-of-funds, which is a fund that invests in a portfolio of PE funds. When
investing in PE, fund-of-funds assume limited partnerships.
Institutional investors have three primary motivations to invest in PE (Cumming
et al. 2011): (1) search costs associated with investor size, type, and location; (2) spe-
cific human capital associated with the investors decision-making structure; and (3)
liquidity time preferences associated with the desire to achieve exposure to PE as soon
as possible.
Jegadeesh, Kraussl, and Pollet (2009) examine the risk and expected returns of PE
investments based on the market prices of publicly traded funds-of-funds that invest in
unlisted PE funds. Results indicate that the market expects LPs of unlisted PE funds to
earn annual positive abnormal returns of about 0.5 percent. The market expects listed
PE funds to earn abnormal returns that are statistically indistinguishable from zero after
fees. Both listed and unlisted PE funds have market betas close to one. PE fund returns
are negatively related to the credit spread and positively related to growth in gross do-
mestic product (GDP). PE investments also suffer from high search costs compared to
listed public equity. The willingness of institutional investors to incur such search costs
depends on the investors size. Most small institutions do not have the time or skill to
identify quality GPs and negotiate favorable limited partnership terms.
Institutional investors who prefer PE typically make the requisite investment
in human capital to help them identify investments in a market beset with large
326 p r i v a t e e q u i t y : u s e s a n d s t r u c t u r e

informational asymmetry. Lerner et al. (2007) show that early adopters of PE spend
resources building teams that have the skills and industry insights to access private in-
vestments. Thus, having a specialized PE team can greatly help institutional investors
select PE investments.
Fully investing all PE capital can take as long as 10 years. The PE firm draws down
or calls the funding obligation or commitment over a period of time after identifying
new investment opportunities. Cumming et al. (2011) suggest that PE firms typically
call 75 to 80 percent of capital committed to their fund over the first five years, and re-
serve the remaining commitment to finance follow-on investments in companies and
management fees over the next five years of the fund. Therefore, investors are required
to pay their commitment over a 10-year period often in an uneven manner. Thus, a dif-
ference exists between the amount of PE committed by investors and the actual amount
invested. Gompers and Lerner (1999) and Ljungqvist and Richardson (2003) model
the drawdowns in PE funds and show that their activities vary according to the supply
of investible opportunities, competition for deals, and cost of financing. Distributions
given to investors from PE firms are dependent on the state of public finance markets
and the economy. Takahashi and Alexander (2002) show that expectations about draw-
downs influence the institutional investors decisions on capital commitment.

Summary and Conclusions


This chapter examines the motivations and types of institutional investors in PE. Man-
aging successful PE investments requires properly selecting the private firm targeted
for investment, managing that investment, and successfully exiting the direct PE in-
vestment. Institutional investors in PE are motivated not only by the usual risk-return
concerns of ordinary investors but also by indirect, non-financial returns. Various costs
influence the decision to invest in PE including search costs, specific human capital, and
liquidity time preferences.
Analysts often use IRRs to determine PE performance. The PrEQIn measures the
return earned by investors on average in their PE portfolios. Other indexes, such as the
market-adjusted models, incorporate the opportunity cost of PE. An alternative profit-
ability index discounts outflows, inflows, and the market index rate at the risk-free rate.
All three methods report wide variation in the performance of PE. A lack of externally
provided data over long periods exists because the industry is relatively young.
Given the wide variation in performance, institutional investors incur greater search
costs to determine good investments in PE. The selection of a good manager is essential
to finding an attractive investment. More academic research is necessary to fully under-
stand the nature of PE and its challenges to institutional investors.

Discussion Questions
1. Describe an institutional investor and list major types.
2. Describe the growth and development of the PE market.
3. Explain the structure of PE.
I n s t i t u t i o n a l I n v e s t o r s a n d P E 327

4. Outline the terms of the covenants that establish the contract between institutional
investors and PE.
5. Distinguish between the behavior of individual and institutional investors in PE.
6. Explain how institutional investors determine their choices in PE.

References
Axelson, Ulf, Per Strmberg, and Michael S. Weisbach. 2009. Why Are Buyouts Levered? Financial
Structure of Private Equity Funds. Journal of Finance 64:4, 15491582.
Bailey, Bronwyn. 2013. Long Term Commitments: The Interdependence of Pension Security and
Private Equity. Private Capital Growth Capital Council White Paper. Available at http://
www.pegcc.org/wordpress/wp-content/uploads/Long-Term-Commitments-The-
Interdependence-of-Pension-Security-and-Private-Equity.pdf.
Barber, Brad, and Terrance Odean. 2008. All That Glitters: The Effect of Attention and News on the
Buying Behavior of Individual and Institutional Investors. Review of Financial Studies 21:2,
785818.
Bennedsen, Morten, and Daniel Wolfenzon. 2000. The Balance of Power in Closely Held Corpora-
tions. Journal of Financial Economics 58:1 113139.
Blume, Marshall, and Donald B. Keim. 2012. Institutional Investors and Stock Market Liquidity:
Trends and Relationships. Working Paper, The Wharton School, University of Pennsylvania.
Bonafede, Julia K., Steven J. Foresti, and Russell J. Walker. 2012. 2012 Report on State Retirement
Systems: Funding Levels and Asset Allocation. Wilshire Consulting.
Braun, Martin. 2011. States Miss Pension Targets by 50% Even with Private Equity. Bloomberg,
July 26. Available at http://www.bloomberg.com/news/2011-2007-26/states-miss-pension-
targets-by-50-with-private-equity-proving-not-enough.html.
Bruce, Brian. 2010. Handbook of Behavioral Finance. Northampton, MA: Edward Elgar Publishing.
Committee on the Global Financial System. 2011. Fixed Income Strategies of Insurance Compa-
nies and Pension Funds. Available at http://www.bis.org/publ/cgfs44.pdf.
The Conference Board. 2010. Institutional Investment Report: Trends in Asset Allocation and
Portfolio Composition. Conference Board Report, November 11. Available at http://papers.
ssrn.com/sol3/papers.cfm?abstract_id=1707512.
Cornelli, Francesca, and David Goldreich. 2001. Bookbuilding and Strategic Allocation. Journal of
Finance 56:6, 23372369.
Cumming, Douglas. 2010. Private Equity: Fund Types, Risks and Returns, and Regulation. Hoboken,
NJ: John Wiley & Sons.
Cumming, Douglas, Grant Fleming, and Sofia A. Johan. 2011. Institutional Investment in Listed
Private Equity. European Financial Management 17:3, 594618.
Cumming, Douglas, Grant Fleming, and Armin Schwienbacher. 2009. Style Drift in Private
Equity. Journal of Business Finance & Accounting 36:5/6, 645678.
Cumming, Douglas, and Sofia Johan. 2009. Venture Capital and Private Equity Contracting: An Inter-
national Perspective. Burlington, MA: Academic Press.
de Zwart, Gerben, Brian Frieser, and Dick van Dijk. 2012. Private Equity Recommitment Strate-
gies for Institutional Investors. Financial Analysts Journal 68:3, 8199.
Easterbrook, Frank, and Daniel R. Fischel. 1986. Close Corporations and Agency Costs. Stanford
Law Review 38:2, 271301.
Ewens, Michael, Charles M. Jones, and Matthew Rhodes-Kropf. 2013. The Price of Diversifiable
Risk in Venture Capital and Private Equity. Review of Financial Studies 26:8, 18531889.
Freiburg, Markus, and Dietmar Grichnik. 2012. Institutional Investments in Private Equity Funds:
Social Ties and the Reduction of Information Asymmetry. Venture Capital 14:1, 126.
Giannetti, Mariassunta, and Andrei Simonov. 2006. Which Investors Fear Expropriation? Evi-
dence from Investors Portfolio Choices. Journal of Finance 61:3, 15061547.
328 p r i v a t e e q u i t y : u s e s a n d s t r u c t u r e

Gompers, Paul, Anna Kovner, Josh Lerner, and David Scharfstein. 2008. Venture Capital Invest-
ment Cycles: The Impact of Public Markets. Journal of Financial Economics 87:1, 123.
Gompers, Paul, and Josh Lerner. 1996. The Use of Covenants: An Empirical Analysis of Venture
Partnership Agreements. Journal of Law and Economics 39:2, 463498.
Gompers, Paul, and Josh Lerner. 1999. An Analysis of Compensation in the U.S. Venture Capital
Partnership. Journal of Financial Economics 51:1, 344.
Gompers, Paul, and Josh Lerner. 2000. Money Chasing Deals? The Impact of Fund Inflows on Pri-
vate Equity Valuations. Journal of Financial Economics 55:2, 281325.
Grinblatt, Mark, Sheridan Titman, and Russ Wermers. 1995. Momentum Investment Strategies,
Portfolio Performance, and Herding: A Study of Mutual Fund Behavior. American Economic
Review 85:5, 10881105.
Gutierrez, Roberto C., Jr., and Christo A. Pirinsky. 2007. Momentum, Reversal, and the Trading
Behaviors of Institutions. Journal of Financial Markets 10:1, 4875.
Harris, Robert S., Tim Jenkinson, and Steven N. Kaplan. 2014. Private Equity Performance: What
Do We Know? Journal of Finance, forthcoming.
Hobohm, Daniel. 2010. Investors in Private Equity. Wiesbaden, Germany: Gabler|GWV Fachverlage
GmbH.
Hochberg, Yael V., Alexander Ljungqvist, and Yang Lu. 2007. Whom You Know Matters: Venture
Capital Networks and Investment Performance. Journal of Finance 62:1, 251301.
Huyghebaert, Nancy. 2010. Institutional Affiliation of General Partners and Private Equity Invest-
ment Choices. Journal of Private Equity 13:4, 2541.
Investment Company Institute. 2012. Investment Company Fact Book. Washington, DC: Investment
Company Institute.
Investopedia.com. 2014. Committed Capital Definition. Available at http://www.investopedia.
com/terms/c/committedcapital.asp.
Jegadeesh, Narasimhan, Roman Kraussl, and Joshua M. Pollet. 2009. Risk and Expected Returns of
Private Equity Investments: Evidence Based on Market Prices. Available at http://ssrn.com/
abstract=1364776.
Jensen, Michael, and William H. Meckling. 1976. Theory of the Firm: Managerial Behavior, Agency
Costs and Ownership Structure. Journal of Financial Economics 3:4, 303360.
Kaplan, Steven N., and Antoinette Schoar. 2005. Private Equity Performance: Returns, Persistence,
and Capital Flows. Journal of Finance 60:4, 17911823.
Kreutzer, Laura, and Hillary Canada. 2014. U.S. Firms Have Bumper 2013 Fundraising Crop. Wall
Street Journal. Available at http://blogs.wsj.com/privateequity/2014/01/13/u-s-firms-have-
bumper-2013-fundraising-crop/.
Lakonishok, Josef, Andrei Shleifer, and Robert Vishny. 1992. The Impact of Institutional Trading
on Stock Prices. Journal of Financial Economics 32:1, 2343.
Lerner, Josh. 1994. Venture Capitalists and the Decision to Go Public. Journal of Financial Econom-
ics 35:3, 293316.
Lerner, Josh, Antoinette Schoar, and Wan Wongsunwai. 2007. Smart Institutions, Foolish Choices?
The Limited Partner Performance Puzzle. Journal of Finance 62:2, 731764.
Liles, Patrick R. 1977. Sustaining the Venture Capital Firm. Cambridge, MA: Management Analysis
Center.
Ljungqvist, Alexander, Felicia Marston, and William J. Wilhelm. 2009. Scaling the Hierarchy: How
and Why Investment Banks Compete for Syndicate Co-management Appointments. Review
of Financial Studies 22:10, 39774007.
Ljungqvist, Alexander, and Matthew Richardson. 2003. The Cash Flow, Return and Risk Charac-
teristics of Private Equity. NBER Working Paper No. 9454. Available at http://ssrn.com/
abstract=369600.
Mayer, Colin, Koen Schoors, and Yishay Yafeh. 2005. Sources of Funds and Investment Activities
of Venture Capital Funds: Evidence from Germany, Israel, Japan and the United Kingdom.
Journal of Corporate Finance 11:3, 586608.
Mergenthaler, Karl, and Chad Moten. 2008. Private Equity for Institutional Investors Current En-
vironment and Trends. J. P. Morgan Investment Analytics and Consulting, 0.3.
I n s t i t u t i o n a l I n v e s t o r s a n d P E 329

Moskowitz, Tobias, and Annette Vissing-Jrgensen. 2002. The Returns to Entrepreneurial Invest-
ment: A Private Equity Premium Puzzle? American Economic Review 92:4, 745778.
Nahapiet, Janine, and Sumantra Ghoshal. 1998. Social Capital, Intellectual Capital, and the Organ-
izational Advantage. Academy of Management Review 23:2, 242266.
Nielsen, Kasper M. 2008. Institutional Investors and Private Equity. Review of Finance 12:1,
185219.
P&I/Towers and Watson. 2013. The Worlds 500 Largest Asset ManagersYear End 2012.
Towers Watson. Available at http://www.towerswatson.com/en-US/Insights/IC-Types/
Survey-Research-Results/2013/11/The-Worlds-500-Largest-Asset-Managers-Year-
end-2012.
Pagano, Marco, and Ailsa Roell. 1998. The Choice of Stock Ownership Structure: Agency Costs,
Monitoring and the Decision to Go Public. Quarterly Journal of Economics 113:1, 187225.
Patterson, Douglas, and Vivek Sharma. 2006. Intraday Trading: Herding versus Market Efficiency.
Working Paper, Virginia Polytechnic Institute and State University.
Phalippou, Ludovic, and Oliver Gottschalg. 2009. The Performance of Private Equity Funds.
Review of Financial Studies 22:4, 17471776.
Phalippou. Ludovic, and Maurizio Zollo. 2005. What Drives Private Equity Fund Performance?
Working Paper, University of Amsterdam and INSEAD. Available at http://fic.wharton.
upenn.edu/fic/papers/05/0541.pdf.
PitchBook. 2013. The 2013 Annual Private Equity Breakdown Report. Available at http://
pitchbook.com/The_Annual_PE_Breakdown_2013.html.
Practical Law. 2014. Choosing an Entity Comparison Chart. Available at http://us.practicallaw.
com/7-381-0701.
Sahlman, William. 1990. The Structure and Governance of Venture-Capital Organizations. Journal
of Financial Economics 27:2, 473521.
Sharpe, William F. 2010. Adaptive Asset Allocation Policies. Financial Analysts Journal 66:3,
4559.
Shleifer, Andrei, and .Robert W. Vishny. 1997. A Survey of Corporate Governance. Journal of Fi-
nance 52:2, 737783.
Sias, Richard. 2004. Institutional Herding. Review of Financial Studies 17:1, 165206.
Solomon, Steven. 2011. Wall St.s Odd Couple and Their Quest to Unlock Riches. New York Times
DealBook, December 13. Available at http://dealbook.nytimes.com/2011/12/13/wall-st-s-
odd-couple-and-their-quest-to-unlock-riches/.
Stowell, David. 2013. Investment Banks, Hedge Funds, and Private Equity. Waltham, MA: Academic
Press.
Takahashi, Dean, and Seth Alexander. 2002. Illiquid Alternative Asset Fund Modeling. Journal of
Portfolio Management 28:2, 90100.
Uzzi, Brian. 1996. The Sources and Consequences of Embeddedness for the Economic Perfor-
mance of Organizations: The Network Effect. American Sociological Review 61:4, 674698.
Wermers, Russ. 2002. Mutual Fund Herding and the Impact on Stock Prices. Journal of Finance
54:2, 581622.
19
Private Equity and Value Creation
PTER HARBULA
Executive Director Corporate Finance, Edenred Group; Lecturer, HEC Paris Business School
The views expressed herein are only those of its author

Introduction
Private companies backed by private equity (PE) funds, just as PE companies, are major
players in many economies. PE or financial buyer sponsored transactions represent an
increasingly important portion of the global mergers and acquisitions (M&A) market
since 1990. The total value of PE-backed deals accounts for about 10 percent of the
cumulated market capitalization of the publicly traded companies in most European
countries (Wright, Renneboog, Simons, and Scholes 2006). Financial buyers represent
a growing force in the market for corporate control. This presence is crucial because the
market for corporate control is an important mechanism of modern capital markets.
The growth in the European market was above the worldwide average: the dollar value
of transactions grew by an average of 40 percent between 2003 and 2007, but declined
thereafter (Botazzi 2010).
Leverage is a critical item in the success of key players in the PE sector. In a leveraged
buyout (LBO), a specialized investment firm acquires a company financing the transac-
tion by using debt (also referred to as the LBO model). Leverage is an important tool for
optimizing the managements performance from the perspective of a firms stakehold-
ers. Jensen (1986) introduces the free cash flow (FCF) theory, which is a much debated
topic in corporate finance. He finds that managers reinvested FCF in poor investment
opportunities in the oil industry instead of returning it to investors. His theory suggests
that debt could be used to prevent managers from squandering resources. Since, many
academics consider debt as a tool to increase the quality of corporate governance. The
interaction between leverage and the quality of corporate governance hence affects the
success of LBO funds.
The success of the LBO model has been instrumental in deals in which value results
from improvements in cash-flow management, bargain purchases, and turnaround situ-
ations. At the same time, by reducing agency costs and improving efficiency, LBOs have
successfully engineered transactions to deliver good returns to their investors (Kaplan
and Schoar 2005; Phalippou 2012). The question is did anything go wrong in recent
years? Using data from different databases and contributing PE houses in Europe, a

330
Pr iv ate E qu it y an d Val u e C re at ion 331

database for continental Europe transactions between 1997 and 2013 is constructed to
analyze the impact of financial leverage on LBO funds in terms of asset prices and value
creation mechanisms as well as corporate governance mechanism in general. Leverage is
the cornerstone of the LBO model and the instrument responsible for previous success.
Was the mechanism inherent to the success of the business model instrumental in fuel-
ing the financial crisis of 20072008 even if the impact was temporary?
If one of the main sources of value creation in LBO transactions is the capacity to
identify assets offering a good potential for cash-flow growth and improvement, the
interaction between leverage and the effectiveness of the corporate governance poli-
cies applied must also be analyzed. During this analysis, consider that the leverage ratio
used for LBO deals is rising and a decrease in cash-flow improvements exists. Originally,
these trends were masked by favorable M&A market trends and the use of cheap lever-
age that led both managers and investors to believe that returns were improving. How-
ever, this situation resulted in LBO funds performance failing to outperform market
indices once the effect of increasing market prices of assets was neutralized.
The purpose of this chapter is to analyze the interaction between value creation of
LBO transactions and the use of leverage. The remainder of the chapter has the follow-
ing organization. The next section reviews the history of LBOs in Europe, followed by
an overview of the modus operandi of LBO funds and value creation levers in LBO
transactions. Next, a section is dedicated to analyzing the data panel used for the anal-
ysis and market data. The following sections discuss the analysis of the interaction be-
tween value creation and leverage and the impact that LBO funds have on the operating
performance of their target companies. The chapter closes with summary remarks.

History of Leveraged Buyouts in Europe


PE and venture capital (VC) can be traced to the 1950s and 1960s in the United States.
Highly leveraged transactions arose in the 1970s by Bear Sterns bankers Jerome Kohl-
berg, Henry Kravis, and George Roberts. They engineered several buyouts of compa-
nies in which the founders searched for an exit solution. In Europe, LBOs emerged in
the 1980s. Initially, many highly leveraged transactions focused on saving distressed
companies after the recession that hit the United Kingdom between 1979 and 1981.
At the end of the 1980s, a second wave of development began driven mainly by man-
agement buyouts (MBOs) sponsored by financial partners. A third wave of develop-
ment began in 1994 coinciding with the development of the European M&A market
supported by strong economic growth. This wave lasted until the burst of the dot-com
bubble in 2000. In 1997, about 700 PE-led LBOs occurred in the United Kingdom
(Wright et al. 2006; Botazzi 2010).
Compared to the United States, M&A and LBO markets were far less developed in
continental Europe and took a longer time to mature. In 1996, only 200 known LBOs
were registered (mainly in France and the Netherlands), which was one-third of the
number in the United Kingdom. In 1998 and 1999, about 250 LBOs occurred in conti-
nental Europe, one-quarter of that in the United Kingdom. The substantial increase in
LBOs began in the early years of the new millennium when publicly listed corporations
started disposing of non-strategic assets to ease their debt burden built up in the huge
332 p r i vat e e q u i t y : u s e s a n d s t r u c t u r e

M&A wave between 1997 and 2000. In 2006, experiencing rapid growth and success
and in a favorable market environment with accessible credit markets, about 40 per-
cent of the European LBO deals took place in continental Europe (Wright et al. 2006;
Botazzi 2010).
Although transactions in the United Kingdom equaled, in terms of the aggregate
value of U.S. dollars spent to those occurring in continental Europe in 2000, the latter
represents three times the value of the former in 2005. However, in the United King-
dom, the value of LBOs amounted to 1.5 percent of the U.K.s gross domestic product
(GDP) while it represented, for example, only 0.75 percent and 0.5 percent of GDP in
France and Germany, respectively.
Most continental European countries began to catch up in subsequent years (Ren-
neboog and Simons 2005). Since 2004, an unparalleled growth occurred in both the
number of transactions and players in the market (Wright et al. 2006). Driven by fac-
tors such as historically low interest rates, low inflation, worldwide GDP growth, strong
inflows of liquidity on the money and debt markets, and record levels of fundraising in
PE funds; LBO deals flourished in continental Europe. Between 2003 and 2007, LBO-
sponsored transactions accounted for 17 percent of the aggregated M&A deal value.
PE-sponsored transactions reached a record-setting level between 2005 and 2007 with
LBOs accounting for more than 20 percent of the total amount invested between 2000
and 2010 (Botazzi 2010).
LBO fund-sponsored transactions also played an important role in keeping the
M&A market afloat in crisis times such as between 2001 and 2004, as long as funding is
available for applying the appropriate financial leverage. These periods offered opportu-
nities for good deals and bargain purchases with targets having sound fundamentals but
with the ultimate shareholders being under pressure (e.g., core business restructuring).
In 2008 and 2009, PE transactions fell dramatically in both volume and value com-
pared to their peak in 2007. Between 2010 and 2012, a modest recovery occurred de-
spite lagging far behind 2006 and 2007. Deal volumes fell slightly for acquisitions while
aggregate deal values remained increased significantly in 2013. At the time of writing
this chapter, preliminary results for 2014 for Europe indicate that deal volumes fell by
17 percent for the first time since 2009, while aggregate deal value remained steady. For
2015, various sources anticipate a rising trend in both deal amount and aggregate value,
but the extent of the recovery still remains unclear.

How Leveraged Buyout Funds Operate


LBO funds enter deals with a limited time horizon, typically between 3 and 10 years.
Investments are financed by a mix between the equity portion provided by the LBO
fund and financial debt (e.g., senior debt, junior debt, mezzanine debt, and convertible
debt) issued by banks and institutional investors. The equity portion is drawn from the
money the LBO fund raised with its investors.
Besides the initial financial valuation, another factor that determines the viability
of an investment is if leverage is maximized to the point at which the target can safely
reimburse the debt. This qualification is because the higher the proportion of debt in
the financing mix of the transaction, the higher is the internal rate of return (IRR) on
Pr iv ate E qu it y an d Val u e C re at ion 333

the equity portion of investment. The IRR relies on the entry value, represented by the
equity portion of the investment, the expected or realized proceeds to the equity pro-
viders from the exit price of the asset, and the flows to equity, if any, between the entry
and the exit. Thus, as opposed to classic corporations that are primarily concerned
with value creation for the stakeholders over the long term, LBO funds already have an
exit strategy in mind at the investment date.
LBO funds are particularly interested in understanding the fundamentals that can
increase the IRR. Such factors include the following:

the capacity to increase its cash-flow generation during the holding period;
optimization possibilities of the FCF yield (e.g., enhancing working capital manage-
ment, monitoring investment policy, and optimizing tax expenses);
the ability to reduce the buyout indebtedness through the holding period by assign-
ing cash flow in priority to reimburse the buyout financial debt;
the valuation potential at the exit date; and
IRR optimization using advanced financial instruments such as options and war-
rants.

General partners (GPs) of LBO funds usually have a remuneration based on the assets
under management but also share an incentive fee based on value creation if the IRR
exceeds a certain threshold known as carried interest (Wright et al. 2006). This fee struc-
ture helps to align incentives between the interests of shareholders and the GPs. PE or
LBO funds also earn remuneration on a fixed percentage of the invested and managed
funds. Some view this mechanism as more controversial in terms of aligning interests
between the principals (i.e., investors in the PE funds) and the agents (i.e., GPs manag-
ing the fund).

Leveraged Buyouts and Value Creation


In two seminal articles, Jensen (1986, 1989) argues that LBOs create value through
high leverage and related powerful incentives. In contrast, corporations are often char-
acterized by a non-shareholder management pursuing their own interests, prone to cash
flow diversion, and averse to taking on sufficient levels of risk. According to Jensens
FCF theory, increased monitoring and management ownership provide incentives for
managers to improve operating performance and generate cash flows. Also, high debt
limits a managers ability to squander free cash on wasteful investments.
Consistent with Jensens view, Kaplan (1989) provides evidence that LBOs, and
thus debt, create value by substantially improving the operating performance of ac-
quired companies and by redirecting cash in the form of high debt payments instead
of other uses. Conversely, Renneboog, Simons, and Wright (2007) do not find support
for the FCF argument as a driver of public to private transactions. While evidence of
gains in operating performance exists, less research concerning the increase in value of
buyout targets has been conducted.
Evidence relating to the performance of exited buyouts seems limited relative to
those that have exited by IPOs in terms of accounting and stock market returns (Lehn
334 p r i vat e e q u i t y : u s e s a n d s t r u c t u r e

and Poulsen 1989; Bruton, Keels, and Scifre 2002; Jelic, Saadouni, and Wright 2005).
Such studies offer mixed support for Jensens FCF theory suggesting that increased lev-
erage and realignment of incentives have positive effects on the operating performance
of LBO companies. Wright, Wilson, and Robbie (1995) show that the size of a buyout
is positively related to a short holding period and a high exit probability. Kaplan (1991)
finds similar evidence, but his study focuses only on larger buyouts.
Studies on buyout returns focus either on limited samples relating to IPOs among
the largest transactions, rather than across the buyout market as a whole (Kaplan 1989)
or on measures of the funds performance (Kaplan and Schoar 2005). Nikoskelainen
and Wright (2006) conduct a study on a U.K.-based sample that analyzes the factors af-
fecting realized buyout returns for target companies. Testing the FCF theory, they show
that value increases and above average returns of LBOs are related to the corporate gov-
ernance mechanisms resulting from LBOs, especially managerial equity holdings. They
also report that financial leverage is not linearly correlated with value creation in LBO
funds: after a threshold, the IRR falls below the sample average. Guo, Hotchkiss, and
Song (2011) identify three main sources of value creation: (1) financial leverage, which
on average results in a tripling of pre-transaction values; (2) managerial control, espe-
cially when the chairman of the board is not the same person as the chief executive
officer (CEO); and (3) enhanced cash-flow generation from improved profitability or
lower working capital levels.
Based on research evidence, the following five actions contribute to creating value in
leveraged transactions. First, agency costs between management and shareholders should
be reduced. This reduction can occur through several means such as through the discipline
effect of leverage. Jensens (1986) FCF theory shows that an increase in indebtedness re-
duces the available cash for management and forces managers to pursue value-increasing
investments, which triggers a restructuring of the company because of rising bankruptcy
costs. A reduction in agency costs can also occur by increasing control and monitoring in
terms of financial incentives and controlling of expenses and investments. Further, agency
costs can be reduced by motivating management through ownership. As in most LBO
deals, management is interested in the LBOs success ( Jensen and Meckling 1976).
Second, value creation can occur by reducing agency costs between creditors and
shareholders using covenants on financial indebtedness and strip financing, which is a
technique allowing investors to hold senior debt, convertible debt, preferred and ordi-
nary shares, and different seniority level between creditors. Third, selecting undervalued
acquisition targets, especially in public to private deals, can lead to higher value. Fourth,
PE funds can benefit from the tax shield on debt financing. Finally, value creation can
result from increasing operating efficiency with improved working capital management,
investment monitoring, and other techniques to maximize the cash-flow yield available
to repay the senior debt (Kaplan 1989; Acharya, Kehoe, and Reyner 2009).

Observations on Leveraged Buyout Transaction Data


For the analysis of LBOs and their financial characteristics, this study includes hand-
collected data from European PE-sponsored transactions between 1997 and 2013 and
compares investment returns, realized (investment), IRR, and value creation using
Pr iv ate E qu it y an d Val u e C re at ion 335

cross-sectional regression analysis. The data set uses data available from different Euro-
pean databases, publicly available data, and 21 European PE houses that kindly shared
information of actual, realized, and exited transactions. Data are cross-checked and har-
monized for each transaction reference when multiple sources are available. Harmoni-
zation of definitions occurred between the different sources to produce a homogeneous
database. The data sample also used publicly available data for reversed LBOs (RLBO),
which refers to an LBO exiting through an IPO.
Sample partitioning occurs based on the following exit types: (1) an IPO on a public
market; (2) a sale to a strategic (corporate) buyer, secondary or tertiary LBO; (3) a
bankruptcy; and (4) an acquisition by a turnaround fund if the lenders take control of
the target in a debt workout scheme. On a limited number of transactions (when infor-
mation was available or was contributed by the PE houses), the sample included CEO
and CFO turnover post-acquisition information on governance specific key metrics,
such as board size and composition (i.e., internal versus external versus independent
members).
In terms of key metrics and comparative statistics, the sample of transactions initi-
ated by strategic buyers was created to mirror the LBO deals. Size, industries, trans-
actions dates, and geographic locations are matched as closely as possible to the
LBO-sponsored transactions. The benchmark is based on public market data. For the
entire observation period, a benchmark index based upon public market companies was
constructed with the aim to screen out and understand the impact of changing market
prices of companies (upward or downward) on the realized IRR of the LBO fund. The
benchmark index consists of 500 European companies for each year, which are kept
the same over the whole period except for the effect of mergers, public offerings, and
delistings. The analysis includes a total of 687 companies. Figure 19.1 provides an over-
view of the key data statistics of the sample. Only exited deals are necessary to perform a
complete value creation and IRR analysis, thus, the last vintage (entry) year considered
was 2010.
Deals are distributed over the entire period, although some concentration exists
within the PE boom period between 2005 and the first half of 2008, in terms of entry
transactions as well as in terms of exits. Deals in the sample have relatively high IRRs,
but exhibit outliers especially for the different sub-samples. IRR data are below the av-
erage values reported in many studies, but above the level observed in studies that were
the most vocal in criticizing the performance of LBO funds (Kaplan and Schoar 2005;
Phalippou and Gottshalg 2009).
Table 19.1 shows that the average returns exhibited by LBO-sponsored transac-
tions are not very different from the returns of the public company benchmark index
on an enterprise value based on IRR. Observing the stock market pricing-adjusted IRR
figures (Adjusted IRR) based upon enterprise value, the results indicate that the index
based on stock market data outperforms the LBO data sample. Analyzing the different
subsets shows that LBO fund returns outperform the stock market index during both
the 2001 to 2004 and 2008 to 2012 subperiods. They are comparable to the subperiod
between 1997 and 2000. They fail to outperform in the 2005 to 2007 subperiod and
during 2013. The mean entry multiple is below the mean exit multiple. This is consistent
with findings of Kaplan (1989), Nikoskelainen and Wright (2007), and Acharya, Hahn,
and Kehoe (2008).
Panel A
12
9.6 10.5
EV / EBITDA multiple 9.0 8.68.9
9.2 9.9 8.7
9.0 9.0
9.5
8.5 8.6 8.5
7.3 7.9 8.0 8.0
8 6.9 7.2 7.0 7.0 7.3 7.0
6.3 5.9 6.7 6.5

0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
Strategic buyers LBO initiated deals

Panel B
8
6.5
6.0
Debt / EBITDA

5.1 5.5
5.2
4.7 4.5 4.4
4.0 4.0 3.9 4.0 4.0 3.8 2.9 3.3 4.0
4 3.5 3.4 3.5 3.5
3.0 3.2 3.2 3.2 3.4
2.5 2.5

0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
Strategic buyers LBO initiated deals

Panel C
12x
Average transaction multiple

2007
10x
(EV to EBITDA)

2006
2005
2008
8x 2010 2004 2013
2011 2012 1999
2009 1998 1997
2003 2000
2001
6x 2002

4x
2x 3x 4x 5x 6x 7x 8x
Financial leverage (Debt to EBITDA ratio)

Figure 19.1 An Overview of the Key Data Statistics of the Sample Panel A.
European Transaction Valuation between 2000 and 2013 This figure shows the average
valuation of European transactions (both LBO and strategic deals) between 2000 and
2013. Panel B. European Transaction Leverage Level between 2000 and 2013This
figure shows the average leverage level of European transactions (both LBO and strategic
deals) between 2000 and 2013. Panel C. European LBO Transaction Financial
Leverage Level Plotted against Valuation Level between 2000 and 2013 This figure
shows the relationship between the average leverage level and valuation multiples paid
for European LBO transactions between 2000 and 2013. Source: Thomson Financial,
Dealogic. Panel D. European Strategic Transaction Financial Leverage Level Plotted
against Valuation Level between 2000 and 2013 This figure shows the relationship
between the average leverage level and valuation multiples paid for European strategic
transactions between 2000 and 2013. Source: Thomson Financial, Dealogic.
Pr iv ate E qu it y an d Val u e C re at ion 337

Panel D
Average transaction multiple 12

10 2007
(EV to EBITDA)

2012 2013
2004 2006
1999
2010 2005 2000
1997 1998
8 2009
2003 2011 2001
2002

4
2 3 4 5 6
Financial leverage (Debt to EBITDA ratio)

Figure 19.1 Continued

In terms of the valuation multiple (EV/EBITDA), strategic transactions occur at an


average multiple that is higher than the average multiple of LBO transactions. However,
the average level of valuation multiples may not be necessarily meaningful in the case
of such a diversified and large sample. Strategic exit LBO transactions seem to offer a
higher relative valuation as opposed to secondary or tertiary LBO and IPO exited trans-
actions. As expected, the public sector benchmark index shows that public companies
are less indebted than those that are a part of an LBO. Similarly, industrial buyers use
debt less aggressively to finance acquisitions. Because information on financing is una-
vailable for all strategic transactions, excerpting and extrapolating data from the acquir-
ing companys financial statements is necessary in some cases.
Table 19.2 shows substantial variation in the financial leverage observed in LBO
transactions across different time periods. These discrepancies and especially the in-
crease of leverage observed in the more recent time periods seem to support the thesis
of the inflationary bubble on asset prices driven by cheap debt (Altman 2007), as it
seems linked with increasing transaction multiples. The Debt/EBITDA ratio is used
to measure the indebtedness in LBO deals because the assumed level of indebtedness
is closely related to the market value of assets. EBITDA seems to be a better proxy for
cash flows in this respect than equity because it corresponds to the metric that is most
commonly used for simplification purposes in the LBO industry. Management stake
increased over the observed period. While divestments are fairly stable over the time
period of the study, build-up activity increased in the more recent years. Furthermore,
between 2005 and 2007, above average deal leverage occurred coupled with below av-
erage IRR measures. Note that enterprise value is based on adjusted stock market index
values. At the same time, a gradual reduction occurs in the average holding period of
LBO funds over the observed period, bottoming out in the 2005 to 2007 period to just
above three years on average, while the full sample average is closer to four years.
In terms of motivational scheme, although the description is incomplete and quan-
titative, LBO funds seem to place more weight on aligning shareholders and manag-
ers interests than public corporations. This observation proves particularly true when
comparing the average value of the package offers to managers. The common tool used
Table 19.1 Data Panel and Descriptive Key Statistics
LBO Sponsored Deals Strategic Deals
Whole Sample IPO Exit Strategic Buy Exit Secondary / Tertiary Bankruptcy /
LBO Exit Turnaround Takeover
Average Median Average Median Average Median Average Median Average Median Average Median
Observations 1052 135 527 334 56 1532
Average deal size 251 256 444 456 231 221 211 237 203 209 265 289
(entry)
Min / Max / Std Dev. 52 / 1056 / 321 124 / 1056 / 189 52 / 925 / 299 78 / 887 / 301 66 / 642 / 144 51 / 1021 / 217
Average deal size 512 537.6 675 785 508.2 521 498 503 45 125 n/a
(exit)
Min / Max / Std Dev. 0 / 1536 / 451 105 / 1523 / 301 21 / 1536 / 405 61 / 997 / 351 0 / 105 / 25 n/a
Holding time (years) 3.8 3.6 4.2 4.4 4.0 3.9 3.3 2.9 3.8 3.6 n/a
Min / Max / Std Dev. 1.4 / 7.6 / 1.6 2.4 / 7.6 / 2.0 1.5 / 5.5 / 2.1 1.3 / 4.0 / 1.5 1.5 / 4.3 / 1.8 n/a
IRR % (Enterprise 15.3% 10.6% 12.5% 8.1% 26.1% 20.0% 3.3% 0.9% 21.1% 28.3% n/a
Value)
Min / Max / Std Dev. 78% / 106% / 55.9% 8% / 56% / 22.3% 1% / 106% / 41.3% 6% / 77% / 33.5% 78% / 19% / 15.2% n/a
IRR% (Equity Value) 30.1% 27.8% 33.5% 27.1% 37.5% 34.8% 21.3% 21.3% 12.0% 13.2% n/a
Min / Max / Std Dev. 52% / 152% / 45.1% 8% / 77% / 32.1% 4% / 152% / 51.1% 7% / 87% / 47.5% 52% / 8% / 11.3% n/a
Index Adj. IRR % 2.1% 1.0% 3.1% 1.3% 12.0% 9.1% 12.1% 15.1% 13.0% 15.2% n/a
(EV)
Min / Max / Std Dev. 89% / 89% / 33.2% 12% / 75% / 31.0% 4% / 89% / 43.3% 13% / 48% / 63.2% 89% / 29% / 12.5% n/a
Table 19.1continued

LBO Sponsored Deals Strategic Deals


Whole Sample IPO Exit Strategic Buy Exit Secondary / Tertiary Bankruptcy /
LBO Exit Turnaround Takeover
Average Median Average Median Average Median Average Median Average Median Average Median
Index Adj. IRR % 18.5% 12.2% 19.3% 14.3% 25.3% 19.0% 10.5% 3.3% 11.0% 13.2% n/a
(Equ. V)
Min / Max / Std Dev. 72% / 78% / 36.2% 1% / 78% / 36.2% 6% / 57% / 29.6% 9% / 65% / 28.6% 72% / 13% / 12.2% n/a
Debt / EBITDA 4.5x 4.5x 4.4x 4.5x 5.0x 4.7x 4.2x 4.2x 4.2x 4.2x 3.3x 3.4x
(entry)
Min / Max / Std Dev. 1.0x / 18.0x / 4.0x 1.3x / 7.2x / 4.1x 1.7x / 8.0x / 4.3x 1.8x / 6.3x / 4.0x 3.2x / 18.0x / 3.5x 0.0x / 8.0x/1.0x
Debt / EBITDA (exit) 3.4x 3.3x 3.3x 3.0x 3.5x 3.2x 3.5x 3.6x 4.0x 4.2x n/a
Min / Max / Std Dev. 0.0x / 19.1x / 2.1x 0.0x / 4.3x / 1.7x 0.2x / 6.1x / 2.9x 0.3x / 4.3x / 3.2x 6.3x / 19.1x / 3.6x n/a
Entry multiple 7.8x 7.8x 7.7x 7.9x 7.6x 8.1x 8.1x 7.3x 7.9x 8.0x 8.5x 8.6x
Min / Max / Std Dev. 1.2x / 25.6x / 3.9x 1.3x / 15.8x / 4.2x 1.2x / 25.6x / 4.1x 1.7x / 15.1x / 4.3x 1.5x / 18.5x / 3.9x 2.0x / 33.2.0x
/ 6.9
Exit multiple 8.5x 8.3x 8.5x 8.2x 8.6x 9.0x 9.0x 8.0x 2.0x 1.7x n/a
Min / Max / Std Dev. 0.0x / 33.2x / 4.3x 5.2x / 14.3x / 5.1x 6.1x / 33.2x / 3.8x 4.8x / 18.5x / 6.1x neg. / 2.2x / 0.9x n/a
Management n/a
905 121 454 302 28
investment
Management stake 11.3% 10.3% 11.8% 10.7% 11.2% 10.7% 11.5% 9.8% 9.1% 9.0% 5.1% 5.2%
Min / Max / Std Dev. 0.5% / 25.3% / 5.2% 0.5% / 25.3% / 5.1% 1.8% / 20.3% / 5.0% 4.5% / 20.2% / 4.8% 0.9% / 19.9% / 3.9% 0.0% / 10.1% /
3.7%
continued
Table 19.1continued

LBO Sponsored Deals Strategic Deals


Whole Sample IPO Exit Strategic Buy Exit Secondary / Tertiary Bankruptcy /
LBO Exit Turnaround Takeover
Average Median Average Median Average Median Average Median Average Median Average Median
Divestment 225 65 63 95 2 n/a
Build-up 178 65 9 101 3 n/a

No of deals 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Total (1)
Entry 51 65 102 95 81 72 66 45 87 89 78 31 25 41 35 34 45 991
Exit 45 50 65 82 77 60 54 55 118 107 169 99 33 31 17 18 17 1052

Note: This table shows the descriptive statistics of the PE data sample for the whole and the sample and by the type of transaction as well as the strategic deal sample that serves as the
benchmark and the reference for each analysis.
(1) 67 deals with an entry date before 1997.
Table 19.2Data Panel Key Statistics
Panel A
LBO Sponsored Deals
Whole Sample 19972000 20012004; 20082012 20052007; 2013
Average Median Average Median Average Median Average Median
Observations 1052 242 310 462
Average Deal Size (Entry) 251 256 239 233 208 203 292 312
Min / Max / Std Dev. 52/ 1056 / 321 61 / 898 / 201 52 / 733 / 177 88 / 1056 / 203
Average deal size (Exit) 512 537.6 501 533 399 415 602 630.4
Min / Max / Std dev. 0 / 1536 / 451 89 / 1023 / 287 0 / 1536 / 423 101 / 1416 / 525
341

Holding time (Years) 3.8 3.6 4.1 4.1 4.2 3.9 3.4 3.1
Min / Max / Std dev. 1.4 / 7.6 / 1.6 1.4 / 6.7 /1.8 1.5/ 7.2 / 2.3 1.4/ 7.6 / 1.9
IRR % (Enterprise Value) 15.3% 10.6% 13.3% 8.7% 17.9% 13.1% 14.7% 10.1%
Min / Max / Std dev. 78% / 106% / 55.9% 65% / 78% / 25.3% 71% / 89% / 27.3% 78% /106% / 26.3%
IRR% (Equity Value) 30.1% 27.8% 29.6% 26.4% 31.8% 28.6% 29.2% 28.1%
Min / Max / Std dev. 52% / 152% / 45.1% 23% / 99% / 29.3% 25% /101% / 31.1% 52% / 152% / 32.1%
Index Adj. IRR % (EV) 2.1% 1.0% 1.0% 0.9% 6.6% 2.6% 0.4% 5.54%
Min / Max / Std dev. 89% / 89% / 33.2% 88% / 71% / 39.9% 52% / 87% / 37.5% 23% / 89% / 42.4%
Index Adj. IRR % (Equ. V) 18.5% 12.2% 18.3% 11.4% 21.6% 14.8% 16.5% 10.8%
Min / Max / Std dev. 72% / 78% / 36.2% 17% / 69% / 29.9% 15% / 75% / 33.1% 72% / 78% / 29.7%
continued
Table 19.2continued
LBO Sponsored Deals
Whole Sample 19972000 20012004; 20082012 20052007; 2013
Average Median Average Median Average Median Average Median
Debt / EBITDA (Entry) 4.4x 4.5x 4.6x 4.5x 4.0x 4.1x 5.0x 4.9x
Min / Max / Std Dev. 1.0x / 18.0x / 4.0x 1.1x / 7.1x / 3.2x 1.1x / 18.0x / 4.9x 1.0x / 17.7x / 3.3x
Entry Multiple 7.8x 7.8x 7.6x 7.5x 7.3x 7.3x 8.4x 8.3x
Min / Max / Std Dev. 1.2x / 25.6x / 3.9x 1.2x / 14.3x / 3.7x 2.1x / 25.6x / 3.6x 3.2x / 25.1x / 4.0x
Management Stake 11.3% 10.3% 10.5% 9.8% 11.3% 10.1% 11.9% 10.8%
Min / Max / Std Dev. 0.5% / 25.3% / 5.2% 0% / 23.6% / 4.7% 0% / 22.3% / 4.9% 0.5% / 25.3% / 4.7%
342

Divestment 215 70 80 65
Build-up 178 65 12 101
LBO Sponsored Deals
Whole Sample 19972000 20012004; 20082010 20052007; 2013
Detailed data available Average Median Average Median Average Median Average Median
Observations 189 46 68 75
Change of CEO 105 29 31 45
Change of CFO 121 32 55 34
Table 19.2continued
LBO Sponsored Deals
Whole Sample 19972000 20012004; 20082010 20052007; 2013
Detailed data available Average Median Average Median Average Median Average Median
Board Size 7.3 7.5 6.9 6.8 7.3 7.1 7.6 8.2
Board External Members 26.3% 25.3% 21.3% 22.6% 24.3% 24.1% 31.2% 28.0%
Board PE Members 25.3% 24.8% 26.3% 25.4% 27.2% 27.6% 22.9% 21.9%
CEO & Chairman Different 64.3% 62.1% 66.3% 67.5% 56.3% 59.6% 70.3% 61.0%
New Incentive Plan 87.5% 87.9% 75.3% 78.8% 87.5% 88.6% 94.9% 92.8%

Notes: Panel A Shows a breakdown of the sample statistics by periods for the LBO transaction subsample.
343

Panel B
Strategic Deals
Whole Sample 19972000 20012004; 20082010 20052007; 2013
Average Median Average Median Average Median Average Median
Observations 1532 405 512 615
Average Deal Size (Entry) 265 289 272 295 270 278 256 294
Min / Max / Std dev. 51 / 1021 / 217 51 / 777 / 199 57 / 998 / 203 55 / 1021 / 211
Debt / EBITDA (Entry) 3.7x 3.9x 3.3x 3.6x 3.0x 3.2x 4.5x 4.7x
Min / Max / Std dev. 0.0x / 8.0x / 1.0x 0.0x / 7.2x / 2.9x 0.3x / 7.7x / 3.8x 0.0x / 8.0x / 3.9x
continued
Table 19.2continued

Strategic Deals
Whole Sample 19972000 20012004; 20082012 20052007; 2013
Average Median Average Median Average Median Average Median
Debt / EBITDA (Exit) n/a n/a n/a n/a
Min / Max / Std dev. n/a n/a n/a n/a
Entry Multiple 8.5x 8.6x 8.3x 8.7x 7.8x 8.0x 9.2x 9.1x
Min / Max / Std dev. 2.0x / 33.2.0x / 6.9x 2.1x / 24.4x / 3.3x 2.0x / 13.9x / 1.7x 2.2x / 33.2x / 3.9x
Management Stake 5.1% 5.2% 4.7% 5.1% 4.8% 5.3% 5.6% 5.3%
Min / Max / Std dev. 0.0% / 10.1% / 3.7% 0% / 9.8% / 4.4% 0% / 10.0% / 4.5% 0.0% / 10.1% / 4.1%
344

Strategic Deals
Whole Sample 19972000 20012004; 20082012 20052007; 2013
Detailed data availabe Average Median Average Median Average Median Average Median
Observations 197 51 62 84
Change of CEO 59 23 15 21
Change of CFO 78 21 20 37
Board Size 10.3 10.5 10.2 10.1
Board External Members 54.6% 54.9% 55.6% 53.2% 54.3% 52.3% 53.9% 58.1%
CEO & Chairman Different 55.6% 66.9% 57.6% 60.2% 53.2% 67.3% 55.4% 71.0%
New Incentive Plan 78.0% 89.2% 74.3% 79.9% 67.5% 86.5% 88.1% 97.5%

Note: Panel B shows a breakdown of the sample statistics on corporate governance by different periods for the strategic transaction subsample.
Pr iv ate E qu it y an d Val u e C re at ion 345

to adjust managements equity stake and its value is based upon the actual IRR realized
by the company, with leverage offered on the valuation of these stocks. This is called
a ratchet security mechanism. Adjusting managements equity stake and basing it upon
actual IRR tend to provide more incentives to managers than do plain vanilla stock
options. This process occurs because managers believe they have more direct control
to meet business objectives, as opposed to larger corporations where performance of
many different divisions can influence stock prices. Further, the exit valuation will be
less dependent on factors outside the true performance of the company based on the
work of managers rather than the more volatile nature of share prices of listed com-
panies that relies on more external factors. Also, informational asymmetry on LBOs
remains present despite buyer-initiated due diligence. The due diligence is perceived
as less intensive than that for listed companies driven by mandatory disclosure rules.

Transaction Value Leverage in Continental European


Leveraged Buyouts
Value creation analysis and the research on factors affecting this change are measured
comparing the return between the time of entry (LBO) and the time of exit (invest-
ment realization). This comparison permits us to measure the success of buyouts and
the relative importance of factors affecting the degree of success. Pioneering research in
this area includes Kaplan (1989), Wright et al. (1995, 2006), Nikoskelainen and Wright
(2007), Acharya et al. (2008), and Acharya et al. (2009).
As a proxy for transaction value, an enterprise value (EV) to EBITDA multiple is
used as the main metric because it represents the most relevant method of simplified
valuation measurement used by practitioners and corporate officers for comparing
the relative level of valuation in an M&A transaction. Financial leverage, differences in
tax rates, and depreciation policies do not affect EBITDA. Harbula (2007) provides a
demonstration of the neutrality of EBITDA multiples versus financial leverage. LBO-
sponsored transactions are compared to strategic transactions by comparing the corre-
sponding average valuation multiples with the average financial leverage for each year.
As previously suggested by the statistical analysis of the data sample, a trend is vis-
ible in Figure 19.1, in which multiples paid in LBO-sponsored transactions rose sub-
stantially since the middle of the first decade of the 2000s, more rapidly than multiples
in strategic deals. Additionally, such increases seem coupled with an increasingly high
leverage in the PE(LBO)-sponsored deals: the correlation coefficient between financial
leverage and relative valuation benchmark is above 0.80. This trend raises the question
of whether LBO funds are driven to overpay targets by misusing leverage when debt
financing is abundant and cheap.
The following sections further discuss the links among leverage, returns, asset valu-
ation, and corporate governance for LBO funds. Historically, asset valuation practices
of LBO funds reached irrational levels because of cheap and abundantly available debt.
To understand the extent to which leverage is involved in the inflation of asset prices
and the consequences on the economics of LBO funds, value creation and IRR will be
decomposed into different explanatory factors.
346 p r i vat e e q u i t y : u s e s a n d s t r u c t u r e

Leveraged Buyouts, Value Creation, and Leverage


This section explores the relationship between different value creation levers and the
performance of LBOs, both at an enterprise value and equity value level, and in terms
of deal performance. Besides the IRR measure, value creation is also measured by eco-
nomic value added (EVA). Table 19.3 presents details of the analysis. EVA is computed
as the difference between operating profits less the cost of all capital charges on the cap-
ital employed to produce those earnings. The EVA analysis is based on a ratio between
net operating profit after taxes (NOPAT) less the theoretical cost of capital employed
and the average capital employed (CE) for the year. The cost of capital is the weighted
average cost of capital. It is computed using a full-year, average market capitalization
for the weight of equity and the average level of net financial debt during a year for the
weight of debt. For all non-LBO, the transaction price is considered to be the market
price. The analysis is based upon the vintage (entry) years of the transactions.
The main leverage factors for value creation in LBO deals selected for this analy-
sis are: (1) enhanced corporate governance mechanisms, (2) cash flow yield improve-
ment, (3) below industry average acquisition multiple (bargain purchase), (4) the
positive impact of leverage, as predicted by the FCF theory, (5) management incentive
for the transactions success (alignment of interests), (6) improved operating perfor-
mance versus the industry average, and (7) changes in the valuation multiples of the in-
dustry between the entry and exit date. To measure the impact of corporate governance,
a dummy variable is created using a cumulative scoring index. The index encompasses
five possible measures for the presence of corporate governance changes: (1) change
in CFO or CEO, (2) presence of a management package, (3) change in the number of
external board members, (4) diminishing board size, and (5) the separation of the CEO
and chairman of the board. For the purpose of the cross-sectional data analysis, a cumu-
lative result for the index is computed by adding up a score (0 or 1) for all five criterion.
Based upon the cumulated result, the index takes a value of minus 1 if it falls in the range
[0 to 1], 0 if it falls in the range [2 to 3], and 1 if it falls in the range [4 to 5].
The analysis indicates that corporate governance mechanisms are major value crea-
tion levers for PE transaction returns as financial theory predicts regardless of the exit
strategy. They seem to play a more important role in LBOs than in public companies.
This observation is noted in the general sample as well as the subsamples between 1997
and 2000 and the periods from 2001 to 2004 and 2008 to 2010. However, a distinguish-
able decline occurs for exited deals between 2005 and 2007 in which corporate govern-
ance mechanism falter. For this same period (2005 to 2007), this leverage plays a more
important role for realized returns, while the role of FCF yield improvement declines
below that of the general sample. Similarly, the average operating performance improve-
ment also lags behind the global average. This finding indicates that debt increased in
this period while improvement in operating performance was weaker than average.
Conversely, the impact of build-up strategies and the change in the industry valuation
multiple seem to have the greatest impact for the 2005 to 2007 period, while below in-
dustry average valuation deals are less present. This result is related to the M&A market
as well as the evolution of market prices during this period. Finally, a favorable change
in the industry valuation multiple benefit returns. This evidence suggests that the
Table 19.3 I mpact of the Main Value Creation Levers on the Performance of Leveraged Buyout Transactions by Entry Year and Leveraged
Buyout Transaction Type
Value creation leversSummary LBO LBO LBO LBO Public to IPO exited Tertiary Control
transaction transactions transactions transactions Private LBO LBO LBOs group:
whole between between between transactions transactions Stock
sample 1997 and 2001 and 2005 and market
2000 2004; 2008 2007 benchmark
and 2010 index
Leverage
Management stake
Corporate governance
mechanisms
Deal size
Build-up strategy / acquisitions
Divestments
Free Cash Flow yield increase
Operating performance
improvement (beyond industry)
Below industry acquisition
multiple
Change in industry valuation
multiple

Note: This table exhibits the levels of statistical significance of the different valuation creation levers used by LBO funds both by the transaction entry (vintage) year as well
as for the different types of LBO transactions.
Statistical significance levels: > 0.1 > 0.05 > 0.01
348 p r i vat e e q u i t y : u s e s a n d s t r u c t u r e

prevailing mechanisms which lead to success of the LBO funds modus operandi failed
during the 20052007 subperiod.
Regardless of the observed period, tertiary LBOs have value creation mechanisms
that are close to those of LBOs between 2005 and 2007. Traditional value creation
mechanisms seem to be below the total sample average. Further, an increased reliance
exists on the use of leverage and the benefits of the change in the valuation multiple of
the industry between entry and exit. This conclusion is based on the idea that compa-
nies that have undergone two successive LBOs effectively manage working capital and
capital expenditures. As a result, bargains for financial sponsors rarely occur. Therefore,
either the capacity to use increased financial leverage or a build-up strategy can reason-
ably explain such investments.
Trends are similar for LBOs having gone through an IPO exit route and for public
to private transactions. If leverage remains an important factor of success, other value
creation levers are also observable. The IRR is driven mostly by (1) enhanced corporate
governance mechanisms, (2) enhanced operating performance, and (3) build-up strat-
egies as well as a favorable evolution between the entry and exit multiples. For most of
these deals, entry occurs below industry average multiples while the fund benefited at the
exit date of situations in which the industry multiples rose since entry. Factors explaining
the valuation multiple paid on transactions included in the data panel yield similar obser-
vations. For the overall sample, various factors explain the change in valuation multiple
(i.e., exit multiple versus entry multiple) including governance, improved performances,
and bargain purchases. However, all these factors seem to have vanished between 2005
and 2007. During that period, only leverage and changes in the industry valuation multi-
ple seem to explain a higher exit multiple compared to the entry level relative valuation.
Table 19.4 also shows the traditional LBO models failing. While leverage plays the
most important role in explaining the valuation multiples paid between 2005 and 2007,
corporate governance and operating performance improvements are the least impor-
tant. Leverage and change in the industry valuation multiples could be compensating
actual IRR returns for all other value creation factors. In other words, realized returns
on the LBO funds are engineered by using above average leverage, which benefits from a
favorable evolution of market multiples in the target industry rather than the traditional
company-specific features required to make an LBO transaction successful. The only
stable factor over time is related to the management involved in the transaction and the
average size of the managements stake. The analysis suggests the same underachieve-
ment of tertiary LBOs as opposed to the general sample for the same reasons as previ-
ously explained.
Following the work of Nikoskelainen and Wright (2007), Acharya et al. (2008),
and Acharya et al. (2009), the IRR is divided into different components in Table 19.5:
(1) impact of leverage, (2) industry- and market-specific impact, and (3) company-
specific performance. For the full sample, leverage explains about one-third of the IRR
performance, while market performance contributes 20 percent. Company-specific
performance explains the remaining 50 percent of variation. Above industry growth con-
tributes to the IRR to a similar extent as governance mechanisms (17 and 16 percent of
the total IRR, respectively), and the sum of these two factors equals the effect of leverage.
In comparison, the performance of the control group (a peer group composed of
stock exchange listed companies) exhibits a lower level of leverage than the overall
Table 19.4 E
 V/EBITDA Multiples in Leveraged Buyout Deals: The Impact of Different Value Creation Levers by Entry Year and
Transaction Type
EV/EBITDA multiple LBO LBO LBO LBO Public to IPO exited Tertiary Control
decomposition-Summary transaction transactions transactions transactions Private LBO LBO LBOs group:
sample between 1997 between 2001 between transactions transactions Stock
and 2000 and 2004; 20052007 market
20082010 benchmark
index
Leverage
ROCE
CAGR EBITDA
Operating Leverage stability
Corporate governance
Size
Build-up / Divestment
FCF yield
Operating performance improvement
(beyond industry)
Below industry acquisition multiple
Change in industry valuation multiple

Note: This table shows the levels of statistical significance of different financial metrics on the EV/EBITDA valuation multiple paid by the transaction entry (vintage) year as
well as for the different types of LBO transactions. The higher the significance level, the more relevant is the factor explaining a high (above industry average) valuation multiple.
Statistical significance levels: > 0.1 > 0.05 > 0.01
Table 19.5 Decomposing Internal Rate of Return in Leveraged Buyout Transactions by Entry Year
LBO signif. LBO signif. LBO signif. LBO signif. Control signif.
Transaction Transactions Group:
(%) (%) (%) (%) (%)
IRR (Equity) (a) 100.0 100.0 100.0 100.0 100.0
Leverage Effect 18.2 17.8 14.2 18.5 7.5
Leverage Effect vs. Sample Average 13.3 15.5 7.2 20.6 3.2
Full leverage Impact (b) 31.5 33.3 21.4 39.1 10.7
IRR (Enterprise Value) (c) = (a)(b) 68.5 66.7 78.6 60.9 89.3
and (c) = (i) + (o)
Market Multiples (d) 11.2 14.2 5.8 17.5 15.2
Adjusted IRR (e) = (c) (d) 57.3 52.5 72.8 43.4 74.1
Industry Performance (EBITDA Margin) (f) 5.3 4.2 4.6 3.2 8.7
Industry Growth (EBITDA CAGR) (g) 3.9 3.4 5.2 3.3 6.3
Industry and Market Impact (h) = (d) + (f) + (g) 20.4 21.8 15.6 24.0 30.2
Above Industry Margin (EBITDA margin) (i) 4.5 3.1 8.7 2.1 6.7
Above industry growth (EBITDA CAGR) (j) 5.6 5.0 6.3 4.0 9.8
Above industry FCF yield increase (k) 7.2 4.2 12.1 4.6 3.2
Corporate governance index (l) 16.3 12.2 19 12.0 14.3
Other (m) 14.5 20.4 16.9 14.2 25.1
Company-specific performance (n) = (i) + 48.1 44.9 63.0 36.9 59.1
(j) + (k) + (l) + (m)

Statistical significance levels: > 0.1 > 0.05 > 0.01


Pr iv ate E qu it y an d Val u e C re at ion 351

sample except for the 2009 to 2013 period. Moreover, internal factors (e.g., growth),
as well as stock market performance, mainly explain their value creation. However, for
LBO deals between 2005 and 2007, leverage explains the IRR to a greater extent than it
does for the full sample. Market and industry performance contribute up to 24 percent
(4 percentage points higher than the sample average) while company-specific perfor-
mance declines by 11 percent, especially on the LBO from an above industry growth
contribution perspective (9 percentage points less). These results are in line with the
previous observations. Leverage serves as a surrogate for other factors to engineer ac-
ceptable returns to investors and was available at low cost with favorable conditions
(e.g., weak or no covenants). By contrast, the improvement of operating performance
seems well below that for other periods or for the overall sample. This finding may be
a result of more favorable industry trends or placing less emphasis on target selection,
monitoring, and performance improvements.
This evidence shows that one of the virtuous items and a cornerstone of the LBO
model contributed to an inflationary asset bubble between 2005 and 2007. One possi-
ble reason for this result is that abundant cheap debt was available and a rise in collater-
alized and syndicated debt occurred (Altman 2007).
Tools other than valuations and leverages are available to analyze the behavior of LBOs.
Improvement in operating performances and growth patterns of companies under LBO
management are also potential ways to analyze the role that LBOs play in the economy. The
next section focuses on how companies under LBOs behave based upon a sample analysis.

The Impact of Leveraged Buyouts on Operating


Performance
Besides the value creation analysis in the previous section, a different possible angle for
the analysis investigates the impact of LBO funds on value creation. Instead of looking
at a funds realized return, investigating the impact on operating statistics and metrics is
also possible. This framework offers a different perspective on companies under LBOs
and is widely used for analysis in academic studies.
Kaplans (1989) investigation of post-buyout operating performance of 48 large MBOs
on public companies completed between 1980 and 1986 is based on an analysis of the op-
erating performance of the companies. He reports that in the three years after the buyout,
companies experienced increases in operating income, decreases in capital expenditures,
and increases in net cash flow. Specifically, operating income, adjusted for industry changes,
remains unchanged in the first two post-buyout years but increased in the third year. Guo
et al. (2011) find that gains in operating performance of companies under PE ownership
are either comparable to or exceed those observed by public benchmark firms. In contrast,
Leslie and Oyer (2008) find only very weak evidence of greater profitability or operating effi-
ciency of LBOs between 1996 and 2004, relative to public companies. Lerner, Sorensen, and
Strmberg (2011) provide evidence in contrast to the often-cited claim that PE has short-
term incentives to reveal data that shows buyout deals as increasing long-term innovation.
The first analysis in this section demonstrates the direct impact of PE ownership
on key operating performance metrics. In Table 19.6, key performance metrics are
352 p r i vat e e q u i t y : u s e s a n d s t r u c t u r e

compared between the pre-transaction and post-transaction period. As opposed to


strategic deals in which the variation between the pre-transaction and post-transaction
performance is the most important factor to the top line growth (in absolute terms or
relative to the industry) and the EBITDA margin, the impact of LBO ownership seems
the most important in EBITDA growth, as well as working capital and capital expendi-
tures efficiency ratios. This finding is consistent with the argument that value creation
leverages used by the funds and made evident in the previous section: improvement
in generating cash is mainly achieved by an improved profitability (EBITDA margin),
optimized cash generation cycle (working capital/sales), and investment policy (capi-
tal expenditures/sales). Also, as Chapman and Klein (2010) show, growth (measured
by EBITDA compound annual growth rate (CAGR)) is a key component for creating
value in PE transactions. However, during 2005 to 2007, LBO funds seem to put less
emphasis on cash management as the importance of working capital and capital ex-
penditure ratios are largely below the overall sample average. The impact of LBO owner-
ship on employment does not differ materially between the time periods. This finding is
consistent with the results reported in the previous section on the value creation levers.
Finally, in Table 19.7, ex ante expectations are compared with the ex post data for
transactions where data are available. Two subsamples are presented, exhibiting either
fully private deals or LBO transactions that end with an IPO or a RLBO. In the latter
case, the data do not comprise the voluntarily disclosed figures of the PE houses but use
publicly available figures, therefore providing a good test against any data selection bias.
According to the analysis, more than one-third of the companies surpass their initial
targets in terms of EBITDA growth and EBITDA margin. Yet, the whole sample falls
about 10 percent short of the initial objectives. The results are better for working capital
and capital expenditures management. In this case, more than half of the sample met or
exceeded the initial targets, despite the fact that the overall sample shows a small under-
performance in average. This finding is related to, among other things, bankruptcies and
distressed cases. Also, in most cases, the expected change between the entry and the exit
EBITDA multiple is attained.
Between 2005 and 2007, the number of companies achieving their initial targets
in terms of sales growth, EBITDA margin, or EBITDA growth lags behind the over-
all sample. Also, the expected change in the EBITDA multiple does not occur as ex-
pected in the original forecasts. This suggests that the initial investments plans included
expectations on continued increase in market multiples. Although this is observed in
the beginning on the period, it does not actually materialize. The results of the RLBO
subsample confirm those of the full data sample, thereby rejecting any risks that the
observations would suffer from a selection bias. PE companies appear to engage in more
aggressive margin preservation actions than their industrial counterparts. For the
cash-flow-generating items such as capital expenditures and working capital, the targets
are met to a similar extent between all samples and time periods. This finding provides
evidence of the great emphasis that LBO funds place on operating performances and
cash generation.
Finally, the EBITDA exit multiple targets are not met from 2005 to 2007, and to a
greater extent than for other subperiods or the overall sample. The subsample consist-
ing of companies choosing an RLBO perform overall on par with the overall sample.
RLBOs between 2005 and 2007, albeit not very numerous, are good performers and
Table 19.6 Analysis of the Impact of Leveraged Buyout Ownership on Operating Performance
LBO Deals (Post- 20052007 LBO Entry Year Strategic deals
TransactionPre-Transaction) Deals (Post-Transaction (Post-transaction
Pre-Transaction) Pre-transaction)
Change Change vs. Change Change vs. Change Change vs.
Different Industry Different Industry Different Industry
Operating Performance and LBO Ownership from Nil Average from Nil Average from Nil Average
CAGR Sales Growth
CAGR Sales GrowthCAGR Industry Growth
EBITDA CAGR
EBITDA CAGREBITDA Industry CAGR
EBITDA Margin (%)
EBITDA Margin (%) Industry EBITDA
Margin (%)
WC / Sales
WC / Sales Industry WC / Sales
Capex / Sales
Capex / Sales Industry Capex /Sales
FTE / Sales
FTE / Sales Industry FTE / Sales

Statistical significance levels: > 0.1 > 0.05 > 0.01


Table 19.7 Expected versus Actual Performance of Leveraged Buyout Deals by Entry Year
Difference between Realized and Full Information Private Full Information Private LBO RLBO RLBO Entry Year 20052007
Projected Operating Performance LBO Deal Deal Entry Year 20052007
vs % vs % vs % vs %
Expected Target Expected Target Expected Target Expected Target
Y+2 Y+2 Y+2 Y+2
(%) (%) (%) (%) (%) (%) (%) (%)
Sales Growth 12 34 22 23 10 33 12 30
Significance
Difference Full Period / 20052007
Significant
EBITDA Growth 9 40 26 19 11 40 17 37
354

Significance
Difference Full Period / 20052007
Significant
EBITDA / Sales 12 41 16 34 12 45 15 36
Significance
Difference Full Period / 20052007
Significant
ROCE 10 48 7 54 11 39 13 41
Significance
Difference Full Period / 20052007
Significant
Table 19.7continued

Difference between Realized and Full Information Private Full Information Private LBO RLBO RLBO Entry Year 20052007
Projected Operating Performance LBO Deal Deal Entry Year 20052007
vs % vs % vs % vs %
Expected Target Expected Target Expected Target Expected Target
Y+2 Y+2 Y+2 Y+2
(%) (%) (%) (%) (%) (%) (%) (%)
WC / Sales 7 58 7 55 8 69 9 63
Significance
Difference Full Period / 20052007
355

Significant
Capex / Sales 5 59 7 55 7 53 8 67
Significance
Difference Full Period / 20052007
Significant
(EBITDA Entry Multiple / 9 60 17 27 n/a n/a n/a n/a
EBITDA Exit Multiple)
Significance
Difference Full Period / 20052007
Significant

Statistical significance levels: > 0.1 > 0.05 > 0.01


356 p r i vat e e q u i t y : u s e s a n d s t r u c t u r e

certainly perform better than the overall sample for the same period. This finding is re-
lated to the fact that these companies had to convince the stock market of capabilities
and capacities to deliver and meet their initial targets. Only companies with a good track
record and solid fundamentals choose such a path.
In general, the impact of PE ownership is quite different from the overall sample
and the specific 2005 to 2007 period. For the latter, the operating performance analysis
seems to corroborate the findings of the section on value creation.

Summary and Conclusions


PE has attracted much attention from the public and scholars since 1990. Unlike in the
United States, the European PE market remains fragmented. In Europe, the amount
of investments, funds under management, and their role in the corporate world have
increased dramatically in the last decade. At the same time, the structure of the whole
industry has evolved with new players and has become more mature and transparent. As
Jensen (1989) contends, a risk exists that the same agency problems that the PE man-
agement seemed to resolve will consume them now if they become listed or more prone
to the principalagent problem.
Previous research considers that PE is either interested in turnaround situations (i.e.,
acquisition of distressed or underperforming assets in low growth but mature indus-
tries) or growth-based transactions in high-growth booming industries. European PE
focuses on these types of companies when investing. At the same time, LBO funds have
successfully engineered transactions that delivered good returns to their investors by
reducing the agency costs and improving efficiency. As financial theory predicts and
describes, those value creation levers were successfully at work in recent years for LBOs.
This explains much of the investment returns.
According to the sample, the main sources of value creation in LBOs are based on
the capacity of the fund to identify assets offering a good potential for growth and
cash flow improvement. This analysis did not consider the impact of target selection
mechanisms and the self-fulfilling impacts of the success of already established LBO
funds. While the overall trends show that PE investments in terms of enterprise value
hardly outperform the benchmark index, the results differ between the beginning and
the end of the analyzed period. In particular, between 2005 and 2007, PE returns are
higher than the sample average and the benchmark index primarily due to the market
asset price effect. However, for all years except between 2005 and 2007, LBOs outper-
form the index on an operating basis. In light of the different analysis performed, this
underperformance is attributed to an aggressive use of leverage to engineer equity re-
turns instead of focusing on the key value creation levers where PE funds have recently
excelled: delivering growth, using operating leverage, and providing an enhanced cash
flow yield. The implication is that increasing leverage levels began in 2013. In early
2014, staple financing (i.e., pre-packaged debt conditions offered by the sell side M&A
advisor in M&A auction processes) was being offered at seven times EBITDA and PE
players were bidding at above 10 times EBITDA in certain industries. Perhaps history
is repeating itself. PE built up massive cash balances recently.
Pr iv ate E qu it y an d Val u e C re at ion 357

Corporate governance, through the reduction of agency costs, is also a key item
in explaining the effectiveness of PE. PE deals exhibit much higher leverage levels
than a sample of strategic transactions. The cross-sectional analysis of returns con-
firms the monitoring and disciplining effect of indebtedness. Also, PE firms imple-
ment more governance measures (e.g., management incentives, management stakes,
and downsized boards) than public companies or companies subject to a strategic
transaction.
An occurrence between reductions of the asset holding period is observed which
seems correlated with the reduction of returns. While the relationship is not entirely
clear or proven, the average holding period decreased substantially between the begin-
ning and the end of the observation period. Moreover, the number of secondary and
tertiary LBOs also increased toward the end of the period. These transactions exhibit
the shortest holding periods from all transactions. This finding seems consistent with
the accused aggressive behavior of some investments funds that are oriented toward
making quick returns with highly leveraged transactions (abusing leverage) instead of
looking and selecting appropriate targets for creating value through governance and
operating leverage. It also raises concerns over an agency problem, as fund manag-
ers are remunerated not only based upon the achieved IRR but also on management
fees that rely on the amount of assets under management and number of transactions
realized.
The LBO model is not dead. Evidence indicates that it can work effectively
if managers consider the appropriate levers for value creation. In an economic
downturn, LBOs provide a powerful tool for saving and restructuring companies
as occurred in Europe. An LBO can also serve as a tool for organizing transi-
tion of small and medium enterprises from current owners to the next generation
of entrepreneurs. However, many more aspects need to be further explored. The
endogeneity between returns, ownerships, and governance needs to be analyzed
considering the potential unobservable facts about LBO funds targeting strategic
choices, and the natural moral hazard that involves fees incurred for managing the
assets, without being contingent (for this portion) on success. If LBO funds con-
sider the factors that drive value creation such as improvement of cash-flow yields
and corporate governance mechanisms when making deals, then the industry has
nothing to fear in the future.

Discussion Questions
1. Identify the main value creation factors for LBO transactions.
2. Discuss the factors contributing to the failure of traditional LBO models as ex-
plained by valuation multiples.
3. Explain the relationship among leverage, the IRR, and industry movements accord-
ing to the study presented in this chapter.
4. Discuss key performance indicators for which PE ownership had the most signifi-
cant impact and compare these indicators to those for strategic transactions.
358 p r i vat e e q u i t y : u s e s a n d s t r u c t u r e

References
Acharya, Viral, Moritz Hahn, and Conor Kehoe. 2008. Corporate Governance and Value Creation:
Evidence from Private Equity. Working Paper, London School of Economics and New York
University Stern School.
Acharya, Viral, Conor Kehoe, and Michael Reyner. 2009. Private Equity versus PLC Boards in the
UK: A Comparison of Practices and Effectiveness. Journal of Applied Corporate Finance 21:1,
4556.
Altman, Edward. 2007. Global Debt Markets in 2007: New Paradigm or the Great Credit Bubble?
Journal of Applied Corporate Finance 19:3, 1731.
Botazzi, Laura. 2010. Private Equity in Europe. In Douglas Cumming, ed., Private Equity: Fund
Types, Risks, Returns and Regulation, 437461. Hoboken, NJ: John Wiley and Sons, Inc.
Bruton, Garry, Kay Keels, and Elton Scifre. 2002. Corporate Restructuring and Performance:
An Agency Perspective on the Complete Buyout Cycle. Journal of Business Research 55:9,
709724.
Chapman, John, and Peter Klein. 2010. Value Creation in Middle Market Buys. In Douglas Cum-
ming, ed., Private Equity: Fund Types, Risks, Returns and Regulation, 229255. Hoboken, NJ:
John Wiley and Sons, Inc.
Guo, Shourun, Edie S. Hotchkiss, and Weihong Song. 2011. Do Buyouts (Still) Create Value?
Journal of Finance 66:2, 479517.
Harbula, Pter. 2007. Multiples de valorisation: Une approche approfondie. Analyse Financire 25
(Winter), 5659.
Jelic, Ranko, Brahim Saadouni, and Mike Wright. 2005. Performance of Private to Public MBOs:
The Role of Venture Capital. Journal of Business Finance and Accounting 32:34, 643681.
Jensen, Michael. 1986. Agency Costs of Free Cash Flow, Corporate Governance and Takeovers.
American Economic Review 76:2, 323329.
Jensen, Michael. 1989. Eclipse of the Public Corporation. Available at http://ssrn.com/
abstract=146149%20or%20http://dx.doi.org/10.2139/ssrn.146149.
Jensen, Michael, and William Meckling. 1976. Theory of the Firm: Managerial Behavior, Agency
Costs and Ownership Structure. Journal of Financial Economics 3:4, 305360.
Kaplan, Steven. 1989. The Effects of Management Buyouts on Operating Performance and Value.
Journal of Financial Economics 24:2, 217254.
Kaplan, Steven. 1991. The Staying Power of Leveraged Buy-outs. Journal of Financial Economics
29:2, 287313.
Kaplan, Steven, and Antoinette Schoar. 2005. Private Equity Returns: Persistence and Capital
Flows. Journal of Finance 60:4, 17911823.
Lehn, Kenneth, and Annette Poulsen. 1989. Free Cash Flow and Stockholder Gains in Going Pri-
vate Transactions. Journal of Finance 44:3, 771787.
Lerner, Josh, Morten Sorensen, and Peter Strmberg. 2011. Private Equity and Long-Run Invest-
ment: The Case of Innovation. Journal of Finance 66:2, 445477.
Leslie, Philip, and Paul Oyer. 2008. Managerial Incentives and Value Creation: Evidence from Pri-
vate Equity. NBER Working Papers 14331. Available at http://ideas.repec.org/p/nbr/
nberwo/14331.
Nikoskelainen, Erkki, and Mike Wright. 2006. Mechanisms on Value Increase in Leveraged Buy-
outs. CMBOR Occasional Paper.
Nikoskelainen, Erkki, and Mike Wright. 2007. The Impact of Corporate Governance Mechanisms
on Value Increase in Leveraged Buyouts. Journal of Corporate Finance 13:4, 511537.
Phalippou, Ludovic. 2012. Private Equity Funds Performance, Risk and Selection. In Phoebus
Athanassiou, ed., Research Handbook on Hedge Funds, Private Equity and Alternative Invest-
ments. Research Handbooks in Financial Law Series. Cheltenham, U.K.: Edward Elgar Pub-
lishing.
Phalippou, Ludovic, and Oliver Gottschalg. 2009. The Performance of Private Equity Funds.
Review of Financial Studies, 22:4, 17471776.
Pr iv ate E qu it y an d Val u e C re at ion 359

Renneboog, Luc, and Tomas Simons. 2005. Public-to-Private Transactions: LBOs, MBOs, MBIs
and IBOs. ECGI Finance Center Discussion Paper No. 200598.
Renneboog, Luc, Tomas Simons, and Mike Wright. 2007. Why Do Public Firms Go Private in the
UK? The Impact of Private Equity Investors, Incentive Realignment and Undervaluation.
Journal of Corporate Finance 13:4, 591628.
Wright, Mike, Luc Renneboog, Tomas Simons, and Louise Scholes. 2006. Leveraged Buyouts in
the UK and Continental Europe: Retrospect and Prospects. Journal of Applied Corporate Fi-
nance 18:3, 3855.
Wright, Mike, Nick Wilson, and Ken Robbie. 1995. The Longer Term Effects of Management Buy-
Outs. Journal of Entrepreneurial & Small Business Finance 5:3, 213234.
20
Compensation Structure
JI-WOONG CHUNG
Assistant Professor, Korea University Business School

Introduction
A private equity (PE) fund usually lasts for about 10 years and is typically organized as a
limited partnership in which fund managers serve as general partners (GPs) and investors
are limited partners (LPs). GPs are responsible for the funds daily management and are
personally liable for the partnerships debts and obligations, although they usually avoid
unlimited liability by forming a general partnership through a limited liability company
(LLC). By contrast, LPs have limited liability and do not participate in managing the part-
nership. Inherently, the GPLP relationship suffers from agency problems because LPs
cannot closely monitor GPs activities. Therefore, the limited partnership agreements
(LPAs) stipulate various conditions to align the interests of GPs and LPs by specifying
covenants, the financing processes, and fee structures. This chapter examines PE fund fee
structure and explores how different fee structures may create different incentives for GPs.
The standard compensation structure of PE funds is known as 2-and-20, where the
2 refers to a management fee of 2 percent of committed capital and the 20 refers to a
GPs carried interest of 20 percent of profits. This structure, including substantial carried
interest that greatly exceeds the usual pay-for-performance sensitivity of chief executive
officers (CEOs) of publicly traded corporations, is a key driver of PE funds success
(Frydman and Jenter 2010).
The actual compensation structure of PE funds is much more subtle and complex
than the 2-and-20 rule implies. Different LPAs specify different rules for how to cal-
culate, when to recognize, and how to split profits and so forth. Many of these detailed
fund terms are open to intense negotiation between GPs and LPs. The expected size of
GP compensation and the effective incentives of GPs vary substantially across funds
depending on how fund terms are arranged and written.
The rest of chapter is organized as follows. The next section explains the structure of
different types of fees that GPs earn, variations of the basic fee structure, and incentive
outcomes of different fee arrangements. Next, the chapter surveys the academic litera-
ture about PE compensation structure and discusses such issues as whether fees gener-
ate enough pay-for-performance for GPs, how fund and market characteristics relate to
fee structure, and how the bargaining power between GPs and LPs alter compensation
practices in the PE industry. The final section provides a summary and conclusions.

360
C om pe n s at ion S t ru ct u re 361

Compensation Structure of PE Funds


This section reviews three major components of PE compensation: (1) management
fees, (2) carried interest, and (3) deal-level fees. The section discusses the basic struc-
ture as well as the variation of each fee component. It also shows how to compute each
fee, and how GPs and LPs interests can be misaligned under particular forms of fee
terms.

MANAGEMENT FEES
LPs pay an annual management fee that is a percentage of the amount invested by the
LP. Management fees cover the ongoing operating expenses of the partnership such as
the salaries of the investment team, rents, and other costs associated with investment
activities. As implied in formulating the management fee, the absolute dollar amount of
the fee is higher for larger funds based on an assumption that greater required resources
are necessary to manage those funds.
The annual management fee percentage usually falls between 1 and 3 percent and is
paid on a quarterly, semi-annual, or annual basis in advance. The fee basis could be (1)
committed capital, (2) a combination of uncalled committed capital and cost basis of
ongoing investments, (3) net invested capital, or (4) net asset value (NAV). The first
two bases usually are used in the funds earlier life.
The fee percentage is either fixed or variable throughout a funds lifetime. In a
variable management fee, a fund reduces its fee level from its initial level after the
investment period, which is generally the first five years of the funds life. The fee
percentage also decreases with the increased fund size on the ground that econo-
mies of scale exist in managing the fund. The fee percentage also may be reduced
when a follow-on fund (new fund) is raised based on the view that a part of the
fixed overhead will not rise simply because a new fund is formed. For example, a
partnership does not necessarily need extra office space to manage an additional
fund.
LPs committing a larger amount of capital could have more favorable fee arrange-
ments by paying a lower management fee percentage. Thus, different investors in the
same fund may pay different fee percentages. Further, PE funds that do not require
greater oversight and management resources demand a smaller management fee per-
centage. For example, buyout (BO) funds, on average, charge a lower fee percentage
than venture capital (VC) funds, and funds-of-funds do not charge as high a fee as reg-
ular PE funds.
PE funds sometimes specify a different fee percentage depending on the aggregate
amount of the fund to be raised. For example, the fee could be set at 2 percent if com-
mitted capital is below $1 billion and 1.5 percent otherwise. Management fees are typ-
ically paid out of committed capital in the early years of the fund or the proceeds from
investments in later years of the fund. However, management fees could be paid to GPs
apart from the stated committed capital. The latter is especially useful when different
investors pay different fee percentages, in which case computing profit-sharing among
LPs becomes challenging.
362 p r i vat e e q u i t y : u s e s a n d s t r u c t u r e

The fee basis can also be variable. It is typically based on committed capital or a com-
bination of uncalled committed capital and the cost basis of unrealized investments
during the five-year investment period. Then it switches to net invested capital, which
is the cost basis of all investments minus the cost basis of realized investments (i.e., the
cost basis of ongoing investments). The rationale is that management fees compen-
sate for the cost associated with continuing to operate a fund. This cost tends to fall in
later years of the funds life when GPs mainly focus on harvesting existing investments
via sales, public offerings, or bankruptcy, which is a period when arguably less effort is
required.
PE funds could introduce a budget-based management fee in which GPs periodically
present their planned budget for operating expenses and get approval from LPs or the
representatives of LPs at annual meetings. However, the drawbacks of this approach are
the added costs associated with budget planning and the concern for sharing sensitive
financial information with outsiders.
LPs should understand the possibility that a different fee basis can cause different
incentives for GPs at the expense of LPs. For instance, when committed capital is the
basis, GPs have an incentive to raise larger, perhaps excessively large, funds, which might
erode future fund performance by paying GPs even during the divestment period when
no investment activity exists.
The management fee as a percentage of uncalled committed capital and the cost basis
of unrealized investments can be used as a basis. This approach excludes the cost basis
of exited investments from the fee basis based on the assumption that monitoring and
investment-related activities is no longer needed for these liquidated investments. Al-
though they may not be materially large if a portion of capital contribution is used to pay
for expenses (e.g., fees), LPs effectively pay a fee on fees.
Net invested capital better reflects the intensity of investment and monitoring activi-
ties by GPs. However, net invested capital is rarely used as a management fee basis in
the early years of a funds life because the partnership does not usually have sufficient
monetary resources to support investing activities during this period when net invested
capital is minimal. Using net invested capital as a fee basis in later years could incentiv-
ize GPs to delay realizing poor investments as late as possible to maximize net invested
capital and related fees (Robinson and Sensoy 2013).
Although funds used NAV as a fee basis in the early 1980s, this practice has virtually
disappeared in the BO and VC industries. NAV is the estimated market value of ongo-
ing portfolio companies (i.e., companies owned by the partnership). Since GPs perform
the NAV valuation, they could overestimate the value of portfolio companies to boost
their fee revenue or to influence future fund raisings (Barber and Yasuda 2013; Brown,
Gredil, and Kaplan 2014).
The LPs should be aware of the incentive effects of the management fee structure. They
negotiate the terms of the contract with GPs in order to better align the interests of GPs
with themselves and yet provide reasonable income to the GPs. Clearly, tension exists be-
tween the LPs and the GPs on the terms of management fee. Excessive fees threaten the
relationship between the LPs and the GPs especially when the expected management fees
are greater than the expected carried interest (performance-based revenue). Conversely,
reasonably low fees can demotivate the GPs by distracting them from focusing on operat-
ing the fund and hampering the GPs from attracting talented investment managers.
C om pe n s at ion S t ru ct u re 363

CARRIED INTEREST
Carried interest, which is also called carry, promote, or override, is the GPs share of the
profits from a PE fund. Carry is a performance-based fee to align the interests of LPs
and GPs. This fee structure generates high-powered pay-for-performance for GPs and is
the key driver of success of PE funds. Most PE partnership agreements stipulate a car-
ried interest of 20 percent. In fact, Robinson and Sensoy (2013) report that more than
90 percent of PE funds choose a 20 percent carry. Similar to management fees, carried
interest is lower for fund-of-funds than regular funds and BO funds than VC funds. Cal-
culating carried interest is complex and involves intense negotiation between LPs and
GPs. Indeed, profit distribution methods and carried interest calculations rank as the
most important provisions in LPAs by LPs and GPs (Tuck 2004).
The implicit amount of incentives provided to GPs varies substantially across funds
depending on the detailed terms of carried interest including the size, timing, and cal-
culation method of carried interest. Partnerships should determine how to calculate
profits, when to recognize profits, and how to split the profits between LPs and GPs.
Partnership agreements that spell out the rules governing payment priorities among
partners are known as waterfall provisions.
When computing profits, one problem is how to treat management fees, organiza-
tional expenses, and other income or expenses. Partners must decide whether these fees
and expenses should be paid by the fund and whether the funds profits should net these
fees and expenses when calculating carried interest. Organizational expenses are the costs
associated with raising and forming the fund. Other income or expenses include inter-
est on short-term investments, dividends from portfolio companies, litigation expenses,
the costs of annual meetings, and deal-level fees such as transaction fees, monitoring
fees, and break-up fees. A later section provides a detailed discussion of these fees.
Consider a simple example of a fund with a committed capital of $100 million, an
annual management fee of 2 percent of committed capital, and a one year life. Over the
life of the fund, total management fees amount to $2 million. The partnership must first
decide whether these fees are paid out of the committed capital or separately by the LPs
and whether shared profits are based on $98 million (i.e., invested capital net of fees) or
$100 million (contributed capital). In the latter case, the GPs must first recoup manage-
ment fees before they share profits.
Often LPs receive a minimum rate of return before GPs start receiving carried interest,
which occurs more so in BO than VC funds. This threshold return is called a preferred return,
which is typically 8 percent per year. In other words, if a fund does not produce more than
8 percent annual return, the GPs do not receive carried interest. This pure preferred return
may come with a catch-up provision, usually called the hurdle rate. BO fund agreements
almost always include a preferred return term while VC funds rarely do so (Fleischer 2005).
Several issues must be considered when a partnership specifies a preferred return
and a hurdle rate. For example, the following questions should be addressed and clari-
fied in the partnership agreement: Should the preferred return be compounded? If so,
how frequently? Does the preferred return accrues to all contributed capital (including
fees and expenses borne by the fund) or invested capital only (capital used for invest-
ments)? Should a partnership have multiple hurdle rates? Schell (2005) and Breslow
(2009) provide more discussion about detailed contractual terms of preferred returns.
364 p r i vat e e q u i t y : u s e s a n d s t r u c t u r e

A catch-up provision allows GPs to quickly recover their carried interest after a pre-
ferred return is delivered to LPs. The recovery speed varies depending on the catch-up
allocation rule. A 100 percent catch-up provision allows GPs to recover 100 percent of
the agreed-upon carried interest of the cumulative profits before sharing the profits with
LPs, according to the carry provision (e.g., an 80/20 split).
An example can help explain the exact mechanism of the catch-up provision. Con-
sider a fund with a 20 percent carried interest and a hurdle rate of 8 percent. The com-
mitted capital is $100 million, and the entire amount is invested on the first day of its
operation. The investment is sold for $130 million in one year, and the fund is liquidated
on that day (i.e., the internal rate of return (IRR) equals 30 percent). Figure 20.1 shows
the profit distribution of $30 million between LPs and GPs. This chapter considers two
scenarios for the catch-up provision.
The first scenario involves a 100 percent catch-up provision. The LPs first recoup the in-
vested capital of $100 million. They also receive 8 percent (the hurdle rate) of the invested
capital (i.e., $8 million). Next, the GPs receive 100 percent of profits until they receive 20
percent of the cumulative profits. Since the $8 million that the LPs recover represents 80
percent of the cumulative profits, the (corresponding) cumulative profits are $10 million
(i.e., $8 million/0.80). Therefore, the GPs receive 20 percent of the $10 million, which is
$2 million. The remaining profits, $20 million (= $30 million $8 million $2 million),
are allocated 80 percent to the LPs ($16 million) and 20 percent to the GPs ($4 million).
In sum, the GPs earn 20 percent of the total profits ($2 million + $4 million = $6 million =
0.20 [$30 million]) and the LPs 80 percent ($8 million + $16 million = $24 million = 0.80
[$30 million]) of the total profits, $30 million.
The second scenario involves a 50 percent catch-up provision. As in the first sce-
nario, the LPs collect the first $8 million. The GPs receive 50 percent of the profits until
they receive 20 percent of the cumulative profits. If $x is the incremental profit (over $8
million), then the cumulative profits are $8 million plus $x. Thus, $x would be deter-
mined algebraically as shown in Equation 20.1:

0.50($x) = 0.20($8 million + $x) (20.1)

The 50 percent of the incremental profits that the GPs receive should equal 20
percent of the cumulative profits. Therefore, $x = $5.33 million. Hence, for the next
$5.33 million in profit, the GPs and the LPs are allocated $2.67 million (= $5.33
[0.50]) each. The remaining profits, $16.67 million (= $30 million $8 million
$2.67 million $2.67 million), are allocated 80 percent to the LPs ($13.33 million)
and 20 percent to the GPs ($3.33 million). Therefore, the GPs earn 20 percent of the
total profits ($2.67 million + $3.33 million = $6 million = 0.20 [$30 million]) and the
LPs earn 80 percent ($8 million + $2.67 million + $13.33 = $24 million = 0.80 [$30
million]) of the total profits of $30 million. Figure 20.1 shows the final split of the
profits between GPs and LPs.
In these examples, both catch-up allocation rules (100 percent and 50 percent)
result in the same profit-sharing between the LPs and the GPs because the fund gener-
ates sufficiently high profits. However, as Figure 20.1 shows, when profits are not high
enough, different catch-up allocation rules will generate different allocation outcomes
C om pe n s at ion S t ru ct u re 365

Case 1. 100% Catch-up Provision

8020 split
$16 m
$4 m
Catch-up 100%
$2 m
Hurdle rate

$8 m
Committed
capital

$100 m

LP GP
Case 2.50% Catch-up Provision

8020 split
$13.33 m
Catch-up 50% $3.33 m
$2.67 m $2.67 m
Hurdle rate

$8 m
Committed
capital

$100 m

LP GP

Figure 20.1 Illustration of Profit-Sharing between Limited Partners and


General Partners This figure considers a fund with a 20 percent carried interest and
a hurdle rate of 8 percent. The committed capital is $100 million and the whole amount
is invested on the first day of its operation. The investment is sold for $130 million in one
year and the fund is liquidated on that day. The distribution of the profits of $30 million is
shown.

for the partners. Specifically, if the cumulative profits are between $8 million and $13.33
million in the example, the two allocation rules yield different allocation of profits to the
LPs and the GPs. Without a catch-up provision, the GPs effectively receive less than a
specified (e.g., 20 percent) carried interest.
Another important issue involves when to recognize and distribute profits. Two
types of waterfall provisions are available: (1) American style, which is also known as
366 p r i vat e e q u i t y : u s e s a n d s t r u c t u r e

deal-by-deal or front-end-loaded distribution, and (2) European style, which is also


known as aggregation method or back-end-loaded distribution.
American style waterfall computes and shares profits on a deal-by-deal basis. Al-
though popular in the infancy of the PE industry, especially in the United States, the
pure form of American style waterfall has become less popular in recent years, as this
waterfall structure overly favors GPs.
In American style waterfall, profits are computed for each portfolio investment. If
profits are earned, GPs share a pre-specified carry after taking into account, if any, a
preferred return or hurdle rate. However, if losses occur, only LPs bear the losses. The
economic interests of LPs and GPs can be misaligned. GPs have a convex compensation
structure as a call option and, all else being equal, have an incentive to increase the risk
in portfolio investments at the expense of LPs.
A deal-by-deal carry may come with a loss carry forward provision to mitigate this
problem in which carry is computed after considering the losses incurred in the previ-
ous investments. For example, suppose the GPs make two investments with the cost
basis of $10 million each and a 20 percent carry on a deal-by-deal basis. The first invest-
ment generates a $5 million loss and the second one earns a $5 million profit. Without
a loss carry forward provision, GPs would receive zero carry in the first deal and a $1
million (= 0.20 [$5 million]) carry in the second deal. However, with a loss carry for-
ward, the loss of $5 million from the first deal will carry forward and net out the profit of
$5 million in the second deal. Hence, the GPs will not share the profit from the second
deal. This example ignores the time value of money.
Another mechanism to alleviate over-paying carry to GPs is to include a clawback
clause in LPAs. A clawback clause allows LPs to collect back overpaid carry from the
GPs at the end of a funds life although interim clawback clauses can be set up. However,
several issues arise in executing a clawback provision such as when to turn on a clawback
and how to make sure LPs receive overpaid fees.
Suppose a fund manager makes two investments simultaneously, each costing $100
million. The first portfolio is liquidated at $130 million in one year, and the second one
is divested at $80 million in two years. The carry is 20 percent and the hurdle rate is
8 percent per year with a 100 percent catch-up. The first divestment results in a $30
million profit and an IRR of 30 percent. The LPs receive the invested capital ($100 mil-
lion) plus an 8 percent preferred return ($8 million). The GPs then receive 100 percent
catch-up on the next profits until they recoup 20 percent of the cumulative profit (i.e.,
$2 million). Next, LPs and GPs share the remaining profit ($20 million) according to
the carry rule: $16 million (= 0.80 [$20 million]) to the LPs and $4 million (= 0.20
[$20 million]) to the GPs.
The second investment generates a loss of $20 million and an IRR of 11 percent.
The loss accrues only to the LPs, and the GPs do not receive a return. At the end of
the second year, partnership should determine the total amount and distribution of
profits to the two groups of the partners. First, with the hurdle rate of 8 percent, the
LPs receive $209.17 million ($100 [1.08] + $100 [1.08]2) including the contributed
capital. Here the assumption is that the preferred return is earned on the invested cap-
ital for the actual investment period. Then, the GPs are supposed to receive 100 per-
cent catch-up, $2.28 million (0.20 [$11.4 million]), in which $11.4 million is equal to
$9.12 million/0.80. However, the total profits only amount to $10 million. Hence, the
C om pe n s at ion S t ru ct u re 367

GPs receive $0.83 million ($10 $9.17). With a deal-by-deal carry structure, the GPs
will have already received $6 million from the first investment. Therefore, according
to the clawback provision, the GPs should return $5.17 million (= $6 million $0.83
million) to the LPs.
Thus, several issues must be considered when implementing the clawback provision:
Should the time value of money that the LPs have forgone be ignored? What is the ap-
propriate way to treat tax payments that the GPs will have already made on their first
carry income? What happens if the GPs (individuals or the general partnership) do not
have enough capital to pay the clawback? All of these contingencies should be written
in LPAs. Mathonet and Meyer (2007) and Schell (2005) offer further discussion on the
complexities and difficulties in enforcing the clawback clause.
Another issue in a deal-by-deal waterfall provision is whether all fees and expenses
or only those fees and expenses associated with exited investments should be recouped
(on a pro rate basis) before profit-sharing. For example, suppose a fund with a com-
mitted capital of $100 million consists of five investments for the first three years of
$10 million each. During this investment period, management fees amount to $6 mil-
lion (= 0.02 [$100 million, 3 years]). If the first investment is sold at $20 million in
the fourth year, without a preferred return, the partnership agreement should specify
whether the amount of profits to be shared is $4 million (profits net of the entire man-
agement fees, i.e., $20 million $10 million $6 million) or $8 million (= $20 million
$10 million [$6 million/3 investments]). In the latter case, fees are spread out on a
pro-rata basis based on the cost basis of the investments.
In a deal-by-deal waterfall provision, the partnership may introduce the fair value
test, also called the NAV test. The fair value test permits GPs to share profits only after
the estimated NPV of existing investments exceeds a preset threshold. The threshold
typically falls between 110 and 125 percent of the cost basis of un-exited investments.
The fair value test reduces the probability of triggering the clawback provision. How-
ever, GPs can manipulate the estimates and reports of the portfolio companies NAVs.
All these issues and complications with profit distributions disappear if a partner-
ship follows a European-style waterfall provision (fund-as-a-whole basis carry). This ap-
proach allows GPs to participate in the carried interest on the whole-portfolio basis at
the end of a funds life. Clearly, this method simplifies fund accounting. Fund managers
do not have to worry about over-paying GPs, less need exists to have a clawback provi-
sion, and GPs have less incentive to manipulate the interim valuations of assets. Hence,
this waterfall provision is simpler. A clear disadvantage of this method from the GPs
perspective is a delay of the receipt of carry for several years. This waterfall is especially
unfavorable for younger GPs who may not have enough monetary resources or income.
The GPs may also have a more difficult time recruiting and retaining talented fund man-
agers with this distribution approach.
A return-all-contributed-capital-first approach lies between the pure American-
style and European-style waterfalls. In a return-all-contributed-capital-first approach,
GPs return the contributed capital to LPs first and then start sharing profits. In the pure
European-style approach, all committed capital is to be returned to the LPs. The interim
investments may generate enough profits to recover contributed capital including fees
and expenses before returning the committed capital to the LPs. Hence, the GPs may
be able to share profits earlier in this case than in the pure European-style approach.
368 p r i vat e e q u i t y : u s e s a n d s t r u c t u r e

DEAL-LEVEL FEES
GPs often receive fees from their portfolio companies, especially in BO funds. These
deal-level fees also called ancillary fees or portfolio company fees include transac-
tion fees, monitoring fees, and break-up fees. These fees represent compensation to the
GPs for the broad range of services they provide to portfolio companies. When acquir-
ing target firms, GPs perform due diligence, negotiate the terms of the acquisitions, and
arrange financing. That is, they provide investment banking services. They also serve as
the board of directors and continue to provide advisory services to portfolio companies
and receive monitoring fees. GPs sometimes receive break-up fees when an acquisition
fails due to the target company (e.g., when the target company accepts a higher offer
from a third party). Hence, a break-up fee is intended to remunerate GPs for their time
and resources spent in preparing the acquisition.
If GPs do not receive deal-level fees, the value of the portfolio companies will in-
crease by the amount of the fees and LPs profit would increase by the corresponding
amount. In the case of break-up fees, portfolio companies do not directly pay break-up
fees. However, LPs have to bear the opportunity costs of forgoing other investment op-
portunities. Therefore, deal-level fees are another layer of compensation that GPs re-
ceive from LPs.
Several problems are present with the practice of paying deal-level fees. First, LPs
already pay management fees to compensate GPs for their efforts in engaging in var-
ious activities to improve portfolio values and enhance fund performance. Hence,
the deal-level fees are redundant. Second, the deal-level fees increase the fixed part
of GPs compensation and weaken aligning interest between LPs and GPs because
if GPs receive deal-level fees, such fees will reduce the likelihood of participating in
carried interest. Third, of all fee terms, deal-level fees are the most opaque. LPAs typ-
ically do not have explicit contractual terms stipulating when, how much, and what
kinds of deal-fees will be charged. This practice has come under close investigation
by the Securities and Exchange Commission (SEC) (Securities and Exchange Com-
mission 2014).
Without adequate monitoring by LPs, an incentive could exist for GPs to improp-
erly maximize deal-level fees. For instance, GPs may collect deal-level fees (especially
transaction fees and monitoring fees) by employing services companies (directly or in-
directly) affiliated with the GPs, thereby compensating themselves secretively.
Deal-level fees should accrue to the fund, be shared with LPs, or be used to offset
management fees (called management fee waiver or offset). Management fee waiver also
helps GPs reduce their tax liabilities by converting ordinary income (management fees)
to capital gains (carried interest). This has been one of the most debated practices in PE
funds (Polsky 2008).

Literature Survey
This section reviews both academic research and industry surveys to explore both
cross-sectional and time-series variations in PE compensation structure. This section
presents findings from previous studies about the following questions: What are the
C om pe n s at ion S t ru ct u re 369

distributional characteristics of compensation structure? Do systematic differences


exist between VC and BO partnerships in their fee structure? Are fees related to other
fund characteristics such as size and age, and if so, why? Are fees designed to generate
enough incentives to GPs? What is the relationship between fee structure and fund per-
formance? Does any evidence support distorted incentives of GPs? Does the relative
bargaining power of LPs and GPs affect fee structure?
The literature on compensation structure of PE funds is relatively small largely due
to the private nature of the industry. Most studies obtain information on fund terms
from proprietary and confidential sources such as large investors in PE funds. Other
than these proprietary sources, perhaps the most comprehensive publicly available
source for fund terms and conditions is Preqins Private Equity Fund Terms Advisor.
However, Preqins data do not disclose fund identities and other detailed fund infor-
mation such as exact fund size and performance, thereby increasing the difficulty of
studying the relationship between fund terms and other fund characteristics including
performance.

C O M P E N S AT I O N S T R U C T U R E I N P R A C T I C E
To date, Metrick and Yasuda (2010) and Robinson and Sensoy (2013) provide the
most extensive studies of PE compensation structure. Both studies are based on fee in-
formation collected from undisclosed but large LPs. Table 20.1 summarizes the distri-
butional characteristics of fee terms from the two studies.
Management fees are about 2 to 2.5 percent of committed capital. Both studies show
that VC funds tend to charge greater management fees than BO funds, with a median
value of 2.5 percent and 2 percent, respectively. This finding is consistent with earlier
studies (Sahlman 1990; Gompers and Lerner 1999).
The fee level or basis typically changes over a funds lifetime. According to Metrick
and Yasuda (2010), 80 percent (for VC) to 90 percent (for BO) of the funds change
their fee level or basis. In Robinson and Sensoy (2013), the rate is about 60 percent for
both VC and BO funds. One interesting asymmetry between VC and BO managers is
that VC managers tend to change their fee level more often than their fee basis and BO
managers vice versa. Given that the better part of PE funds choose to change either fee
level or basis, the rationale behind this choice would be an interesting area for further
investigation.
Carried interest is almost always 20 percent. Metrick and Yasuda (2010) find that
95 percent of VC managers and 100 percent of BO managers choose a 20 percent carry.
According to Robinson and Sensoy (2013), more than 90 percent of the funds choose
20 percent. The cross-sectional variation of carried interest is much smaller than that of
management fees. The highly concentrated carry level may reflect a lack of transparency
on GPs ability in the industry. If this is true, as the industry matures, greater variation
in carried interest across different PE partnerships may be observed. Most funds have a
hurdle rate of 8 percent. While most BO funds stipulate a hurdle rate in their partner-
ship agreement, less than half of VC funds contain this term.
Studies on deal-level fees such as transaction fees, monitoring fees, and termina-
tion fees are rare. Frth, Rauch, and Umber (2013) examine the deal-level fees of 93
portfolio companies that went public from 1999 to 2008. They collect detailed data on
370 p r i vat e e q u i t y : u s e s a n d s t r u c t u r e

Table 20.1 Summary Statistics of Fee Terms


Metrick and Yasuda Robison and Sensoy
Sample Period 19892012 19842010
VC BO VC BO
Number of funds 94 144 295 542
Fund size 322 1,238 207.96 987.98
[225] [600] [106.12] [312.91]
Management fee % % % %
2.24 1.78
[2.50] [2.00]
<2% 9.6 64.9 5.0 23.0
2% 46.8 51.8 26.0 43.0
>2% 42.6 9.6 46.0 7.0
Fee % change 55.3 45.1 53.0 40.0
Fee basis change 42.6 84.0 14.0 43.0
Change % and basis 16.0 38.9 6.0 24.0
Carried interest
Percentage % % 20.44 19.96
[20.00] [20.00%]
<20% 1.1 0.0 1.0 2.0
20% 94.7 100.0 89.0 97.0
>20% 4.3 0.0 10.0 1.0
% hurdle rate 44.7 92.4

Note: This table compares descriptive statistics of fee terms from Metrick and Yasuda (2010) and
Robinson and Sensoy (2013). The numbers in brackets are medians.
Management fee is exactly equal to 1.5 percent.
Management fee is exactly equal to 2.5 percent.
Source: Table 2 in Metrick and Yasuda (2010) and Table 2 in Robinson and Sensoy (2013).

deal-related fees from the initial public offering (IPO) prospectuses of these reverse lev-
eraged buyouts (RLBOs). The study documents the following: (1) 57 percent of the
portfolio companies pay deal-level fees; (2) the proportions of deals paying monitor-
ing, transaction, and termination fees are 54.8 percent, 38.7 percent, and 24.7 percent,
respectively; (3) total deal-level fees amount to, on average, 2.32 percent of enterprise
value (in dollar terms, $10.79 million in which the average deal size is $923.3 million);
and (4) about 60 percent of the funds have transaction fee rebate rules. The fee rebate
rules specify whether transaction fees should accrue to the fund and how LPs and GPs
should share the fees.
C om pe n s at ion S t ru ct u re 371

Dechert and Preqin (2011) survey 143 BOs and document that transaction fees
amount to an average of 1.09 percent of deal size from 2005 to 2010 and monitoring
fees (for 95 BO deals) are on average 1.78 percent of earnings before interest, taxes,
depreciation, and amortization (EBITDA). Fee rebate rules exist for about 81 percent
of the deals. These statistics are largely consistent with anecdotal evidence showing
transaction fees are typically 1 to 2 percent of enterprise value, and monitoring fees
are 1 to 5 percent of EBITDA of a portfolio company (Metrick and Yasuda 2011; Stoff
and Braun 2014).
This section explores fee structure in PE funds. Notably, it identifies several differ-
ences in fee structures between VC and BO funds. Management fees are higher and
the change in fee levels is more frequent in VC funds. Hurdle rates and deal-level fees
occur more often in BO funds. Although this asymmetry may reflect different invest-
ment technology and operation/production processes of the two types of funds, a more
thorough theoretical consideration would provide further understanding about the ec-
onomics of PE funds.

THE DETERMINANTS OF FEE STRUCTURE


Several studies document the relationship between fee structure and fund or market
characteristics with little inconsistency. This section summarizes the relationships doc-
umented in Gompers and Lerner (1999), Robinson and Sensoy (2013), and Stoff and
Braun (2014). Several industry surveys also confirm the findings in these studies (SCM
2009; Dechert and Preqin 2011).
In general, fund size (committed capital) and age (fund sequence number) have a
negative relationship with management fees (as a percentage of fund size) for both VC
and BO funds. Additionally, capital flows into the PE industry are positively associated
with management fees. However, carried interest is not significantly related to capital
flows to the industry, but it is positively associated with fund size and fund age especially
for VC funds.
Several patterns emerge. First, management fees seem more cyclical than carried
interest implying that GPs negotiate for higher fixed pay when they have greater bar-
gaining power (i.e., when larger capital inflow occurs) (Robinson and Sensoy 2013).
Second, younger and smaller funds have higher management fees and less carried in-
terest (i.e., smaller pay-for-performance). Gompers and Lerner (1999) interpret this
result as consistent with a learning model in which initially neither LPs nor GPs know
the GPs ability, and GPs have a greater incentive to make an effort to establish a reputa-
tion. Over time, however, as a GPs reputation builds, extra incentive via greater pay-for-
performance is required to induce greater efforts from that GP. Hence, compensation
packages include larger variable components (i.e., carried interest) as a fund becomes
larger and older. The positive relationship between experience and pay-for-performance
also appears stronger for VC funds than BO funds. This asymmetry may imply greater
information opaqueness on venture capitalists ability in early stage funds and greater
informativeness of experience of venture capital funds.
In a unique study of PE funds in a cross-country setting involving 50 funds from
17 countries, Cumming and Johan (2009) show that a countrys legal condition affects
compensation structure. They find that management fees are larger in countries with
372 p r i vat e e q u i t y : u s e s a n d s t r u c t u r e

weaker legal environments, whereas carried interest is smaller in those countries. Their
interpretation is as follows: When a country with a poor legal condition has an opaque
information environment, evaluating the GPs skill becomes more difficult. Basing the
GPs pay on noisy signal about their efforts or skills is not optimal. Therefore, GPs prefer
that a greater fraction of their pay comes from fixed components.
One important missing variable in examining the relationship between fee structure
and other fund characteristics in the previous studies is GPs (ex ante) ability. Thus,
their analyses are subject to an omitted variable bias. In an attempt to address this issue,
Cumming and Johan (2009) examine how GPs personal characteristics relate to fee
structure. They find weak evidence that GPs with PhD degrees in science have higher
carried interest.
Overall, the information environment in the PE industry appears to be poor. Assess-
ing managers ability is challenging given the secretive nature of their investments (with
regards to due diligence, selection, and monitoring processes), long-term investment
horizon (in which performance is materialized long after inception and informative sig-
nals about managers cannot be observed frequently enough), and a lack of secondary
markets (from which fund managers skill is recurrently evaluated). Thus, compensa-
tion contracts have evolved to incorporate these imperfections of PE markets.

SIZE OF GP INCENTIVES
An important issue in compensation contracts is whether the compensation struc-
ture gives strong incentives for agents to make efforts for the benefit of principals (the
LPs). In the context of PE funds, fee structure should align the interests of GPs with
LPs. In particular, the LPs would want to have a greater fraction of variable compo-
nents (e.g., carried interest) relative to fixed components (e.g., management fees) in
the GPs total pay.
Metrick and Yasuda (2010) estimate the expected revenue of GPs coming from fixed
components (e.g., management fees and transaction fees) and variable components
(e.g., carried interest and monitoring fees). Under several reasonable assumptions about
investment speed, volatility, and exit timing, they show that the expected fee revenue
from fixed components is twice as large as that from variable components. This result is
discomforting and is largely consistent with the view that pay-for-performance may not
be large enough in PE funds.
Yet, explicit incentives generated from fund terms and conditions are only a part
of the total incentives that GPs face. When GPs manage a fund, they are concerned
with raising a follow-on fund. Current and past performance affects the success and
size of the next fundraising. The expected GPs revenue coming from fixed compo-
nents of the total fees from future funds will become greater if GPs can raise more
funds. Therefore, this indirect and implicit pay-for-performance also incentivizes
GPs, although it is missing in the previous analyses. Addressing this gap, Chung,
Sensoy, Stern, and Weisbach (2012) estimate the magnitude of the indirect pay-
for-performance and document that the indirect incentives are as significant as the
direct incentives from carried interest. Taking the results of the two studies, the ex-
pected revenue from variable components appear as large as that from fixed compo-
nents of total fees.
C om pe n s at ion S t ru ct u re 373

T H E R E L AT I O N S H I P B E T W E E N F E E S A N D F U N D P E R F O R M A N C E
LPs want to structure fees in ways that incentivize GPs to improve fund performance.
Does any relationship exist between fee structure and fund performance? Gompers and
Lerner (1999) examine the relationship between carried interest and fund performance as
measured by the size of IPO relative to committed capital, but they do not find a positive
association between the two variables. Similarly, Robinson and Sensoy (2013) estimate
the relationship between various fee components and net-of-fee fund returns and do not
find that high ex ante pay-for-performance generates high ex post returns.
These findings do not necessarily imply that a performance-based fee is not an im-
portant element to incentivize GPs to work harder so that their funds produce higher
returns. First, the relationship between compensation structure and fund returns is
endogenous. Especially, as pointed out in the previous section, the previous empir-
ical studies exclude many variables. For example, the GPs ability is hard to measure
and challenging to control in empirical investigations. Also, other incentive-alignment
mechanisms such as covenants and preferred returns are not considered. Such missing
factors could obscure the effect of the expected pay-for-performance on performance.
Second, as the authors of the two studies argue, compensation structures may be in
equilibrium. Gompers and Lerner (1999) contend that the learning model does not nec-
essarily imply a positive association between carried interest and performance because
compensation is properly structured to incentivize agents based on market information
on their ability. For example, younger GPs do not require strong performance-based
pay because they have a strong reputational incentive to work diligently. As Robinson
and Sensoy (2013) maintain, higher pay-for-performance may generate higher returns
gross-of-fees but GPs simply collect all economic rents that their efforts have produced.

F E E S A N D I N C E N T I V E D I S TO R T I O N
Although the GPLP relationship is fraught with potential agency problems, not much
systematic empirical evidence exists on whether fee structure directly affects GP behav-
ior at the expense of LPs. Robinson and Sensoy (2013), who conducts the only study
exploring this issue, find that GPs delay exits when fees are based on invested capital,
while they expedite exits when they are owed profit distribution around waterfall dates.
Ample evidence exists that shows other types of conflicts of interest between LPs and
GPs. For example, as Gompers (1996) shows, younger VC partners take their portfolio
companies public earlier and at a higher cost than older counterparts (i.e., greater underpric-
ing) to establish a reputation and facilitate new fundraising. Cumming and Walz (2010) doc-
ument that PE funds have an incentive to overstate the value of their existing investments to
help their follow-on fundraising especially in countries with weaker legal conditions and less
strict accounting rules. Barber and Yasuda (2013) and Brown et al. (2014) also find that PE
managers manipulate reported NAV before a follow-on fundraising occurs.

TIME SERIES DYNAMICS AND THE FUTURE OF FEE STRUCTURE


Since the financial crisis of 20072008, much debate has occurred about compensa-
tion practices in the PE industry. As market liquidity compromised these markets,
374 p r i vat e e q u i t y : u s e s a n d s t r u c t u r e

LPs have gained an advantage when negotiating contractual terms with GPs. In par-
ticular, the fees and expenses that PE funds charge have come under closer scrutiny
(Katz 2014; Maremont 2014). A trend has emerged toward lower management fees
and carried interest and more transparent disclosure requirements for deal-related
fees (Hudec 2010; Leamon, Lerner, and Garcia-Robles 2012; Mahieux 2013). The In-
stitutional Limited Partners Association (ILPA) also published a guideline for LPAs
to promote the interest of limited partners (Institutional Limited Partnership Asso-
ciation 2011).
Whether this movement will dramatically alter PE compensation practices is un-
certain. The PE industry is characterized by its boom and bust cycle (Kaplan and
Strmberg 2009). Since its inception, the industry has experienced three distinctive
crashes: one in the early 1990s (the collapse of the high yield bond market), another
in the early 2000s (the burst of the dot-com bubble), and the third in the late 2000s
(the financial crisis of 20072008). Each failure put downward pressure on fees earned
by GPs (Fenn, Liang, and Prowse 1995), followed by a recovery of fees to pre-crash
levels, except the most recent one. In other words, fees also appear to be cyclical. Rob-
inson and Sensoy (2013) document that LPs pay smaller percentage management fees
during bust periods.
If history is any guide, as the economy rises and fundraising activities pick up (Preqin
2012; Bain & Company 2014), fees will revert to their pre-2008 level. However, other
controversial fund terms (e.g., deal-level fees and management fee waivers) may be ad-
justed to be more favorable to the LPs. For example, Stoff and Braun (2014) show that
although fund terms have remained remarkably stable since 2007, progress has occurred
toward more LP-favorable terms, especially, in waterfall provisions and deal-level fees.

Summary and Conclusions


A distinctive feature of the PE industry is its secrecy. Investors concede this private
nature of the business so that fund managers can creatively capitalize in a timely
way on investment opportunities with minimal constraints. However, the secrecy
and flexibility granted to the fund managers come at a cost to investors in the form
of costly contracting and slow learning of GPs ability. Partnership agreements
should be carefully written to protect the interest of the investors by considering
various circumstances in which conflicts of interest between LPs and GPs may
arise. Besides, informative signals containing fund managers abilities are scant,
which impedes the investors learning process. Perhaps this lack of information is
why contractual terms are remarkably homogeneous across funds and the industry
is so cyclical.
This chapter reviews the basic structure and variants of different types of fees that
GPs earn. Current and potential investors in PE funds should meticulously examine fee
terms, understand the intricacies of fee structures, and demand transparency in distri-
bution methods when needed. As the industry matures, a secular movement may occur
in this direction. As more information about the workings of PE funds becomes accessi-
ble and the learning process improves, a greater heterogeneity of fund terms and condi-
tions adequately reflecting GPs characteristics may emerge.
C om pe n s at ion S t ru ct u re 375

Discussion Questions
1. Consider a fund with a 20 percent carried interest and a hurdle rate of 8 percent. The
committed capital is $100 million, and the entire amount is invested on the first day
of its operation. The investment is sold for $110 million in one year, and the fund is
liquidated on that day. Describe how profits are split between LPs and GPs under a
100 percent catch-up provision and a 50 percent catch-up provision.
2. Consider two funds that each invest $100 million today and produce $120 million
over the next three years. Both have an 8 percent hurdle rate, a 100 percent catch-up
premium, and a clawback provision. The two funds generate cash flows at different
times as shown below. Compute how the LPs and GPs will share the profits.
0 1 2 3

Fund A 100 +108 +2 +10


Fund B 100 +100 +10 +10

3. Denote c as the carried interest, h as the hurdle rate, and u as the catch-up. Deter-
mine the IRR of a fund at which the catch-up provision expires (i.e., GPs and LPs
share profits according to the preset carry rule).
4. Discuss how the relative bargaining power of LPs and GPs affects various fee terms
in LPAs.

References
Bain & Company. 2014. Bain Global Private Equity Report. Available at http://www.bain.com/
publications/business-insights/global-private-equity-report.aspx.
Barber, Brad M., and Ayako Yasuda. 2013. Interim Fund Performance and Fundraising in Private
Equity. Working Paper, University of California, Davis.
Breslow, Stephanie R. 2009. Private Equity Funds: Formation and Operation. New York: Practicing
Law Institute.
Brown, Gregory W., Oleg Gredil, and Steven N. Kaplan. 2014. Do Private Equity Funds Game
Returns. Working Paper, University of North Carolina at Chapel Hill and University of
Chicago.
Chung, Ji-Woong, Berk A. Sensoy, Lea Stern, and Michael S. Weisbach. 2012. Pay for Performance
from Future Fund Flows: The Case of Private Equity. Review of Financial Studies 25:11,
32593304.
Cumming, Douglas J., and Sofia A. Johan. 2009. Venture Capital and Private Equity Contracting: An
International Perspective. Burlington, MA: Academic Press.
Cumming, Douglas J., and Uwe Walz. 2010. Private Equity Returns and Disclosure around the
World. Journal of International Business Studies 41:4, 727754.
Dechert and Preqin. 2011. Transaction and Monitoring Fees: On the Rebound? Preqin Ltd.
Available at https://www.preqin.com/docs/reports/Dechert_Preqin_Transaction_ and_
Monitoring_Fees.pdf.
Fenn, George W., Nellie Liang, and Stephen Prowse. 1995. The Economics of the Private Equity
Market. No. 168. Board of Governors of the Federal Reserve System Staff Report.
376 p r i vat e e q u i t y : u s e s a n d s t r u c t u r e

Fleischer, Victor. 2005. The Missing Preferred Return. Journal of Corporation Law 31, 77117.
Frydman, Carola, and Dirk Jenter. 2010. CEO Compensation. Annual Review of Financial Econom-
ics 2:1, 75102.
Frth, Sven, Christian Rauch, and Marc Umber. 2013. The Role of Deal-Level Compensation in
Leveraged Buyout Performance. Working Paper, Goethe University.
Gompers, Paul A. 1996. Grandstanding in the Venture Capital Industry. Journal of Financial Eco-
nomics 42:1, 133156.
Gompers, Paul, and Josh Lerner. 1999. An Analysis of Compensation in the U.S. Venture Capital
Partnership. Journal of Financial Economics 51:1, 344.
Hudec, Albert J. 2010. Negotiating Private Equity Fund Terms: The Shifting Balance of Power.
Business Law Today 19:5, 4549.
Institutional Limited Partnership Association. 2011. ILPA Private Equity Principles. Toronto,
Canada: ILPA. Available at http://ilpa.org/ilpa-private-equity-principles/.
Kaplan, Steven N., and Per Strmberg. 2009. Leveraged Buyouts and Private Equity. Journal of Ec-
onomic Perspectives 23:1, 121146.
Katz, Alan. 2014. Bogus Private-Equity Fees Said Found at 200 Firms by SEC. Available at http://
www.bloomberg.com/news/2014-200407/bogus-private-equity-fees-said-found-at-
200-firms-by-sec.html.
Leamon, Ann, Josh Lerner, and Susana Garcia-Robles. 2012. The Evolving Relationship between
LP & GPs. Multilateral Investment Fund. Available at http://lavca.org/2012/09/19/16354/.
Mahieux, Xavier. 2013. Changes in the American Private Equity Industry in the Aftermath of the
Crisis. Comparative Economic Studies 55:3, 501517.
Maremont, Mark. 2014. Private-Equity Firms Fees Get a Closer Look. Available at http://online.
wsj.com/news/articles/SB10001424052702303743604579354870579844140.
Mathonet, Pierre-Yves, and Thomas Meyer. 2007. J-Curve Exposure: Managing a Portfolio of Venture
Capital and Private Equity Funds. Chichester, U.K.: John Wiley & Sons.
Metrick, Andrew, and Ayako Yasuda. 2010. The Economics of Private Equity Funds. Review of Fi-
nancial Studies 23:6, 23032341.
Metrick, Andrew, and Ayako Yasuda. 2011. Venture Capital and Other Private Equity: A Survey.
European Financial Management 17:4, 619654.
Polsky, Gregg D. 2008. Private Equity Management Fee Conversions. Working Paper, Florida
State University.
Preqin. 2012. PE Fundraising Gains Momentum in Q2 2012. Available at https://www.preqin.
com/item/preqin-pe-fundraising-gains-momentum-in-q2-2012/102/5354.
Robinson, David T., and Berk A. Sensoy. 2013. Do Private Equity Fund Managers Earn Their Fees?
Compensation, Ownership, and Cash Flow Performance. Review of Financial Studies 26:11,
27602797.
Sahlman, William A. 1990. The Structure and Governance of Venture-Capital Organizations. Jour-
nal of Financial Economics 27:2, 473521.
Schell, James M. 2005. Private Equity Funds: Business Structure and Operations. New York: Law Jour-
nal Press.
SCM. 2009. Annual Review of Private Equity Terms and Conditions. Zurich, Switzerland: Strate-
gic Capital Management. Available at https://www.yumpu.com/en/document/view/
15028517/2009-annual-review-of-private-equity-terms-and-conditions-scm-ag.
Securities and Exchange Commission. 2014. Examination Priorities for 2014. Washington, DC:
Securities and Exchange Commission. Available at http://www.sec.gov/about/offices/ocie/
national-examination-program-priorities-2014.pdf.
Stoff, Ingo, and Reiner Braun. 2014. The Evolution of Private Equity Fund Terms Beyond 2 and
20. Journal of Applied Corporate Finance 26:1, 6575.
Tuck. 2004. Limited Partnership Agreement Conference: Proceedings. Center for Private Equity
and Entrepreneurship, Tuck School of Business, Dartmouth.
21
Global Regulatory and Ethical Framework
HENRY ORDOWER
Professor of Law, Saint Louis University School of Law

Introduction
Private equity (PE) funds follow one of two basic business models. In some cases, they
operate as venture capital (VC) funds by providing capital to enterprises without ac-
quiring control in exchange for a large share of the enterprises profits when it becomes
financially successful. In their more common role, PE funds acquire control and, gener-
ally, complete ownership of an enterprise (a target) so the PE fund may restructure the
enterprise and resell it. Accordingly, much of the regulation of PE funds is regulation of
corporate takeovers. Major regulatory events address mainly the acquisition of control
and the elimination of minority ownership of a target.
Restructuring often involves redefining the targets business plan, reducing or ex-
panding the targets labor force, altering compensation of both managerial and rank and
file employees, and leveraging the target to improve the targets rate of return on equity
capital. It also involves disposing of the targets inefficiently deployed assets and occa-
sionally liquidating an inefficient target to unlock the embedded value of its underly-
ing assets. Since any of those restructuring activities affect other regulatory frameworks
such as labor and employment laws, agencies governing the credit markets, or corporate
rules on liquidation, the PE fund also becomes subject to those regulatory frameworks.
If the PE fund succeeds in making the target more productive and efficient than it was
before acquisition and restructuring, the PE fund may resell the target, either privately
or publicly, at a substantial profit to the PE funds investors.
Regulatory intervention affects the PE fund at several stages of its operation: (1)
assembling capital to fund an acquisition, (2) acquiring the target, (3) restructuring
the target, and (4) reselling the target following restructuring. At the acquisition stage,
the intensity of regulation depends on the dispersion of ownership of the target. When
management of a closely held enterprise invites a PE fund to provide capital and inter-
vene because of the enterprises need for funding, the PE fund acts as a venture capital
fund. In those instances, regulation is light. The PE funds intervention by invitation is
subject to arms-length negotiation between the owners of the closely held enterprise
and the fund. When the PE fund acquires a public company, heavier regulation to pro-
tect shareholders and possibly other stakeholders such as employees comes into play.

377
378 p r i vat e e q u i t y : u s e s a n d s t r u c t u r e

Many regulatory impediments take effect when incumbent managers of publicly traded
enterprises oppose the takeover. Often the interests of existing management do not
align perfectly with the shareholders interests (Easterbrook and Fischel 1981). Seeking
the best price for shareholders may conflict with existing managers desire to protect
their own employment and compensation or to capture favorable departure compen-
sation (Macey 1988). Managements evaluation and recommendation to owners for
the takeover may not alter the need for heavy regulation to protect shareholders even
though managements approval of a takeover removes impediments to the takeover.
Legislatures react to media attention and often discuss the need for increased PE
fund regulation whenever PE funds take over public corporations without the consent
of the corporations management. This type of takeover, which is called a hostile take-
over because it is without managements consent, is a controversial topic with view-
points widely ranging on its ethicality. Some commentators consider the activities of
the PE fund industry generally valuable to both the national economy and the invest-
ing public and thus favor limited additional regulation (Easterbrook and Fischel 1981).
Those commentators sometimes recommend increased transparency in the form of
more complete disclosure of financing and objectives of each fund. For example, a 2006
Danish report evaluating the need for regulation of the PE fund industry notes that cor-
porate takeovers by PE funds make the target company more efficient and profitable but
decrease tax revenues because of increased use of debt with its deductible interest pay-
ments (konomi og erhvervsminitriet 2006). Franz Mntefering, formerly a German
Minister of Labor and Social Affairs and head of the Social Democratic Party, repre-
sents the view that PE funds require tight regulation to protect the public. He compares
PE funds to a swarm of locusts destroying healthy enterprises critical to the national
economy (Seith 2005), a comparison that continues to resonate among scholars and
legislators in the European Union (EU) (Couret 2012).
The phenomenon of the hostile takeover of publicly traded corporations histori-
cally took root in the U.S. Regulation of those takeovers developed and matured under
U.S. law. So, this chapter focuses mainly on U.S. law. In the AIFM Directive (2011),1
the European Union addresses the role of PE funds and issues a framework for limited
regulation of PE funds and their investment activities through requirements imposed
on the managers of alternative investment funds including PE funds.

Private Equity Fund Structure for Capital Assembly


and Income Taxation
PE funds assemble equity capital by offering non-managerial interests in the fund to
investors. Limited partnerships are the most common entity choice for PE funds. In
a limited partnership, limited partner (LP) equity investors are passive and the fund
promoter controls the management function as the general partner (GP) and, consist-
ent with the funds stated investment objectives, the investment of the PE funds assets.

1
AIFM Directive, European Parliament and European Council Directive of 8 June 2011 on
Alternative Investment Fund Managers, Directive 2011/61/EU.
Gl obal Reg ul atory an d E t h ical F ram e work 379

The limited partnership also offers tax transparency so that the LPs and GPs are taxable
on their shares of the funds income, including the anticipated substantial gain from
resale or syndication of the underlying target operating entity, while the fund itself is
not taxable. Using tax transparent entities enables fund managers to receive most of
their management compensation as a share of the PE funds profita so-called car-
ried interestthat often yields tax-favored capital income (long-term capital gain in the
United States) from the sale of the underlying PE fund investments rather than ordi-
nary, fully taxed, fee income (Fleischer 2008).
Recently a U.S. court held that a fund was engaged in the active conduct of the tar-
gets operating business. That holding exposed the PE fund to the targets liabilities
under the Employee Retirement Income Security Act of 19742 (Sun Capital Partners
III, LP v. New England Teamsters & Trucking Indus. Pension Fund, 20133). Similarly, a
Danish tax ruling imputes the permanent establishment of the fund manager to LPs in
a fund (Riis and Khan 2014). Both decisions have uncertain implications for the con-
tinued predictability of tax and business outcomes for passive investors in PE funds.
The decisions might undermine the favored capital income character of the gain from
a successful PE fund investment and may subject all or part of that gain to taxation as
ordinary income in the targets home country. Investors anticipate taxation only in their
own countries of residence (Internal Revenue Code 1986, section 875).4
PE funds assemble debt capital by borrowing from institutional lenders or syndi-
cating debt instruments, often below investment grade junk bonds. PE funds using
substantial debt often insert a wholly owned intermediary corporation (PE fund corpo-
ration) as the immediate acquisition vehicle between the fund and the target corpora-
tion. The PE fund contributes its equity capital to the PE fund corporation, and the PE
fund corporation issues debt for the acquisition. A PE fund corporation may merge with
the target and shift the burden of repaying the acquisition indebtedness to the target
corporation by operation of law. The shift of the indebtedness to the target completes
a leveraged buyout (LBO) using the targets own assets and future income to repay the
acquisition indebtedness.

Regulation of Capital Assembly


Both the equity and debt interests in PE funds and PE fund corporations are securi-
ties (Securities Act of 1933 (SA), section 2(a)),5 so laws regulating the sales of secu-
rities govern issuing (SA) and reselling the interests (Securities Exchange Act of 1934
(SEA)).6 Securities regulation generally follows a disclosure model rather than imposing

2
Employee Retirement Income Security Act of 1974, Pub. L. 93406 (1974), codified in various
titles of the United States Code.
3
Sun Capital Partners III, LP v. New England Teamsters & Trucking Indus. Pension Fund, 725 F.3d
129 (1st Cir. 2013).
4
Internal Revenue Code, as amended 1986, 26 U.S.C. 1 et seq.
5
Securities Act of 1933, 15 U.S.C. 77a et seq.
6
Securities Exchange Act of 1934, 15 U.S.C. 78a et seq.
380 p r i vat e e q u i t y : u s e s a n d s t r u c t u r e

quality or other substantive restrictions on issuing and reselling interests. Nevertheless


disclosure-based regulation is rarely free from substantive restrictions.
The disclosure model of securities regulation assumes that financially sophisticated
participants in the financial markets assimilate information, use it to make efficient, ra-
tional investment decisions, and disseminate investment analysis to guide less sophis-
ticated investors to rational investment decisions. If professional participants in the
financial markets have access to all material information about the investment, they
can evaluate the information and enable the markets to price investment opportunities
correctly.
Because PE funds invest their assets in shares of operating companies, the funds
should be investment companies subject to investment company regulation (Invest-
ment Company Act of 1940 (ICA), section 3).7 Instead, statutory exemptions inform
and limit the pool of PE fund investors to enable PE funds to avoid classification as
(regulated) investment companies (ICA, section 3(c)(1), (7)). Investment company
regulation supplements securities regulation with substantive restrictions on the man-
agement, operations, and fees of investment companies.
Substantive restriction assumes a greater regulatory role where disclosures would
fail to protect the investing public or the financial markets, even in fully efficient capital
markets. For equity investments in pooled investment vehicles such as mutual funds,
substantive restrictions limit risk-taking and fraudulent behavior. In the United States,
legislation limits the size of the position that a fund may take in any single investment,
sharply restricts the debt a fund may incur, and bans short selling (ICA, section 12).
Related legislation prevents investment advisers from receiving result-based compensa-
tion that might cause the investment advisers to take unwarranted investment risks with
a funds assets to enhance investment return and generate a greater result fee (Invest-
ment Advisers Act of 1940 (IAA), section 205(a)(1)).8

E Q U I T Y C A P I TA L
PE funds assemble their equity capital from professional (AIFM Directive, Art. 4,
1.(ag)) or accredited investors (Regulation D, Rule 501)9 who have substantial income
or assets. In the United States, PE funds are exempt from the registration requirement
for their interests (SA, section 5) because they do not make public offerings (SA, sec-
tion 4(2)) of the interests. Rather under the Regulation D safe haven for non-public
offerings of securities, there is no public offering if the issuer sells its securities only to
accredited investors. Accredited investors include most institutional investors, charities,
pension plans, and individuals with net worth exceeding $1 million, exclusive of their
personal residence, or income exceeding $200,000 annually ($300,000 for married
couples). In issuing the regulation, the Securities Exchange Commission (SEC) views
accredited investors as individuals and entities that do not require the extensive pro-
tection of the securities laws because accredited investors can understand and evaluate
the risks of an investment or would employ a competent adviser to evaluate it, have the

7
Investment Company Act of 1940, 15 U.S.C. 80a-1 et seq.
8
Investment Advisers Act 1940, 15 U.S.C. 80b-1 et seq.
9
Regulation D, CFR 17: 230.501 et seq. (1982).
Gl obal Reg ul atory an d E t h ical F ram e work 381

bargaining power to ask questions and receive answers from the investment promoter,
and can bear the economic risk of loss.
Despite the registration exemption, PE funds remain subject to the securities laws
in all other respects. They must disclose all information that might be material to the
decision to invest or the retention or resale of the investment once made, and may be
subject to civil liability and criminal penalties for omission or misstatement of material
facts (SEA, section 10(b)).
PE funds that sell their interests only to qualified purchasers such as individuals who
have investments of at least $5 million (compare the $1 million of assets for accredited
investors) (ICA, section 2(a)(51)) are not investment companies (ICA, section 3(c)
(7)). Therefore, PE funds are free from regulation under the ICA allowing them to take
controlling positions in their target companies and to employ substantial borrowing
leverage.

F U N D M A N A G E R S A N D C O M P E N S AT I O N
Investment company and mutual fund advisers, for example, may not receive an incen-
tive fee based on the capital appreciation of the funds assets (IAA, section 205(a)(1)).
Incentive fees are critical to PE fund managers and permissible under an exemption
(IAA, section 205(b)(4)) for advisers to funds admitting only qualified purchasers as
investors (ICA, section 3(a)(7)). PE fund managers charge as much as 20 percent of
the increase in value of the investment as an incentive fee in addition to a 12 percent
annual fee on the value of the funds assets. Before the Dodd-Frank Wall Street Reform
and Consumer Protection Act of 2010 (Dodd-Frank),10 the result fee exemption was
available to investment advisers not required to register with the SEC and most PE fund
advisers were not required to register.
Dodd-Frank now requires most PE fund investment advisers to register (IAA, section
203). Recently, regulatory legislation has focused on the relationship between excessive
leverage and the stability of the financial markets, a concept referred to as systemic risk
(Dodd-Frank 2010; AIFM Directive 2011). Dodd-Frank requires advisers to maintain
records focusing specifically on systemic risk (IAA, section 204), and, under the Vol-
cker rule (Dodd-Frank, section 619 adding section 13 to the Bank Holding Company
Act of 195611), bans commercial banks and their affiliates from sponsoring PE funds.
The Volcker rule imposes broad control on systemic risk by prohibiting banks from in-
vesting speculatively. Similarly, in mid-2014, the German Ministry of Finance released
proposals that, if enacted, prohibit German insurers and pension funds from investing
in alternate investment funds unless the funds managers are subject to regulation in a
member country in the European Economic Area (Debevoise & Plimpton LLC 2014).
The AIFM Directive regulates the activities of EU and non-EU fund managers oper-
ating in member states and requires their registration with national authorities. A man-
ager of the portfolio must manage risk under the AIFM Directive. Annex II of the AIFM
Directive elaborates on general compensation recommendations in the European

10
Dodd-Frank Wall Street Reform and Consumer Protection Act of 2010, Pub. L. 111203, July
21, 2010.
11
Bank Holding Company Act of 1956, as amended, 12 U.S.C. Chapter 17.
382 p r i vat e e q u i t y : u s e s a n d s t r u c t u r e

Union (Remuneration Recommendation 2009)12 and addresses incentive compen-


sation directly. The annex requires no less than 50 percent of incentive compensation
be in interests in the advised fund, ensuring manager participation in the continuing
fund risk. Alternative investment funds should defer at least 40 percent of the incentive
compensation to prevent the fund manager from receiving incentive compensation on
transient increases in value. Instead, incentive compensation becomes a function of the
increase in the value of the fund over its normal life cycle. Life cycle incentive compen-
sation contrasts with the practice of many U.S. funds in which the manager receives the
incentive compensation as the fund value increases, but the fund does not recapture
or claw back compensation when the fund decreases in value. Rather, paying incentive
compensation often fixes fund value referred to as a high water mark up to which the
manager receives no further incentive compensation (Ordower 2012).

I S S UA N C E O F D E BT
While securities laws govern issuing debt instruments into the public markets, financial
rather than securities regulation precludes institutional lenders from excessive lending
to a single borrower (United States Code (U.S.C.) Title 12, section 84 (2014); Code
of Federal Regulations (CFR) 12: Part 32 (2013)). Lending limits protect the lenders
stability on which the public relies. For fully secured lending, a national bank may not
extend credit to a borrower that exceeds 15 percent of or have loans outstanding that
exceed 10 percent of the lenders unimpaired capital (U.S.C. 12, section 84(a)). For
PE fund acquisitions of large, public corporations lending limits prove a challenge for
financing the purchase with institutional borrowing.

Acquisition Regulation
While purchases of large positions or control in closely held corporations have at-
tracted minimal regulation, PE funds acquisitions of publicly traded corporations have
prompted legislation at both federal and state level in the United States and directives
in the European Union to regulate those takeovers. This section addresses that legisla-
tion and the litigation that ensued from hostile takeover activities. The section explains
how legislation and judicial decisions shifted the balance of power in hostile takeover
attempts from the hostile acquirer to incumbent management.

C L O S E LY H E L D B U S I N E S S A C Q U I S I T I O N
Regulation of closely held business takeovers is minimal. When the PE fund acts as
a venture capital fund, it provides financing for the enterprise and may receive both
a fixed return on its invested funds and a share of the future profitability of the enter-
prise. In those instances, the PE fund may play no role managing the enterprise. Existing

12
Remuneration Recommendation, Commission of the European Communities, Recommenda-
tion of 30 April 2009 on Remuneration Policies in the Financial Services Sector (2009/384/EC).
Gl obal Reg ul atory an d E t h ical F ram e work 383

management continues to operate the enterprise. In some cases, the managers of the PE
fund serve an advisory function in developing the enterprise.
When a PE fund acquires control of a closely held corporation, the former control-
ling owners typically receive a price premium for selling control in the company. Minor-
ity owners enjoy few protections when the controlling owner sells his or her interest.
Since minority shareholders are not selling their shares, securities laws are irrelevant. A
limited body of decisional law from state courts offers some protection to the minority
shareholders in instances where the buyer wastes or steals the corporate assets following
an acquisition of controlling share ownership. If the seller recklessly evaluated the buyer
or the offer to the detriment of the enterprise and the minority owners, courts have im-
posed liability on the seller (DeBaun v. First Western Bank (1975)13). Courts have rarely
required the controlling owners to share their opportunity and control premium with
minority owners, as the court did in Perlman v. Feldman (1955).14

P U B L I C LY T R A D E D C O R P O R AT I O N A C Q U I S I T I O N : H O S T I L E
TA K E O V E R S
When the PE fund selects a public corporation to acquire, a formal regulatory network
embodied in the United States in the Williams Act of 1968,15 in the European Union
under the Takeover Directive (2004),16 the laws of the EU member states consistent
with the Takeover Directive, and various antitakeover statutes in most of the U.S. states
come into play. Before the U.S. Congress passed the Williams Act to regulate hostile
takeovers, the securities laws required formal disclosure procedures only for takeovers
by proxy fights for voting control (SEA, section 13) and exchange offers, in which target
shareholders exchanged shares for the shares of the buyer. An exchange offer is a public
offering of shares of the acquiring corporation to a target corporations shareholders.
Exchange offers are subject to registration requirements (SA, section 5) and their ac-
companying disclosures. While general antifraud rules (SEA, section 10(b)) applied
to cash purchases of target shares and prevented buyers from trading on non-public in-
formation concerning the issuer, those rules required neither disclosure of non-public
information about the buyer because the buyer was not selling its shares nor the buyers
analysis of the target based on information available publicly.
Pre-Williams Act hostile acquisitions employed a three-step takeover paradigm that
developed in the 1960s. First, the PE fund or other acquirer purchased a position in the
target on the market and in privately negotiated transactions. Second, the PE fund made
a tender offer for sufficient target voting shares (sometimes all) to enable it to compel
the third step. In that step, the target combines with the PE fund corporation in a stand-
ard or short-form merger. For a standard merger, that control threshold lay between
50 and 67 percent depending on the targets state of incorporation. In some states, an

13
DeBaun v. First Western Bank and Trust Co., 46 Cal App 3d 686, 120 Cal Rptr 354 (1975).
14
Perlman v. Feldman, 219 F.2d 173 (2nd Cir. 1955).
15
Williams Act of 1968, Pub. L. No. 90439, 82 Stat. 454 (1968). Codified in sections 13(d) and
14(d) of the Securities Exchange Act.
16
Takeover Directive, European Parliament and European Council Directive of 21 April 2004 on
Takeover Bids (Directive 2004/25/EC).
384 p r i vat e e q u i t y : u s e s a n d s t r u c t u r e

affirmative simple majority vote in favor of a merger was enough to complete the merger
(Delaware Code Online, Title 8,17 Delaware General Corporation Law (Del GCL),
section 251(b)). Other states had a super-majority requirement for mergers. Most
common was a two-thirds requirement (Missouri Revised Statutes, section 351.425).18
Most states, however, allowed a short-form merger without a vote of the sharehold-
ers if one of the combining corporations owned at least 90 percent of the shares of the
other (Missouri Revised Statutes, section 351.447). The merger eliminated minority
shareholders in the target. Minority shareholders received compensation for their target
shares but could not prevent the merger. If the minority shareholders were unhappy
with the price, their recourse was to state law appraisal rights (Del GCL, section 262).
Appraisals tended to be an unsatisfactory remedy as courts relied on appraisal tech-
niques that rarely maximized value.

T E N D E R O F F E R S A N D R E G U L ATO R Y B A L A N C I N G
A tender offer takes place when a buyer publicly offers to buy a certain number of shares
at a specific price and under the specific sale requirements in the offer. Since the ten-
dered shares give the acquirer control, tender offers generally are at a price higher than
the current trading market because they include part of the control premium normally
absent from the market trading price.

The Williams Act


Pre-Williams Act tender offers tended to be coercive because shareholders had only a
short time in which to accept by depositing their shares into the offer. The offer would
close once shareholders deposited the number of shares the buyer wanted to purchase.
Often the tender offer would issue at the close of business on Friday in anticipation
that it would complete before incumbent management had an opportunity to respond.
Participants in the industry referred to those offers as Saturday night specials. Even if
the offer was not completed immediately, the short acceptance period, which was never
more than one week as required by an early version of the Williams Act, left the targets
management with little opportunity to analyze the value of the shares under the circum-
stances of a takeover. Finding out whether the price offered was fair or whether some
other buyer might be willing to offer a higher price was often difficult.
Despite the above market price, tender offers were coercive because generally they
were for fewer than all the public shares. A shareholders failure to accept the offer might
cost the shareholder the premium tender offer price. Quick acceptance was critical to
participation in the tender offer because the offer ended when shareholders tendered
the buyers desired number of shares. If the buyer did not receive tenders for sufficient
shares, it could either terminate the offer and return the tendered shares or increase the
offer price for additional shares without paying the new higher price to earlier sellers.
Shares not accepted during the tender offer would receive less desirable consideration
in the merger stepeither smaller in amount or in a mix of consideration that included

17
Delaware General Corporation Law, undated. Title 8 of the Delaware Code Online. Available at
http://delcode.delaware.gov/.
18
Missouri Revised Statutes 2013. Available at http://www.moga.mo.gov/statutes/statutes.htm.
Gl obal Reg ul atory an d E t h ical F ram e work 385

long-term below market grade debt. Additionally, the less desirable consideration for
the remaining shares in a merger would lead even well-informed shareholders not to
delay in accepting the offer.
The Williams Act sought to balance the interests of the buyer and the existing man-
agers of the target enterprise by requiring disclosure of the buyers intentions and delay-
ing the acquisition. The Act gave the targets management an opportunity to respond to
the tender offer, make a recommendation to shareholders, and, in some cases, make or
seek a competing offer.
Under the Williams Act, once the acquirer owns more than 5 percent of the targets
shares, it must disclose its beneficial ownership of the target, as well as its source of
funds for the acquisition and its intentions for the purchases of shares (SEA, section
13(d)). As prices on the market rise in reaction to the 5 percent disclosure that a buyer
intends to acquire control of the target, further market purchases in any substantial
volume become impractical. A tender offer becomes critical to assembly of a control
block. The Williams Act (SEA, section 14(d)) undercuts the coercive impact of tender
offers by requiring offers to remain open for 20 days. During the 20 days, sharehold-
ers may withdraw tendered shares. The Williams Act also obligates the buyer to accept
tendered shares throughout the offer period pro rata at identical prices. If management
develops or makes a competing offer, early tenderers are not bound by their tender
but may accept the competing offer. If the tender offeror increases the price to secure
enough shares, each tendering shareholder gets the higher price. The pro rata rule forces
the buyer to accept each tender in proportion to the total number of shares tendered by
all tendering shareholders eliminating the pressure for shareholders to make a quick de-
cision to tender. The pro rata rule creates an incentive for selling shareholders to tender
more shares than they own when they believe the offer will be oversubscribed. The SEC
countered this so-called short tendering with a regulation that bans the practice (CFR
17: 240.14e-4(b)).

The EU Takeover Directive


In the European Union, the Takeover Directive (2004) seeks to coordinate corporate
acquisition practice. The directive establishes rules similar to those of the Williams Act
to give incumbent management an opportunity to respond to the takeover bid and
shareholders to evaluate the offer (Takeover Directive, Article 6). Besides sharehold-
ers, however, the directive reaches out to employees by requiring the targets board to
communicate the acquirers disclosures and terms of the offer to give employees an op-
portunity to respond to the offer and exercise any rights they may have under national
law. The offer must remain open for at least two weeks (Takeover Directive, Article 7).
Unlike the Williams Act that leaves corporate defensive mechanisms to state law (see
Corporate Takeover Defenses below), Article 9 of the Takeover Directive forbids incum-
bent management from initiating an antitakeover strategy, other than seeking compet-
ing purchase offers, when a bid is in process without an affirmative, contemporaneous
shareholder vote. Similarly, Article 11 of the Takeover Directive nullifies defenses, other
than those adopted by a contemporaneous vote of the shareholders, during the pen-
dency of a bid. Further, once an acquirer owns 75 percent of the target, the acquirer may
call a shareholders meeting on short notice to replace the targets board of directors
even if the targets governing documents otherwise would delay that replacement.
386 p r i vat e e q u i t y : u s e s a n d s t r u c t u r e

C O R P O R AT E C O M B I N AT I O N S : E L I M I N AT I N G M I N O R I T Y
SHAREHOLDERS
Once the acquirer gains control of a U.S. target through its tender offer, it may capture
complete ownership by combining the target with the PE fund corporation. The combi-
nation, usually a merger, freezes out the continuing ownership interest of the remaining
target shareholders. Those shareholders receive cash, debt of the surviving corporation,
or a combination of cash and debt for their target shares. The debt often is subordinate
to the claims of creditors lending money to effect the tender offer and often consists of
below investment grade junk bonds or short-term indebtedness.
Shareholders often were unsatisfied with the merger consideration since it rarely
gave them as much cash per share as the preceding tender offer. By dissenting from the
merger, minority shareholders became entitled to a judicial determination of the value
of their shares under the law of the targets state of incorporation (Del. GCL, section
262). This judicial appraisal remedy did not necessarily provide a more satisfactory
price than the merger consideration that the acquirer offered.
Delaware always has had management favorable corporate laws, so publicly traded
corporations including more than 60 percent of Fortune 500 companies, are incorpo-
rated in Delaware. With so many corporations, Delaware decisional law dominates in
determining appraised value. Until 1983, Delaware exclusively used the Delaware Block
Method of appraisal. The Delaware Block Method required the appraiser to weight
market value, the public trading price, asset value based primarily on book value, which
measures historical cost, and earnings value in determining the price of dissenting
shares. No clear standard was available for weighting these factors and the resulting ap-
praisal was not forward-looking, so the appraisal rarely determined a value equal to the
buyers analysis of the underlying value of the enterprise.
Shareholders sought remedies under the federal securities laws to block the merger
and improve their bargaining power. In 1977, however, the U.S. Supreme Court in Santa
Fe Industries v. Green (1977)19 held the shareholders had no federal remedy under the
securities laws because there was no federal law of corporations. These laws were re-
served to the states. The decision seemed to alert the Delaware Supreme Court to the
inadequacy of state protections of minority shareholders. In the same year as the U.S.
Supreme Court decision in Santa Fe Industries v. Green, the Delaware Supreme Court
held in Singer v. Magnavox (1977)20 that shareholders remedy in a hostile takeover no
longer was only an appraisal of their shares. Any transaction the corporation undertakes
to freeze-out minority shareholders, even a short-form merger, required a business pur-
pose. In evaluating that business purpose, lower courts could fashion relief to achieve a
fair outcome, including blocking or unwinding a merger where appropriate. The threat
of such radical relief chilled the corporate acquisition industry.
In Weinberger v. UOP, Inc. (1983),21 the Delaware Supreme Court retreated from the
open-ended relief possibilities it allowed in Singer v. Magnavox (1977). The court re-
versed its holding that a business purpose was necessary for any transaction eliminating

19
Santa Fe Industries v. Green, 430 US 462 (U.S. 1977).
20
Singer v. Magnavox, 380 A.2d 969 (Del. 1977).
21
Weinberger v. UOP, Inc., 457 A.2d 701 (Del. 1983).
Gl obal Reg ul atory an d E t h ical F ram e work 387

remaining minority shareholders. Instead, the court focused its attention on appraisals,
determining the Delaware Block Method was inadequate for the modern market. The
Delaware Supreme Court held the lower state courts should use the method that best
determines value, while including future growth in the evaluation. Use of methods in-
cluding forward-looking discounted cash-flow valuation was appropriate. In Cede & Co.
v. Technicolor, Inc. (1996),22 the court determined that minority shareholders forced out
in the business combination stage of the takeover were entitled to a value reflecting any
increase in value of the target following the tender offer and the accompanying change
in control. As a result, the acquirers business plan for the target might enhance the value
of minority shares.
In the European Union, Article 5(4) of the Takeover Directive requires the acquir-
ers offer to include all shares of minority shareholders at the highest price per share the
acquirer paid during a fixed period preceding and including its bid. National law deter-
mines that period, which must be at least 6 months and may be as long as 12 months
before the bid. The acquirer has the right to squeeze out all remaining shareholders at a
fair price once the acquirers share ownership reaches a 90 percent threshold. National
law may set the threshold as high as 95 percent (Takeover Directive, Article 15). Mi-
nority shareholders have the right to have the acquirer buy their shares at a fair price
(Takeover Directive, Article 16). The voluntary bid or mandatory offer under Takeover
Directive Article 5(4) is a fair price for purposes of both the squeeze-out and the minor-
ity shareholders right to sell.

C O R P O R AT E TA K E O V E R D E F E N S E S : P O I S O N P I L L S
As the Williams Act gave management an opportunity to respond to tender offers and
influence shareholders to accept or reject the offers, managers developed an arsenal
of private defensive strategies to combat unwanted acquisitions. These strategies are
known as poison pills because they tended to diminish the value of the target. Poison
pills discouraged those offers and suitors that management considered inadequate or
unsuitable.
Some poison pills altered the targets balance sheet both pre- and post-acquisition.
Others limited and delayed the hostile acquirers ability to complete the takeover and re-
structure the target. Alone or in combination with state antitakeover statutes, discussed
in the next section, poison pills give incumbent management broad power to choose
whether a suitor might acquire the target and, when an acquisition does occur, who the
successful suitor will be. Both poison pills and the antitakeover legislation deter hostile
takeovers because they may affect corporate transactions after the unwanted suitors ac-
quisition of a large or controlling position in the target.
Corporations with large amounts of liquid assets and little debt are attractive for
leveraged acquisitions because the acquirer can use the targets liquidity to repay its
acquisition indebtedness following the merger step in the acquisition. The defensive
technique of causing the target to repurchase large numbers of its own shares at a pre-
mium price reduces any cash position and often requires sizable borrowing to fund the
repurchases. Target indebtedness limits the hostile acquirers ability to borrow to fund

22
Cede & Co. v. Technicolor, Inc., 684 A.2d 289 (Del. 1996).
388 p r i vat e e q u i t y : u s e s a n d s t r u c t u r e

the acquisition because the target, already burdened with debt, becomes a doubtful
source of repayment. In Unocal Corporation v. Mesa Petroleum Co. (1985),23 the Dela-
ware Supreme Court not only permitted the share repurchase defense strategy but also
allowed the target to exclude the unwanted suitor from the class of shareholders who
might participate in the share repurchase. The court noted pejoratively that the suitor
was a well-known greenmailer, a term referring to someone who buys a share position in
a corporation and resells the position to the corporation at a premium in order for man-
agement to free itself from the greenmailers interference with corporate management.
The Unocal decision included extensive discussion of the board of directors primary
obligation to protect the corporate enterprise. In this evaluation, the court observed,
the board should consider, among other factors, the takeovers impact on constituen-
cies other than shareholders (i.e., creditors, customers, employees, and perhaps even
the community generally) (Unocal at 955). While the obligation to take other con-
stituencies into account has not gained any major presence in the Delaware decisional
law, it has become a factor in the Minnesota antitakeover law and the AIFM Directive,
both discussed below.
Staggering the dates when board members terms end is another effective method
used to prevent hostile takeovers because it delays the time when the buyer can gain
control of the targets board of directors. A more aggressive technique involves locating
a more desirable suitora white knightand encouraging that suitor to investigate
purchasing the target by offering a large breakup fee. A breakup fee is a cash payment to
compensate the white knight for evaluating the target and offering to buy it if the white
knight does not win the target in a takeover contest. Management also might offer the
white knight an option to buy the targets most desirable assets or line of businessits
crown jewelsat a favorable price.
Acquirers challenges to the defensive techniques failed to gain the support of the
Delaware Supreme Court despite how poison pills can tip the balance in a takeover
contest. In Moran v. Household International, Inc. (1985),24 the Delaware Supreme Court
rejected a challenge to a poison pill giving current shareholders the right to buy an inter-
est in the buyer at a bargain price, so-called flip-over rights, when the buyer acquired a 20
percent of the targets shares or made a tender offer for at least 30 percent.
However, in Revlon v. MacAndrews & Forbes Holdings, Inc. (1985)25 the Delaware
Supreme Court held that once management decides to sell the target corporation, man-
agement must treat all suitors equally. It must give access to corporate information to all
potential buyers and must not favor one buyer over another. Lock-ups on crown jewels
and breakup fees to favor one suitor are impermissible. The best financial offer for the
shareholders determines the buyer.
Incumbent management, through its control of the corporate board of directors,
usually has the power to remove the poison pill if the buyer makes a sufficiently advanta-
geous offer to the target. While its fiduciary obligations to the shareholders restrain the
board from accepting offers without thorough analysis (Smith v. Van Gorkom (1985)26),

23
Unocal Corporation v. Mesa Petroleum Co., 493 A.2d 946 (Del. 1985).
24
Moran v. Household International, Inc., 500 A.2d 1346 (Del. 1985).
25
Revlon v. MacAndrews & Forbes Holdings, Inc., 506 A.2d 173 (Del. 1985).
26
Smith v. Van Gorkom, 488 A.2d 858 (Del. 1985).
Gl obal Reg ul atory an d E t h ical F ram e work 389

the business judgment rule tends to insulate the board from liability for inopportune
decisions if it has evaluated any purchase offers diligently (Grobow v. Perot (1987)27).
The Takeover Directive in the European Union suspends the initiation or operation
of takeover defenses while an offer is pending (Takeover Directive, Articles 9,11). Al-
though a corporations board of directors may structure a corporation to make it un-
attractive as a takeover target, while complying with applicable national law and its
obligations to shareholders, the board may not initiate action during a takeover attempt
as U.S. corporations may. Thus, for example, no barrier exists to maintaining a high in-
debtedness in the corporation, but the corporation may not engage in increased bor-
rowing to frustrate a pending offer.

S TAT E A N T I TA K E O V E R L AW S
Parallel with development of poison pills, states in the United States regulate corporate
acquisitions and, concomitantly, PE funds by enacting antitakeover laws. Those antita-
keover laws shift the advantage in takeover contests to incumbent management and have
nearly eliminated hostile takeover opportunities in the United States (Macey 1988). In
Edgar v. Mite Corp. (1982),28 the U.S. Supreme Court held the Illinois Takeover Act,
one of the early antitakeover statutes, unconstitutional. That Illinois statute applied to
both Illinois corporations and out-of-state corporations. Since Illinois was neither the
state of incorporation nor the state in which the target had its principal office, the statute
impermissibly burdened interstate commerce. Later, however, in CTS Corp. v. Dynamics
Corp. of America (1987),29 the U.S. Supreme Court upheld the Indiana Control Shares
Acquisition Chapter of the Indiana Business Corporation Law against a challenge that
the Williams Act occupied the regulatory field for corporate takeovers and preempted
the Indiana statute. Since the statute applied only to Indiana corporations, it did not
burden interstate commerce impermissibly.
Since the CTS Corp. (1987) decision, most states have enacted some form of anti-
takeover legislation. Change in corporate control may trigger various restrictive pro-
visions of those statutes, including limiting the acquirers right to vote with acquired
shares, delaying any business combination with the acquirer or a related party, or re-
quiring the acquirer to buy all shares at a premium price. Delawares antitakeover law
delays business combinations unless the acquirer either gains the consent of the board
of directors before gaining control, the acquirer owns at least 85 percent of the target,
or, after gaining control, the board and a two-thirds vote of shareholders approves the
combination (Del GCL, section 203). The two-thirds shareholder vote requirement is
a common supermajority requirement in many other states (Missouri Revised Statutes,
section 351.425). Delawares antitakeover law has not been tested, as management of
Delaware corporations generally rely on poison pills to thwart takeovers.
Under most antitakeover statutes, the board of directors may remove the impedi-
ment to the business combination when management views the takeover as sufficiently

27
Grobow v. Perot, 526 A.2d 914 (Del. Ch.1987), aff d, Del. Supr., 539 A.2d 180 (1988).
28
Edgar v. Mite Corp., 457 U.S. 624 (U.S. 1982).
29
CTS Corp. v. Dynamics Corp. of Am., 481 U.S. 69 (U.S. 1987).
390 p r i vat e e q u i t y : u s e s a n d s t r u c t u r e

favorable to current shareholders. The board has a fiduciary obligation to evaluate acqui-
sition offers in the best interests of the shareholders. Minnesota law, however, empowers
directors, possibly at the expense of shareholders, to consider the interests of other con-
stituencies, including employees and the community at large in evaluating any takeover
offer (Minnesota Statutes, section 302A.351, subd. 5).30

T H E A I F M D I R E C T I V E , TA K E O V E R S , A N D P R I VAT E E Q U I T Y
F U N D S U N D E R N AT I O N A L L AW
The AIFM Directive requires member states to designate a competent authority to regis-
ter alternative investment fund managers (AIFMs) consistent with the AIFM Directive.
The member states must impose minimum capital requirements on funds and managers
and ensure the AIFMs meet various operational standards to manage each fund fairly.
AIFMs must perform risk management, as well as portfolio management, and set levels of
leverage that are consistent with the operation of the fund. The directive requires AIFMs
to make annual disclosures to investors on the managed funds, including all fees, and must
report financial information especially about funds use of leverage to national authorities.
Member states must impose disclosure requirements to shareholders and employees
as a PE fund achieves specific levels of ownership of companies that are not publicly
traded (AIFM Directive, Article 27). The Takeover Directive already requires disclo-
sures for publicly traded companies. Most important to the acquisition activity of PE
funds is the moratorium on distributions and share redemptions for 24 months if the
distribution impairs corporate capital following acquisition of control (AIFM Directive,
Article 30). This limit may deter some takeovers of both public and private companies
because the acquirer may use only accumulated profits of the target to repay its acquisi-
tion indebtedness. That limit resembles some state antitakeover laws. One commentator
describes the distribution restriction as quite benign but sees it as creating a potential
conflict of interest between the fund and its manager (Wymeersch 2012b). By limiting
some rapid takeover restructurings, the directive protects certain other constituencies,
especially labor, that rely on the continued and undiminished operation of the enterprise.
The AIFM Directive contemplates that each EU member state will issue rules con-
sistent with the directive. Member states will use their own regulatory authorities to en-
force the rules. In the United Kingdom, which is second only to the United States in the
amount of PE fund activity, the Financial Services Authority regulates all activities in-
volving securities and financial management. Under its statutory authority, it issues rules
consistent with the AIFM Directive for PE funds and their managers (McVea 2012).

I M PA C T S O F T H E W I L L I A M S A C T, T H E TA K E O V E R D I R E C T I V E ,
P O I S O N P I L L S , A N D A N T I TA K E O V E R L AW S
By rejecting challenges to both poison pills and antitakeover statutes, courts have regu-
lated takeovers and the activities of PE funds. PE funds must eschew hostile takeover

30
Minnesota Statutes, 2013, Revisor of Statutes. Available at https://www.revisor.mn.gov/
statutes/.
Gl obal Reg ul atory an d E t h ical F ram e work 391

activity in most instances. Negotiating with incumbent management often generates a


more favorable price for shareholders than the acquirer otherwise would pay (Com-
ment and Schwert 1995) and yields improved corporate information reporting (Arm-
strong, Balakrishnan, and Cohen 2012).
On the other hand, removing the hostile takeover threat entrenches management,
leading to increased incentive compensation for top executives (Cheng and Indjejik-
ian 2009), a decrease in overall productivity (Bertrand and Mullainathan 2003), and
diminution in executives and board members exposure to the enterprises risks in
the form of share ownership (Cheng, Nagar, and Rajan 2005). Compare the view that
the Annex to the AIFM Directive reflects with respect to incentive compensation. Ex-
ecutives who share the risks of the venture through ownership of interests in their
funds are more aligned with the risk management process. Incumbent management,
through its control of the board and corporate communications, may easily (and often
does) capture benefits for itself when it removes takeover barriers for a specific suitor
(Macey 1988).
By neutralizing the coercive nature of the tender offer, the Williams Act and the Take-
over Directive in the European Union gave management an opportunity to respond to
the tender offer and seek a better price for shareholders. The Takeover Directive, like
the Delaware Supreme Court decision in Revlon v. MacAndrews & Forbes Holdings, Inc.
(1985), focuses on assuring that all shareholders receive a fair price for their shares.
Poison pills and antitakeover laws, as well as the 24-month capital impairment and dis-
tribution moratorium in the AIFM Directive, seem to give incumbent management
too much power over acquisitions enabling managers to capture part of the corporate
value for themselves (Macey 1988). Incumbent managements use of its power to block
takeovers remains a critical ethical concern. Whether the shareholders would receive
a better price if the target simply became subject to a bidding contest is often unclear.

Restructuring the Acquired Companies


Once the PE fund has completed its acquisition, it may proceed to restructure the
target company. Since the target is no longer publicly traded, the PE fund will cease
filing public reports under the securities laws and will delist the target from any ex-
change on which its shares trade. Change in control may trigger regulatory compliance
requirements under licenses the company may hold. Similarly, the PE fund may have
to renegotiate contracts and borrowing agreements. Changes in operations that impact
employees may require compliance with union contracts and the intervention of agen-
cies regulating labor relations. Many regulated industries will require the PE fund to
work with the regulator for that industry.

Disposition of Acquired Companies


In restructuring the acquired companies, the PE fund may increase the companys in-
debtedness, replace its key management often paying incumbent management large de-
parture payments, change compensation structures of rank and file employees, relocate
392 p r i vat e e q u i t y : u s e s a n d s t r u c t u r e

the headquarters, and dispose of inefficiently deployed corporate assets. Pre-acquisition


publicly traded corporations may become private corporations, often wholly owned by
the PE fund.
Disposition may involve a sale to some other enterprise with which the target enjoys
synergies that make the combination attractive. In this instance regulation of the dis-
position should be minimal. The PE fund management negotiates a sale price with the
purchasing enterprise and sells the portfolio company. Depending on the terms of the
PE funds organizational provisions, it either distributes the sale proceeds among its
investors or holds the proceeds to fund another acquisition.
PE funds may sell a portfolio company in exchange for shares of the acquiring com-
pany. Often that type of sale will be a tax-deferred corporate reorganization (Internal
Revenue Code, section 368). Since the PE fund is normally tax transparent, it may pass
the buyer shares to its investors in a nontaxable distribution. If the buyer is a publicly
traded corporation and the PE fund receives the same class of shares as those that are in
the public market, investors may sell the acquiring company shares, subject perhaps to
trading limits, into the market without registering the specific shares under the securi-
ties laws (SA, section 4(1)).
Alternatively, the PE fund may make a public offering of ownership in the portfolio
company. The PE fund will register the shares of the target under the securities laws of
the countries in which the public offering will take place. In the United States, registra-
tion involves the formal disclosure of a prospectus approved by the SEC. The PE fund
may retain some shares and distribute those retained shares, a nontaxable distribution,
with the proceeds from the public offering to the PE funds investors.

Ethical Framework
The swarm of locusts analogy used earlier frames the fundamental ethical concern
about the PE fund industry and corporate acquisitions. Some view PE funds as preda-
tory because they acquire successful operating enterprises and use them to reap a wind-
fall profit for PE fund investors at the expense of the enterprise, its workers, and the
community. The windfall arises as PE funds denude the enterprise of its liquid assets
and saddle it with debt, leaving the enterprise weakened and vulnerable to failure in
an economic downturn. PE funds lack ties or commitment to the community and the
workers. Once in control, PE funds may sell or discontinue insufficiently profitable por-
tions of the enterprise without considering the loss of employment for workers and loss
of businesses critical to the communities where they are located. PE funds can be ruth-
less in disposing of less productive assets or even in liquidating an enterprise if the PE
fund can secure an acceptable profit. Like a swarm of locusts, they simply leave waste in
their pathunemployment, vacant business structures, and communities without the
businesses on which they depend.
Others view PE fund takeovers as beneficial. PE funds expose ineffective manage-
ment and root out inefficient economic behavior. Change in ownership and control
most often makes the business more profitable and creates jobs, though possibly at
lower pay levels than pre-takeover. Improved economic activity following the take-
over tends to be beneficial overall but may affect certain constituencies adversely.
Gl obal Reg ul atory an d E t h ical F ram e work 393

Takeovers also improve value to existing shareholders, as they capture part of the con-
trol premium that market trading does not and often part of the premium value to
the PE fund that arises from the PE funds business plan for the enterprise. In a com-
petitive takeover environment, the target enterprise and its shareholders derive the
economic benefit of the acquisition, not the acquirer (Andrade, Mitchell and Stafford
2001; Larcker 2011).

Summary and Conclusions


This chapter reviewed the regulatory frameworks that affect PE funds at each stage of
their operations. Securities and investment company laws confine PE funds selection
of investors, so PE funds need not register under the securities laws and are not subject
to substantive regulation as investment companies. In their acquisition and restructur-
ing of target companies, securities laws define the timing of their acquisitions and the
disclosures they must make, forcing PE funds to give incumbent management the op-
portunity to evaluate takeover offers and solicit competing offers. Antitakeover legis-
lation and court approval of defensive mechanisms regulate the manner in which PE
funds may acquire targets and indirectly influence the prices paid for target companies.
Antitakeover laws and defensive strategies tip the balance to incumbent managements
favor, enabling it to extract maximum value from the takeover for both the sharehold-
ers of the target and the incumbent management. Securities laws also often affect the
manner and structure of PE funds divestitures after they have restructured acquired
companies.
Ethical concerns arise as to PE funds treatment of acquired enterprises, their labor
force, and the communities in which they operate. An increase in an enterprises indebt-
edness in a takeover risks its continuing viability and raises a broader ethical concern in
the systemic risk that arises from the use of heavy leverage for the acquisitions. Ethical
concerns also arise about incumbent managements ability to seize benefits for itself at
the cost of the enterprises shareholders when a PE fund seeks to acquire an enterprise.
Further ethical questions arise from certain poison pill practices. Lock-ups of crown
jewels and break-up fees for failed white knight takeovers transfer corporate value to
outsiders.

Discussion Questions
1. Discuss whether PE funds are investment companies subject to regulation under
the Investment Company Act of 1940 and whether they must register offerings of
their investment interests under the securities laws.
2. Explain the laws and private strategies limiting the success of hostile corporate take-
overs in the United States.
3. Discuss whether the laws and strategies used in the European Union are similar to
those in the United States.
4. A former German minister described PE funds as a swarm of locusts. Interpret the
meaning of his statement and discuss whether that analogy is correct.
394 p r i vat e e q u i t y : u s e s a n d s t r u c t u r e

References
Andrade, Gregor, Mark Mitchell, and Erik Stafford. 2001. New Evidence and Perspectives on
Mergers. Journal of Economic Perspectives 15:2, 103120.
Armstrong, Christopher S., Karthik Balakrishnan, and Daniel Cohen. 2012. Corporate Gover-
nance and the Information Environment: Evidence from State Antitakeover Laws. Journal of
Accounting and Economics 53:1, 185204.
Bertrand, Marianne, and Sendhill Mullainathan. 2003. Enjoying the Quiet Life? Corporate Gover-
nance and Managerial Preferences. Journal of Political Economy 111:5, 10431075.
Cheng, Shijun, and Raffi Indjejikian. 2009. The Market for Corporate Control and CEO Compen-
sation: Complements or Substitutes? Contemporary Accounting Research 26:3, 701728.
Cheng, Shijun, Venky Nagar, and Madhav V. Rajan. 2005. Identifying Control Motives in Manage-
rial Ownership: Evidence from Antitakeover Legislation. Review of Financial Studies 18:2,
637672.
Comment, Robert, and G. William Schwert. 1995. Poison or Placebo? Evidence on the Deterrence
and Wealth Effects of Modern Antitakeover Measures. Journal of Financial Economics 39:1,
343.
Couret, Alain. 2012. France. In Eddy Wymeersch, ed., Alternative Investment Fund Regulation,
111152. Alphen aan den Rijn : Kluwer Law International.
Debevoise & Plimpton LLC. 2014. Client Update: German Insurers and Pension Funds May Effec-
tively Be Banned from Investing in Non-EEA Private Funds. Available at http://www.
debevoise.com/files/Publication/23f41601-41673f0-4ca8-a257-200eeb3ed8cc4/Presentation/
PublicationAttachment/39418642-c1a3-41b9-b994-939b6e2e41e5c/20140624A%20-%20
Proposed%20Amendment%20of%20German%20Insurance%20Regulation.pdf
Easterbrook, Frank H., and Daniel R. Fischel. 1981. The Proper Role of a Targets Management in
Responding to a Tender Offer. Harvard Law Review 94:6, 11611204.
Fleischer, Victor. 2008. Two and Twenty: Taxing Partnership Profits in Private Equity Funds. New
York University Law Review 83:1, 159.
Larcker, David F. 2011. The Market for Corporate Control. Corporate Governance Research
Program, Stanford Graduate School of Business. Available at www.gsb.stanford.edu/cldr/
research/research.html.
Macey, Jonathan R. 1988. State Anti-Takeover Statutes: Good Politics, Bad Economics. Wisconsin
Law Review 1988:3, 467490.
McVea, Harry. 2012. United Kingdom. In Eddy Wymeersch, ed., Alternative Investment Fund Reg-
ulation, 343399. Alphen aan den Rijn: Kluwer Law International.
konomi og erhvervsminitriet. 2006. Kapitalfonde i Danmark, konomisk Tema, Nr. 4. Available at
http://www.evm.dk/publikationer/2006/~/media/oem/pdf/2006/Pressemeddelelser-
2006/Kapitalfonde-pdf.ashx.
Ordower, Henry. 2012. United States. In Eddy Wymeersch, ed., Alternative Investment Fund Regu-
lation, 401431. Alphen aan den Rijn: Kluwer Law International.
Riis, Arne, and Babar Khan. 2014. Denmarks Permanent Establishment Trap for Passive Foreign
Investors. Tax Analysts News Analysis. Available at http://services.taxanalysts.com/taxbase/
tni3.nsf/SearchIndex/581EC05192BDAF9085257CBC00070B15?OpenDocument&
highlight=0,denmark,s.
Seith, Anna, ed., 2005. Gesagt ist gesagt: Wirtschaftweisheiten des Jahres, Spiegel Online. Available
at http://www.spiegel.de/wirtschaft/gesagt-ist-gesagt-wirtschaftsweisheiten-des-jahres-a-
390044-390048.html.
Wymeersch, Eddy, ed. 2012a. Alternative Investment Fund Regulation. Alphen aan den Rijn: Kluwer
Law International.
Wymeersch, Eddy. 2012b. The European Alternative Investment Fund Management Directive. In
Eddy Wymeersch, ed., Alternative Investment Fund Regulation, 433496. Alphen aan den RijnT:
Kluwer Law International.
Part Six

TRENDS IN PRIVATE EQUITY


22
Private Investment in Public Equity
NA DAI
Associate Professor of Finance, SUNY at Albany

Introduction
Private investment in public equity (PIPE) has emerged in the last decade as an impor-
tant source of financing, particularly for companies with high information asymmetry
and weak operating performance (Brophy, Ouimet, and Sialm 2009; Chaplinsky and
Haushalter 2010; Chen, Dai, and Schatzberg 2010). From 1995 to 2012, more than
16,000 U.S. PIPE offeringsclosed or with a definitive agreementoccurred resulting
in more than $400 billion raised.
This chapter provides an overview of the development of the PIPE market, its con-
tractual structure, the role of the placement agent, the effect of enforcement by the Secu-
rities and Exchange Commission (SEC), and recent innovations in the market. The rest
of the chapter is organized as follows. The first section describes the legal background
and unique features of PIPE issuances. The second section discusses the major inves-
tors in the PIPE market. The third section provides details on the contractual structure
of a typical PIPE transaction. The fourth section discusses the services that placement
agents offer to PIPE issuers. The fifth section reviews the SEC enforcements in the PIPE
market since 2002 and discusses the impact of the enforcements on the development
of the PIPE market. The chapter ends with a discussion on the most recent regulations
and innovations in the market.

Overview of the Private Investment in the Public


Equity Market
PIPE securities are typically issued under Regulation D, which allows public companies
to issue stocks privately to accredited investors without the need for public registra-
tion as long as the seller follows a set of listed requirements. Regulation D Rule (501)
defines accredited investors from the following categories: banks; brokers or dealers; in-
surance companies; registered investment companies or business development compa-
nies; small business investment companies; pension funds; directors, executive officers,

397
398 t r e n d s i n p r i vat e e q u i t y

or general partners of the issuer; corporations; limited liability companies, and trusts or
partnerships with total assets more than $5 million not formed for the specific purpose
of acquiring the securities offered. Accredited investors also include any natural person
whose individual net worth, or joint net worth with that persons spouse, when the pur-
chase exceeds $1 million, or income or joint income exceeds $200,000 or $300,000,
respectively, in each of the two most recent years; and any entity for which all equity
owners are accredited investors. The major investors in the PIPE markethedge funds,
venture capital (VC) funds, and private equity (PE) fundsall meet at least one of
these criteria but typically the last.
Before the secondary market sale of securities issued under Regulation D, the issuer
must file a resale registration statement with the SEC. Following the closing of a PIPE
transaction, the issuer prepares and files a resale registration statement with the SEC.
In contrast to a traditional private placement, such a closing does not depend on the
SEC review process, making PIPE issuance a time-efficient mechanism by which small
companies that would have difficulty paying for SEC registration can raise capital. How-
ever, investors cannot resell purchased securities until the SEC declares the registration
statement is effective.
The PIPE market became popular because it solved an important matching prob-
lem: small, badly performing companies in dire need of external financing used PIPEs
to create a match with hedge funds, PE funds, or other investor types wanting to invest
in publicly traded securities (Brophy et al. 2009). Among the advantages of a PIPE
is that the offering can be completed even before the issuer files a resale registration
statement with the SEC, thereby giving issuers faster access to the cash companies so
badly need. Another advantage is that the financial contracting template used in a PIPE
allows for several state-contingent terms, which can be finely tailored to match the par-
ticular needs of a given investment (Chaplinsky and Haushalter 2010; Bengtsson and
Dai 2014).
As Table 22.1 and Figure 22.1 show, the PIPE market has grown from 114 trans-
actions in 1995 to about 1,000 deals in 2012. The total amount of capital raised has
increased 35 fold, from $1.3 billion in 1995 to $46 billion in 2012. The average offer size
has quadrupled, from $12 million to $47 million. In 2008, the total amount of capital
raised through the PIPE market reached $80.8 billion, a size that is comparable to the
seasoned equity offering (SEO) market, which raised $88 billion in the same year. More
companies have been using PIPEs to raise equity capital. For instance, in 2007, more
than 1,000 public companies issued PIPEs. Overall, the PIPE market has become an
important financing alternative for U.S. public companies.
PIPEs are particularly popular among healthcare companies. As Table 22.2 and
Figure 22.2 show, healthcare companies issued 4,562 PIPE deals, which account for
about 28 percent of the total market. The financial sector has been using PIPEs heavily
especially during and after the financial crisis of 20072008. Financial companies raised
$164 billion from 1995 to 2012, with an average offer size of $128 million.
Many variations in security structure exist across PIPE deals. Common stock PIPEs
are dominant in both the number of deals and the total amount of capital raised. Of the
more than 16,000 PIPE deals from 1995 to 2012, about 50 percent were common stock
based. Convertible preferred stock PIPEs account for about 18 percent of PIPE deals
and 28 percent of capital raised.
Pr iv ate In v e s t m e n t in P u bl ic E qu it y 399

Table 22.1 Size of the Private Investment in Public Equity Market, 1995 to 2012
Year Number of PIPE Total Amount of Capital Average Offer Size
Transactions Raised (USD billion) (USD million)
1995 114 1.33 11.67
1996 306 4.13 13.50
1997 456 4.88 10.70
1998 440 3.17 7.20
1999 691 10.62 15.37
2000 1,255 28.21 22.48
2001 1,036 17.38 16.78
2002 755 12.87 17.05
2003 880 12.03 13.67
2004 1,286 14.78 11.49
2005 1,325 18.74 14.14
2006 1,348 24.72 18.34
2007 1,391 58.89 42.34
2008 1,018 80.80 79.37
2009 1,026 29.96 29.20
2010 1,202 9.85 33.15
2011 1,026 38.10 37.13
2012 979 45.97 46.96
Total 16,534 446.43 27.00

1600 90.00
1400 80.00

1200 70.00
60.00
1000
50.00
800
40.00
600
30.00
400 20.00
200 10.00
0 0.00
95
96
97
98
99
00
01
02
03
04
05
06
07
08
09
10
11
12
19
19
19
19
19
20
20
20
20
20
20
20
20
20
20
20
20
20

N of PIPE Transactions Total Amount of Capital Raised ($B)

Figure 22.1 Size of the Private Investment in Public Equity MarketThis


figure presents the volume and the total amount of capital raised through PIPEs in the U.S.
market from 1995 to 2012.
400 t r e n d s i n p r i vat e e q u i t y

Table 22.2 Industry Distribution of Private Investments in Public Equity


Sector Number of PIPE Total Amount of Capital Average Offer Size
Transactions Raised (USD billion) (USD million)
ConsumerNon- 4,562 68.57 15.03
Cyclical (Healthcare)
Communications 2,681 55.22 20.60
Industrial 1,844 25.96 14.08
Technology 1,837 22.95 12.49
Energy 1,308 52.35 40.02
Financial 1,277 163.75 128.23
Other 1,944 35.37 18.19

Consumer -Non-Cyclical (Healthcare)

Industrial

Technology

Energy

Financial
0 1,000 2,000 3,000 4,000 5,000
Number of PIPE Transactions

Financial

Communications

Energy

Industrial

Technology
0 50 100 150 200
Total Amount of Capital Raised ($Billion)

Figure 22.2 The Distribution of Private Investments in Public Equity by


Industry These two figures present the top six industries in terms of the number of PIPE
transactions and the total amount of capital raised in billions of dollars.

A structured equity line enables a public company to periodically sell at the issuers
election a certain dollar amount or number of shares of its common stock directly to a
private investor by exercising drawdowns. The issuer is under no obligation to sell any
shares under the equity line. A formula tied to the market price of the shares over a spec-
ified pricing period determines the purchase price for the shares in a particular draw-
down in which the firm requests the investors to pay the committed capital. Previously,
a structured equity line had a tarnished reputation as toxic PIPE because of faulty con-
tract design. For instance, issuers did not have price protection via a floor price allowing
investors to manipulate the issuers stock price. This situation has changed considerably
since the SEC investigation in 2002 and is discussed later in the chapter. Table 22.3 and
Table 22.3 Security Structure of Private Investments in Public Equity, 1995 to 2012
Year Common Stock Convertible Preferred Stocks Common Stock Shelf Structured Equity Lines Common Stock + Warrants Other
N Amount N Amount N Amount N Amount N Amount N Amount
(USD billion) (USD billion) (USD billion) (USD billion) (USD billion) (USD billion)
1995 48 0.58 24 0.21 0 0.00 0 0.00 7 0.03 35 0.51
1996 87 1.34 120 1.24 0 0.00 1 0.05 10 0.04 88 1.46
1997 117 1.58 209 2.06 1 0.00 8 0.14 6 0.02 115 1.08
1998 128 0.98 205 1.49 1 0.00 12 0.17 8 0.04 86 0.49
1999 298 3.56 209 5.43 3 0.04 15 0.36 34 0.21 132 1.02
2000 542 11.43 286 9.29 11 0.09 149 3.88 95 0.62 172 2.90
2001 427 6.8 214 4.6 58 0.96 144 2.83 44 0.20 149 1.99
2002 306 3.73 145 4.24 58 1.20 65 0.85 34 0.12 147 2.73
2003 433 4.98 155 2.96 80 1.89 32 0.42 44 0.22 136 1.56
2004 576 6.73 194 2.85 77 1.40 76 1.13 83 0.46 280 2.21
2005 500 7.03 202 3.48 91 2.01 86 1.83 57 0.42 389 3.97
2006 497 13.04 161 2.90 95 2.84 81 2.09 116 0.93 398 2.92
2007 558 25.62 165 20.27 100 3.61 50 1.17 145 1.07 373 7.15
2008 332 9.82 149 45.34 87 3.31 42 1.08 73 0.61 335 20.64
2009 255 7.6 139 7.73 186 2.93 77 1.05 75 0.52 294 10.13
2010 327 13.43 136 5.50 169 2.19 129 2.05 78 0.54 363 16.14
2011 274 7.84 105 2.92 94 1.19 98 1.45 90 0.70 365 24.00
2012 215 4.66 96 4.94 88 1.83 90 1.04 67 0.33 423 33.17
402 t r e n d s i n p r i vat e e q u i t y

100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
Common Stock Common Stock + Warrants Common Stock -Shelf
Convertible Preferred Stocks Structured Equity Lines Other

Figure 22.3 Security Structure of Private Investments in Public


Equity This figure shows the relative popularity of different security structures in the
PIPE market over time including common stocks, common stocks with warrants, shelf sale,
convertible preferred stocks, and structured equity lines.

Figure 22.3 show the volume of structured equity lines has rebounded since 2003 and
has become increasingly popular since the financial crisis of 20072008.
Similar to the public equity market, the PIPE market was hit by the financial crisis
of 20072008. The volume of PIPEs declined by 25 percent in 2008 compared to
2007 as shown in Table 22.4. The dollar amount increased because several financial
companies use PIPEs to raise capital. In 2008, the financial sector alone raised $58
billion, accounting for more than 70 percent of the total proceeds raised. Companies
initiated 12 transactions in which the offer size exceeded $1 billion with 11 of these
transactions issued by financial companies. The sizes of these PIPE transactions
were unusually large compared to other years. In fact from 1995 to 2012, only 36
PIPEs occurred with offer size exceeding $1 billion, one-third of which took place in
2008. In 2009, the number of transactions remained about 25 percent less compared
to 2007. Further, the amount of capital raised also declined to less than half of the
amount in 2007.
Bengtsson and Dai (2014) document that during the financial crisis of 20072008,
investors in the PIPE market requested more contractual protections when they pro-
vided capital to companies of similar quality before the crisis period. However, they
do not find significant difference in pricing as measured by the discount. This finding
is consistent with the notion that during a tough financing environment, investors have
greater negotiation power and request more protections against risks.

Private Investment in Public Equity Investors


In terms of capital invested, the major investors in the PIPE market are hedge funds,
pension and government funds, corporations, mutual funds and institutional advisors,
buyout firms, and PE firms, VC firms, brokers and dealers, banks, insurance compa-
nies, and charitable, educational, and family trusts. As Figure 22.4 and Table 22.5 show,
Pr iv ate In v e s t m e n t in P u bl ic E qu it y 403

Table 22.4Private Investments in Public Equity in the Financial Sector,


1995 to 2012
Year Number of PIPE Total Amount of Capital Average Offer Size
Transactions Raised (USD billion) (USD million)
1995 9 0.02 2.67
1996 9 0.94 104.37
1997 11 0.29 26.12
1998 15 0.43 28.64
1999 29 0.60 20.65
2000 29 0.99 34.29
2001 55 1.79 32.49
2002 35 0.74 21.05
2003 54 1.77 32.70
2004 59 1.01 17.09
2005 90 2.24 24.94
2006 61 3.47 56.92
2007 65 26.53 408.14
2008 123 58.20 473.18
2009 122 8.47 69.41
2010 206 21.53 104.54
2011 148 18.28 123.53
2012 132 16.21 122.84
Total 1,252 163.52 130.61

hedge funds dominated the PIPE market until 2008. Since 2009, VC firms and corpora-
tions have played a more important role. According to Anderson and Dai (2014), PIPE
issuers with strategic investors such as corporations and VC funds often offer investors
more control rights but less superior cash-flow rights than those with financial inves-
tors such as hedge funds. This finding is consistent with the different investment objec-
tives of these two groups of investors. Strategic investors are often involved in actively
monitoring management following a PIPE issuance and often strive to keep their invest-
ment in the PIPE issuer for a relatively long time horizon. In contrast, financial investors
prefer short-run cash profits from PIPE transactions and are relatively more passive in
nature even when they hold block stakes.
Anderson and Dai (2014) document that firms financed by strategic PIPE inves-
tors significantly outperform those with financial investors. For instance, measured by
equally weighted market-adjusted cumulative abnormal returns (CARs), PIPE issuers
financed by strategic investors outperform issuers financed by financial investors by
11 percent at 6 months, 21 percent at 12 months, and 28 percent at 24 months following
a PIPE offering.
404 t r e n d s i n p r i vat e e q u i t y

Figure 22.4 Investors in the Private Investment in Public Equity


Market This figure shows the relative market share of different institutional investors in
the PIPE market over time.

Table 22.5 Total Amount in USD Billion Invested by Investor Type, 1995 to 2012
Year Corporations Hedge Fund Mutual Venture Capital Others
Managers Funds and Buyout Funds
1995 0.25 1.24 0.18 0.07 1.47
1996 0.69 15.49 5.00 0.91 19.30
1997 2.61 20.19 10.16 2.16 21.14
1998 0.32 7.52 0.79 0.82 3.34
1999 1.55 21.73 3.67 11.09 23.93
2000 7.30 55.14 24.41 13.62 38.32
2001 5.83 35.60 5.44 6.41 18.86
2002 5.15 31.07 8.68 4.75 14.24
2003 4.85 57.97 7.76 8.90 25.74
2004 4.08 126.72 7.68 3.54 27.06
2005 5.73 111.37 12.35 5.32 29.96
2006 8.57 148.44 11.38 4.98 41.11
2007 21.48 244.08 42.50 12.46 200.50
2008 634.28 696.97 560.26 121.08 709.69
2009 24.22 5.30 4.67 213.89 29.54
2010 100.80 32.88 3.29 209.35 20.01
2011 83.82 39.51 0.50 159.87 27.55
2012 46.83 23.42 2.06 5.84 13.84
Pr iv ate In v e s t m e n t in P u bl ic E qu it y 405

The Contractual Structure of Private Investment


in Public Equity Deals
PIPE issuers and investors have many choices about security types: plain vanilla
common stock, common stock resets, common stock shelf sales, company installment
convertibles, fixed-price convertibles, floating price convertibles, convertible resets, and
structured equity lines. Traditional PIPEs do not allow investors to adjust purchase or
conversion prices including plain vanilla common stocks, common stock shelf sales, and
fixed-price convertibles. In contrast, PIPEs that give investors the right to reset prices
are called structured PIPEs including floating price convertibles and resets.
In most PIPE offerings, issuers offer a discount on the price of publicly traded shares
to PIPE investors as compensation for the due diligence costs and the illiquidity risk
they accept, as investors cannot resell PIPE shares to the public before the SEC declares
the associated registration statement effective. Chaplinsky and Haushalter (2010) show
that PIPE investors mainly obtain terms that allocate supra cash-flow rights, while only a
small percentage (8 percent) involves management or investor board participation. This
finding is in contrast to venture capitalists that make heavy use of control rights (Kaplan
and Strmberg 2003, 2004),
PIPE contracts also typically involve various state-contingent contractual terms. Over-
all, substantial variation exists in how terms are included in PIPE contracts. The common
structure of PIPE contracts is investors get superior rights if a company performs poorly.
As company performance improves, an investor must give up these superior rights.
Many contract terms are designed to overcome agency and information problems. This
structure is not surprising given that these problems are particularly severe for the types
of companies that issue PIPEs. For presentation purposes, these terms are grouped into
three categories: (1) investor protections, (2) trading restrictions, and (3) issuer rights.
Following sections described the individual contractual term in more detail.

I N V E S TO R P R OT E C T I O N S
The first categoryinvestor protectionscontains terms that attach various protec-
tions to a PIPE investors stock. Terms in this category are favorable to investors at the
expense of issuers.

Dividend, Interest, and Warrants


Dividends, interest, and warrants are key deal features in PIPE offerings. Dividends
are sometimes cumulative, which means that an investor does not receive regular divi-
dends, but rather a lump sum at a later time. Many PIPE contracts include warrants that
allow investors to purchase predetermined numbers of certain securities at specified
prices in the future.

Investor Registration Rights


PIPE offerings allow firms to close offerings before filing registration statements with
the SEC making PIPE offerings more time-efficient than traditional SEOs. However,
406 t r e n d s i n p r i vat e e q u i t y

this solution means that investors must assume the illiquidity risk because they cannot
resell the acquired securities before the registration statement becomes effective. To
mitigate this risk, PIPE contracts often include investor registration rights that force
issuers to file registration statements within a short time period after offer closings. In
some cases, such a contract places a cap on the amount of capital an issuer can draw
down before the registration statement takes effect. Some contracts include penalty
terms if registration fails, such as canceling the financing.

Anti-Dilution Protection
Anti-dilution provisions protect PIPE investors against future financing at lower valua-
tions than that of current offerings. In its harshest form, anti-dilution provisions ban an
issuer from issuing or selling any equity securities or securities convertible into equity
during a certain period after a PIPE offering. Such a contract could also ban an issuer
from issuing or selling securities at a price below what the PIPE investor paid or below
a specified benchmark price. In a less harsh form, anti-dilution terms protect investors
from future price decreases by reducing offer prices (or, alternatively, conversion prices)
to equal the lowest prices paid for any equity securities in future financing. In such a
case, an investor could also have the right to receive cash or additional common shares
without additional consideration.

Right of First Refusal and Investor Call Options


Investor call options and right of first refusal give investors the right to purchase addi-
tional shares of a companys security during a certain period in the future. Like war-
rants and anti-dilution provisions, these contract terms are in place to protect investors
against future dilution from price decreases or equity offerings at below-market prices.

Redemption Rights
Optional redemption is sometimes used to strengthen the liquidation rights of an in-
vestment. This protection gives investors the right to demand that a firm redeem their
claim upon a change of control. The conversion price is typically set at face value or at
a certain percentage above face value plus the value of any accrued unpaid interest. Re-
demption rights matter because PIPE contracts often do not specify any contracted pay-
ments on which the issuer can default. Redemption rights may thereby offer investors
the only available means by which to force an issuer to repay an investment.

TRADING RESTRICTIONS
Many PIPE offerings include provisions restricting how investors can trade the underly-
ing stock for a certain period after offer closing. The most common trading restrictions
ban investors from engaging in short transactions, hedging a companys common stock,
or taking a position that is more than the value of shares owned (i.e., an offsetting long po-
sition before the effectiveness of the registration statement). Sometimes, a contract also
requires investors not to engage in shorting or hedging for a longer period than the SECs
requirement, sometimes as long as the purchased PIPE security remains outstanding.
An additional trading restriction applies a lock-up period to a PIPE transaction. Lock-
up periods ban investors from selling any shares of their common stocks purchased or
Pr iv ate In v e s t m e n t in P u bl ic E qu it y 407

received through the exercise of warrants for a duration typically lasting a few months
following the closing. In rare cases, PIPE contracts prohibit investors from affecting any
sales to the public of a companys shares for a certain number of days after the registra-
tion statement takes effect. This restriction is useful if a company plans a public offering
(i.e., an SEO) shortly after closing the PIPE offering because it avoids price pressure
from investors resale of shares to the public.

ISSUER RIGHTS
The third categoryissuer rightscontains terms that give an issuer the right to force
an investor to take certain actions. The terms in these latter categories favor issuers at
the expense of investors.

Company-Forced Conversion
PIPE contracts sometimes include company-forced conversion options. These options
convert shares held by investors into common stock under certain conditions, typically
related to issuers stock performance during a stated period following a PIPE offering.
For instance, investors may have to convert their shares if the stocks price or weighted
average stock price exceeds a certain benchmark price. In an alternative formula, issu-
ers may have to convert their shares if daily trading volumes exceed certain levels for a
specified number of consecutive trading days. The effect of company-forced conversion
provisions is to require investors to give up their contractual protections when compa-
nies get a desired level of performance.

Company Put Option and Optional Redemption


Some PIPEs include company optional redemption provisions, which provide issuers
with the right to force PIPE investors to exercise redemption rights after a certain date
or upon certain events. The effect of put options and optional redemption is that an in-
vestor would receive less favorable cash flow rights if a company were to achieve a strong
performance.

The Role of the Placement Agent in the Private


Investment in Public Equity Market
The PIPE advisory market is fragmented. Placement agents compete against each other
for advising companies that seek to issue a PIPE. The group of agents includes large in-
vestment banks, specialized corporate finance advisory firms, and boutique advisors
who specialize in PIPEs and other types of private placements.
The relationship between the issuer and the placement agent may start in one of
two ways. One possibility is that the agent contacts the issuer with a proposal of a
PIPE offering. The agent can identify prospective clientsstruggling companies in
dire need of capitalusing information from SEC filings and news reports. Another
possibility is the relationship starts with the issuer contacting the agent for advice
on a planned offering. The relationship between the issuer and the agent begins with
408 t r e n d s i n p r i vat e e q u i t y

informal discussions that quickly evolve into a formalized relationship. The mile-
stone for this formalization is when the parties sign a private placement agreement
(sometimes also called an engagement letter), which is the legal document specifying
the responsibilities of each party. Under the agreement, the issuer allows the agent
to be the exclusive advisor for the offering, except in deals where more than one ad-
visor is engaged. The issuer also promises to give the agent access to the companys
detailed financial reports, management team, auditor, legal advisor, and consultant.
This agreement allows the agent to prepare the material that is needed for pitching
the deal to potential investors. To safeguard this confidential information, the agent
agrees to owe duty of confidence to the issuer.
The agent promises the issuer to use best efforts to identify potential investors and
solicit offers from them. In most PIPE offerings in the United States, the agent does not
promise to underwrite the offering or purchase any shares for its own account as after-
market support. Hence, no firm commitment exists from the agent to invest its own
capital in the deal.
The agent takes a leading role in initiating, progressing, and closing the PIPE deal.
The agent controls the interactions among issuer, investor, law firm, auditor, and other
parties. This role differs from what advisors do such as PE in which the lead investor is
more in charge of the transaction.
The first task of the agent is to conduct its business and financial due diligence on
the issuer. In some instances, the agent draws on this information to prepare a formal
private placement memorandum that will be circulated to interested investors. In other
cases, the agent only prepares an informal teaser that summarizes the main items of
the company. Importantly, these documents do not reveal the identity of the issuer. This
anonymity is a safeguard against any investor trying to exploit the information about the
PIPE offering to trade in the issuers stock.
The second task for the agent is to identify a group of investors interested in the deal.
Here the agent faces a delicate trade-off. If the agent solicits the deal to too few investors,
then the offering could fail. In contrast, if the agent solicits the deal to too many investors,
then the SEC could view the PIPE as being a general solicitation of the sale of securities,
which is illegal. This illegality follows from the fact the SEC requires much more exten-
sive reporting requirements and scrutiny for a deal with general solicitation such as SEOs.
The third task, which some view as the hardest, is to bring the deal over the wall,
which is industry jargon to identify when the investor expresses serious interest in the
PIPE offering. Once the deal is over the wall, the agent will reveal the identity of the
issuer and give the investor more detailed financial and operational information. Before
this disclosure, the agent must secure a promise that the investor will not disclose the
information and avoid trading the issuers shares. These promises could be made orally,
or in a non-disclosure agreement and a non-trading agreement.
The fourth task of the agent is to advise the issuer on how to price the PIPE offering
and which contract terms to use. Bengtsson and Dai (2014) show that agents fulfill this
task in a way that benefits the issuer. The agent provides advice to the issuer both before
and during negotiating the deal structure that takes place with the investors. In particu-
lar, the agent explains the meaning of the typical PIPE contract terms. Because many of
the terms are esoteric in the sense that they are used mainly in PIPE offerings, the issuer
may not understand what it gives up by agreeing to a particular term.
Pr iv ate In v e s t m e n t in P u bl ic E qu it y 409

Another part of the advising function is to explain the payoff effects of each nego-
tiable term. Calculating payoff effects can be very complicated. Each term has conse-
quences that depend on the state of the world that is realized, the investors choice to
use the right implied by the term, and which other terms are included in the contract. In
practice, even expert agents may be unable to accurately calculate payoff implications,
but their experience with the contract terms means that they are better equipped than
their issuer clients.
The private placement agreement outlines how the agent will be compensated. The
main part of the compensation is a fee. As is typical in investment banking transac-
tions, the advising agent receives full compensation for its expenses related to prepar-
ing materials, registration costs, legal fees, and related costs. The fee is expressed either
as a percentage of the offer size (i.e., amount of gross proceeds of the offering) or as a
fixed dollar amount. Huang, Shangguan, and Zhang (2008) and Dai, Jo, and Schatzberg
(2010) study these fees and find that the mean and median fee to the placement agent is
about 6 percent of the offer size. The issuer typically pays fees at closing of the PIPE of-
fering, but sometimes the agent negotiates a good faith retainer fee that the issue pays
shortly after signing the agreement.
Besides the fixed fee, agents can receive compensation that is tied to contingent out-
comes. The simplest contingent compensation consists of the agent receiving warrants
of the same type issued to the PIPE investor. Using warrants to compensate the agent
ensures that it has an incentive to structure the deal in such a way that the issuers share
price will rise in the future. A more complex and commonly used form of contingent
compensation is to give the agent a tail. The agent then receives the right to receive a
fee from future financing events, in particular other PIPEs that occur within a short time
period. Alternatively, the agent may negotiate a right of first offer or a right of first refusal
to be the advisor or investor in the issuers future offerings. Although these features do
explicitly tie the agents compensation to performance, they ensure the agent has an
incentive to structure the deal so the issuer survives, which is not obvious given the pre-
carious state of the typical PIPE issuer.
According to Dai et al. (2010), some of the placement agents are well-known names
in the IPO and SEO underwriting business such as Citigroup and UBS. Others, such as
Coastline Capital Partners, Halpern Capital, and ThinkEquity Partners are less familiar
and are specialized players in this market. They also relate PIPE agents to the Carter and
Manaster (C&M) ranking, which is commonly used to represent the participation and
reputation of IPO/SEO underwriters and typically ranges from 1.1 to 9.1. The most
reputable underwriters get a C&M ranking of 9.1. Among the 215 PIPE placement
agents in their sample between 1996 and 2005, 121 agents have C&M ranking with a
mean (median) ranking of 5.4 (5.1). A total of 20 placement agents have a C&M rank-
ing of at least 8.1. In very rare cases (only 1.5 percent), PIPE placement agents are the
issuers IPO underwriters or previous SEO underwriters.
Bengtsson and Dai (2014) measure a PIPE placement agents reputation based on
market share. They calculate agent market share by comparing an agents PIPE volume
(in dollars) in the three preceding years to the total volume of intermediated PIPE of-
ferings during the same period. To measure the reputation and its stability over time,
they further count the number of times the agent was on the top 15 lists over the sample
period of 1999 to 2012. The 10 agents that appear on the annual top 15 lists most often
410 t r e n d s i n p r i vat e e q u i t y

Table 22.6 Ranking of Placement Agents by Market Share


Agent Name Number Sum of Gross Proceeds C&M
of PIPEs (USD million) Ranking
Goldman, Sachs & Co. 17 23,092 9
J.P. Morgan Chase & Co. 37 8,175 9
Credit Suisse Securities (USA) LLC 27 5,007 9
UBS Investment Bank 34 3,801 8
Citigroup Global Markets, Inc. 17 3,773 9
Rodman & Renshaw, LLC 183 2,912 2
Lehman Brothers, Inc. 48 2,670 8
Morgan Stanley 10 2,314 9
Banc of America Securities LLC 50 1,613 9
Bear, Stearns & Co. Inc. 12 809 9

Source: Bengtsson and Dai (2014).

are designated as reputable agents. Table 22.6 shows that most of these agents have
a C&M ranking of 8 or 9, which is typically regarded as reputable agents in the IPO
literature.
The literature documents several important roles of placement agents in the PIPE
market, conditional on their reputation and expertise. For instance, Dai et al. (2010)
examine the certification role of placement agents and examine how their reputation
affects the PIPE transactions discounts, agent fees charged, and information symmetry
of the firm before and after the PIPE offering. They find that reputable placement agents
are associated with larger offers and firms with less risk. More reputable agents offer
higher quality services in that their deals are priced at lower discounts and improved
information symmetry. Nevertheless, more reputable agents do not charge a price pre-
mium. Rather than fees per se, the quality of the issuing firm and the reputation of the
placement agent are the key factors that drive the equilibrium in the PIPE market.
Bengtsson and Dai (2014) examine the role of placement agents in PIPE contract ne-
gotiations. They document that issuers advised by more reputable agents provide investors
with greater contractual protections than do those advised by less reputable agents. More-
over, reputable agents allow issuers to extract more compensation (a lower discount) in ex-
change for investor-friendly contract terms. Also, both agent ranking and more contractual
protections for investors are associated with stronger (i.e., less negative) long-run stock
performance following PIPE offerings. The authors contend that PIPE investors are famil-
iar with complicated contract designs and can correctly understand the consequences of
their typically esoteric terms, while PIPE issuers that are often small, distressed companies,
suffer from bounded rationality on their ability to decipher and evaluate contract terms.
In such a contracting environment, placement agents play an important role by
bridging the contract knowledge gap between these two parties knowledge about con-
tracts. This agent role allows contracts to include more cash-flow contingencies, features
that many contract-theoretical models viewed as optimal. Therefore, reputable agents
Pr iv ate In v e s t m e n t in P u bl ic E qu it y 411

provide higher quality services, because they can more credibly and accurately send val-
uable information to their clients.

Securities and Exchange Commission


Enforcement Actions
Responding to concerns about the propriety of so-called death-spiral PIPE transac-
tions entered by hedge funds (Hillion and Vermaelen 2004), the SEC initiated a series
of enforcement actions (Sjostrom 2007), which began with its investigation of Rhino
Advisors, a hedge fund, and its President Thomas Badian, in June 2002 (Sedona Corp.
v. Ladenburg Thalmann & Co.1 and SEC v. Rhino Advisors and Thomas Badian2). In No-
vember 2000, Amro (a Rhino Advisors hedge fund participating in the PIPE trans-
action at issue) and Sedona entered an agreement granting Amro debentures and
warrants with conversion rights tied to the volume-weighted average price of Sedona
stock in the five days before conversion in exchange for Sedona shares. From March
to April 2001, Badian allegedly shorted Sedona stock to benefit Amro in violation of
the explicit anti-shorting provision in the PIPE agreement, driving the price of Sedona
stock from $1.43 per share on March 1 to less than $0.76 per share by April 5. As a
result, Rhino and Badian consented to an injunction for violations of the anti-fraud
provisions of federal securities laws and a $1 million civil penalty (SEC v. Rhino Advi-
sors and Thomas Badian, 2003).
This SEC complaint was an issue of first impression, meaning that no court had
addressed similar claims in the past, and the investigation and resulting litigation
called for increased SEC enforcement efforts on PIPE transactions in business
media. By early 2003, the SEC had publicly heeded these calls. Civil litigation sur-
rounding the underlying transaction (conducted by the SEC and private parties)
remains unresolved as shown in Amro International, S.A. v. Sedona Corp.3 and SEC
v. Badian.4
The Rhino Advisors case initiated a flurry of investigations into short sales surround-
ing PIPE offerings that resulted in civil suits filed by the SEC. The intent behind the
SECs enforcement actions was to reduce opportunities for investors to manipulate
stock prices, a strategy that seemed especially attractive because of financial contracts
that gave investors aggressive repricing rights (i.e., structured PIPEs). However, believ-
ing that proving intent to manipulate prices in court would be difficult, the SEC in-
stead chose to crack down on a specific mechanismshort sellingthat could be used
for price manipulation. The SEC argued that in some cases such PIPE-related trading
violated the Securities Exchange Acts of 1933 and 1934, specifically the regulations on
insider trading and the sale of unregistered securities. The SEC also argued that short

1
Sedona Corp. v. Ladenburg Thalmann & Co., Complaint, Civ. Action No. 03-civ-3120 (S.D.N.Y.
filed May 5, 2003).
2
SEC v. Rhino Advisors, Inc. and Thomas Badian, Civ., Action No. 03 civ 1310 (RO) (S.D.N.Y. 2003).
3
Amro International, S.A. v. Sedona Corp., Slip Copy, 2010 WL 2813452 (S.D.N.Y. 2010).
4
SEC v. Badian, 2010 WL 1028256 (S.D.N.Y. 2010).
412 t r e n d s i n p r i vat e e q u i t y

selling could be outright fraudulent if a contract involved expressed or implied promises


to take a long position on a given stock.
The SEC has not been successful defending its theories when bringing some of the
civil enforcement actions to a federal district court. In fact, the SEC lost all four cases
reaching a final judgment. The only successful prosecution of a case arose from the
claims put forth in SEC v. Berlacher.5 The court found that Berlacher and co-defendants
had committed fraud, but only for instances in which they represented themselves as
having held no short position in the PIPE issuers securities while they did in fact hold
such a position. The court found the defendants guilty of insider trading and associated
(shoehorned) fraud, as well as materiality.
Despite the lack of success in court, the SECs enforcement actions made investors and
issuers pay closer attention to the SECs arguments about PIPEs. In response to the SECs
actions, PIPE investors began to avoid risk mitigation through aggressive repricing rights
and short selling, which the SEC targeted, while pursuing such risk mitigation through
other (non-targeted) contractual rights. For instance, Bengtsson, Dai, and Henson (2014)
document a strong shift from structured PIPEsfeaturing investor-friendly terms that
grant investors conversion rights that adjust in investor-friendly waysto traditional
PIPEs that have less-onerous conversion rights following the SECs enforcement initiative.
Also, they document a significant change in contractual structure in PIPE transactions.
In general, a marked increase occurred by restricting investors from trading in
a companys stock during the critical stages of a PIPE offering. At the same time, in-
cluding investor protections increased while that of issuer rights decreased. PIPE con-
tracts changed partly because investors who continue to be active following the SECs
enforcement initiatives changed their behavior and partly because new investors used
contracts that differed from those of the old investors they replaced. One effect of the
SECs enforcement action was to deter, at least partly, investors accustomed to heavy
contract terms from continuing in the PIPE market.
The previously mentioned contractual changes happened across all investor types
(i.e., hedge funds, mutual funds, corporations, PE funds, VC funds, and brokers), but are
more pronounced among PIPEs associated with hedge funds because the SEC clearly
targeted this investor type. Further, hedge funds demanded more favorable pricing than
other investor types in the post-action period. This finding is consistent with the argu-
ment that changes in PIPE contract design associated with the SECs actions were more
costly for hedge funds, which called for more attractive pricing as compensation. Even
though most of the SECs actions formally targeted hedge funds or individuals affiliated
with hedge funds, enforcement did not deter this investor type from later participating
in the PIPE market. As Bengtsson et al. (2014) document, 531 hedge-fund-led PIPEs
existed in the pre-enforcement period (1999 to 2002) and 758 in the post-enforcement
period (2003 to 2006). Hedge-fund-led PIPEs represented 45 percent and 67 percent,
respectively of all PIPE transactions in these periods.
Another effect that the SECs actions brought to the PIPE market is the increased
use of placement agents particularly when they were marketing PIPEs to hedge funds.
A possible explanation of this result is that the SECs litigations persuaded issuers that
they needed expert advice to guide them through PIPE contract negotiations. Another

5
SEC v. Berlacher, 2010 WL 3566790 (E.D.Pa. 2010).
Pr iv ate In v e s t m e n t in P u bl ic E qu it y 413

potential reason is as new terms became more prevalent, issuers may have sought expert
advisors to explain such changes.
Some changes also appear in the characteristics of PIPE issuers in the post-action
period. Bengtsson et al. (2014) find that companies that issued PIPEs during the period
20032006 are smaller, have less analyst coverage, lower pre-announcement CARs,
more intangible assets, and lower valuations. This pattern shows that issuers had overall
weaker characteristics in the post-action period compared with the pre-action period.
Given that PIPEs are an important financing tool for small, struggling, cash-starved
companies, this change suggests that the SECs actions had no major adverse effects for
issuer participation in the PIPE market.

New Regulations and Innovations in the Private


Investment in Public Equity Market
In 2005, the SEC passed the Securities Offering Reform (Reform), which relaxes re-
strictions on firms disclosures before public equity offerings and simplifies the regis-
tration procedure. The new rules specify three major groups of issuers with different
obligations and benefits: well-known seasoned issuers (WKSI), seasoned issuers, and
unseasoned issuers. WKSI are those issuers with market capitalization of $700 million
or more. Under the new rules, an issuer that qualifies as a WKSI is eligible to conduct an
offering for any security on an automatically effective shelf registration statement. Sea-
soned issuers are those with public float of at least $75 million. These issuers are eligible
for primary offering on Form S-3 or Form F-3. Unseasoned issuers are those with public
float of less than $75 million. These issuers were ineligible for primary offering on Form
S-3 or Form F-3 before the 2008 amendments.
In May 2008, the SEC adopted amendments to allow smaller public company issuers
access to the public securities markets that have traditionally been limited to only the
larger companies. Before adopting the amendments, for an issuer to meet the trans-
action requirement to register its securities in a primary offering for cash (excluding
an offer of nonconvertible investment-grade securities), the issuer had to have a public
float of at least $75 million. The 2008 amendments expanded the list of eligible transac-
tions, which permits issuers with a public float of less than $75 million to qualify for
Form S-3 or Form F-3 eligibility for primary offerings up to one-third of their public
float in one or more direct offerings over a rolling calendar year.
With the opportunity to conduct registered primary offerings, unseasoned firms will
have greater access to the public markets. Before the 2008 amendments, many unsea-
soned firms elected to raise financing through PIPEs. Under the new regime, once a
shelf registration statement is declared effective by the SEC, these firms could capitalize
on favorable market conditions through one or more takedowns without requiring fur-
ther SEC review or approval.
As Table 22.7 and Figure 22.5 show, since the 2008 Form S-3 amendments, a large
increase is observed in SEOs by small firms with market capitalization less than $75 mil-
lion. At the same time, the volume of common stock PIPEs by small firms has declined
since 2007.
414 t r e n d s i n p r i vat e e q u i t y

Table 22.7 Alternative Financings for Small Firms, 2011 to 2012


Year SEO Common Stock PIPE Registered Direct CMPO
2001 2,697.03 6,578.34 933.97 0.00
2002 3,562.39 3,502.62 1,125.14 0.00
2003 2,013.7 3,791.45 935.96 0.00
2004 4,109.34 5,468.52 1,107.15 30.00
2005 3,890.98 3,690.19 950.43 0.00
2006 3,914.00 3,742.54 1,284.93 0.00
2007 4,583.90 3,861.48 1,251.03 0.00
2008 1,447.48 1,907.11 672.14 40.80
2009 8,420.22 1,776.19 1,288.75 769.71
2010 8,338.88 1,807.97 994.44 1402.38
2011 6,998.57 2,299.13 613.38 1223.46
2012 11,233.80 1,101.29 453.05 1741.06

Note: Small firms are public firms with market capitalization less than $75 million.

Figure 22.5 Alternative Financings for Small Firms: Unseasoned


Issuers This figure shows the amount of capital ($millions) raised by small firms through
different financing mechanisms such as SEO, PIPE, registered direct (RD), and CMPO.
Small firms are public firms with market capitalization less than $75 million at the time of
financing.

In response to the regulatory changes, several innovations have emerged in the


equity offering market over the past a few years. One such innovation is the at-the-
market (ATM) offering. An ATM offering enables a public company to sell, periodi-
cally during the term of the facility, registered equity securities directly to the public
through a broker-dealer in amounts and at times and prices at the companys discretion.
The company is under no obligation to sell any shares under the ATM offering. Strictly
Pr iv ate In v e s t m e n t in P u bl ic E qu it y 415

speaking, an ATM offering is not a private placement because the company is selling
registered shares to the public directly. However, Sagient Research, an institution that
specializes in collecting PIPE transaction data, includes ATM offerings under the PIPE
structure.
The advantages of the ATM offering include the following:

The impact of offering on issuers stock price is smaller. Companies do not have to
disclose the precise timing of issuances under an ATM offering at the commencement
of the program. ATM stock issuances generally are not publicly announced at the time
of sale. Instead, companies disclose these sales at the end of each fiscal quarter. This
timing consideration minimizes possible price pressure at the time of issuance.
The company has greater flexibility with regard to the timing and the amount of issu-
ance. The company can raise the capital when it is needed or when the stock is priced
in favor to the company.
The cost of capital is lower than a traditional SEO and a typical PIPE. The
broker-dealer commissions in an ATM offering typically range from 2 to 5 percent of
the gross sales price of the securities. No discounts apply to market price, and no
warrants exist that are typically expected in a typical PIPE.

Billett, Floros, and Garfinkel (2014) empirically compare ATM offerings and SEOs
and report the issuance cost of an ATM offering is lower by anywhere from 25 percent
to more than 50 percent than a SEO. They further find that announcement of the ATM
program is accompanied by a significantly less negative abnormal return (3.7 percent)
compared to the announcement of an SEO (4.2 percent) on average.
Given these advantages, the ATM market is growing quickly. As Table 22.8 shows, a
total of 326 ATM arrangements had been made between 2009 and 2012 through which
the issuers had the opportunity to raise $47 billion at the time of their choice. For the
same period of 2009 to 2012, the average offer size ($143 million) is much greater than
that of PIPEs ($37 million), and smaller than traditional SEOs ($205 million).
Another innovation is the so-called a confidentially marketed public offering (CMPO).
CMPOs are sometimes referred to as wall-crossed, pre-marketed, or overnight of-
ferings. Regardless of the terminology used, the key feature of the CMPO is that it in-
cludes a public offering phase following a private marketing phase to a selected group of
accredited institutional investors. During the confidential phase, investors have to agree

Table 22.8 The Emergence of at-the-Market Offering, 2009 to 2012


Year Number of Deals Total Amount Raised Average Offer Size
(USD billion) (USD million)
2009 30 3.9 130.00
2010 77 8.35 108.44
2011 97 14.1 145.36
2012 122 20.27 166.15
Total 326 46.62 143.01
416 t r e n d s i n p r i vat e e q u i t y

to keep negotiations secret and not to trade on the information outside of the negoti-
ated offer. After a potential investor is brought over the wall by agreeing to keep the
information about the offering confidential, the potential investor can receive material
non-public information about the offering. Once the underwriter has received a sign
of interest from these potential investors, the issuer will announce the offering to the
public at the closing of the market and will file selling documents. The underwriters will
then market the offering more broadly to institutional and retail investors. This public
phase will continue into the evening and the pricing will be finalized and announced
before the market opens the next day. A CMPO is typically priced at a discount com-
pared to the current market price. Its public phase is similar to a traditional SEO.
CMPOs have certain advantages over traditional SEOs and standard PIPEs. For in-
stance, CMPOs are executed very quickly, often just days in comparison to a fully mar-
keted SEO that often takes weeks or months. The confidential feature of the first phase
also allows the issuer to test investors appetite for its security and negotiate the pric-
ing and terms without risking a potential decline in stock price due to speculative trad-
ing before the announcement. Additionally, since the shares sold in CMPOs are freely
transferrable, no illiquidity premium exists that investors usually request in a standard
PIPE.
Like a registered direct offering, issuers need to have an effective registration state-
ment to conduct a CMPO. Although registered direct offerings share many advantages
with CMPOs, such as the confidentiality and the liquidity of shares offered, CMPOs
allow issuers to avoid the compliance of the 20 percent rule since it includes the public
offering phase. The 20 percent rule requires an issuer to get approval from shareholders
when it sells 20 percent or more of its common stock at a discount to the market price.
As Table 22.9 shows, the volume of CMPOs has greatly increased since 2009. By
2012, the number of CMPOs almost doubled. The average offer size is about $44 mil-
lion. The average pricing discount is about 5.4 percent, which is much lower than what is
documented for a typical PIPE offering. The average dilution effect (i.e., shares issued/
shares outstanding after the offering) is around 28 percent, showing that issuers are
taking advantage of CMPOs to avoid complying with the 20 percent rule.

Table 22.9The Emergence of Confidentially Marketed Public Offerings,


2009 to 2012
Year Number Total Amount Raised Average Offer Size Average Average
of Deals (USD billion) (USD million) Dilution (%) Discounts (%)
2009 54 2.50 46.31 24.84 7.84
2010 93 4.37 47.00 23.77 5.58
2011 91 3.91 42.99 33.36 2.67
2012 105 4.21 40.14 31.01 6.15
Total 343 15.00 43.73 28.50 5.43

Note: Dilution is measured as the ratio of the new shares issued in the CMPO to the total shares
after the offering. Discounts are measured as the percentage difference between closing price and
purchase price.
Pr iv ate In v e s t m e n t in P u bl ic E qu it y 417

Summary and Conclusions


The development of the PIPE market is shaped by at least three factors: (1) the fund-
ing needs of small firms and distressed firms that traditionally have difficulty accessing
the SEO market, (2) investors preference for a risk premium especially among hedge
funds, and (3) regulators concerns about market efficiency and fairness. This chapter
describes how these three forces have evolved and affected the PIPE market.
One major advantage of the PIPE is that the offering can be completed before the issuer
files a resale registration statement with the SEC, which gives issuers quick access to the
cash they need so badly. The customized contracts allow for state-contingent terms which
can be tailored to match the needs of both investors and issuers. These features allow PIPEs
to solve an important matching problem: small public companies in dire need of external
financing and investors such as hedge funds and PE funds seeking a high risk premium.
The PIPE market has seen the increasing usage of placement agents in transactions.
The placement agents not only help to match PIPE issuers and investors but also to
bridge the knowledge gap involving the complicated contractual terms between the
usually savvy institutional investors and the relatively less informed PIPE issuers. The
increasing usage of placement agents helped the maturing of the PIPE market.
The problems with the death-spiral PIPEs in the late 1990s raised concerns among
the regulators on whether certain investors illegally took advantage of issuers and the
SEC initiated a series of enforcement actions. These actions have had profound effects
on various aspects of the PIPE market. Traditional PIPEs have become predominant
and contracts have included more protections for both issuers and investors. These
changes also have attracted more strategic investors to the PIPE market.

Discussion Questions
1. Describe a PIPE offering.
2. Explain how a PIPE offering differs from a SEO and why PIPEs have become an al-
ternative to SEOs.
3. Identify the major investors in the PIPE market and how their objectives affect PIPE
deal structure.
4. Identify the three categories of contingent contractual terms in a typical PIPE of-
fering, give an example for each category, and explain how that term affects the cash
flow or control rights of investors and issuers.
5. Discuss the role that placement agents play in a typical PIPE deal.
6. Define a CMPO and discuss its major advantage relative to a traditional PIPE and SEO.

References
Anderson, Christopher W., and Na Dai. 2014. Investor Objectives and Financial Contracting: Evi-
dence from the PIPE Market. Working Paper, SUNY at Albany.
Bengtsson, Ola, and Na Dai. 2014. Financial Contracts in PIPE Offering: The Role of Expert Place-
ment Agents. Financial Management, 43:4, 795832.
418 t r e n d s i n p r i vat e e q u i t y

Bengtsson, Ola, Na Dai, and Clifford Henson. 2014. SEC Enforcement into the PIPE Market: Ac-
tions and Consequences. Journal of Banking and Finance 42:2, 213231.
Billett, Matthew T., Ioannis V. Floros, and Jon A. Garfinkel. 2014. At the Market (ATM) Offerings.
Working Paper, Iowa State University.
Brophy, David J., Paige P. Ouimet, and Clemens Sialm. 2009. Hedge Funds as Investors of Last
Resort. Review of Financial Studies 22:2, 541574.
Chaplinsky, Susan, and David Haushalter. 2010. Financing under Extreme Uncertainty: Contract
Terms and Returns to Private Investments in Public Equity. Review of Financial Studies 23:7,
27892820.
Chen, Hsuan-Chi, Na Dai, and John D. Schatzberg. 2010. The Choice of Equity Selling Mecha-
nisms: PIPEs versus SEOs. Journal of Corporate Finance 16:1, 104119.
Dai, Na, Hoje Jo, and John D. Schatzberg. 2010. The Quality and Price of Investment Banks Ser-
vice: Evidence from the PIPE Market. Financial Management 39:2, 585612.
Hillion, Pierre, and Theo Vermaelen. 2004. Death Spiral Convertibles. Journal of Financial Econom-
ics 71:2, 381415.
Huang, Rongbing, Zhaoyun Shangguan, and Donghang Zhang. 2008. The Networking Function
of Investment Banks: Evidence from Private Investment in Public Equity. Journal of Corporate
Finance 14:5, 738752.
Kaplan, Steven, and Per Strmberg. 2003. Financial Contracting Meets the Real World: An Em-
pirical Analysis of Venture Capital Contracts. Review of Economic Studies 70:2, 281316.
Kaplan, Steven, and Per Strmberg. 2004. Characteristics, Contracts, and Actions: Evidence from
Venture Capitalist Analyses. Journal of Finance 59:5, 21772210.
Sjostrom, William K., Jr. 2007. PIPEs. Entrepreneurial Business Law Journal 2:1, 381413.
23
Listed Private Equity
DOUGLAS CUMMING
Professor and Ontario Research Chair, York University, Schulich School of Business

GRANT FLEMING
Partner, Continuity Capital Partners

SOFIA A. JOHAN
Adjunct Professor, York University, Schulich School of Business

Introduction
This chapter analyzes the capital-raising activities of companies that invest in other
companies (i.e., financial intermediaries that provide capital to non-listed companies).
Private equity (PE) funds are known within financial markets as financial intermediar-
ies that provide much needed capital and value added services largely to private com-
panies. Due to the risky nature of the investments made by PE funds, they are typically
limited to raising their funds on the private placement market through limited partner-
ship vehicles with sophisticated investors such as large institutional investors and high
net worth individuals as limited partners (LPs). As sophisticated investors, investors in
limited partnerships are accorded benefits of negotiated covenants between fund man-
agers and LPs that best suit the needs of the transacting parties, tax, and market condi-
tions at the time of fundraising.
Further, limited partnerships are autonomous and enable fund managers to invest
over the long term for 10 to 13 years without interference from shareholders. Incentives
are aligned in a limited partnership with payment schedules that commonly afford fund
managers with a fixed fee of 2 percent of committed capital and a carried interest fee
of 20 percent. Given the private nature of PE, why PE funds would want to be pub-
licly listed and deal with the range of regulatory and public disclosure requirements on
public stock exchanges seems puzzling. What are the motivations for investing in listed
PE given that private placements via limited partnerships are prevalent? Whatever the
motivations of PE funds, for the retail market, access to the investment capabilities of
PE funds has been welcomed. Retail investors can mimic the portfolios of more sophis-
ticated investors, seek higher returns, and diversify their equity portfolios to access parts
of the economy unavailable on public stock exchanges.

419
420 t r e n d s i n p r i v a t e e q u i t y

Not all listed PE, however, is created equally (Bergmann, Christophers, Huss, and
Zimmerman 2010; Cumming, Fleming and Johan, 2011; Huss and Zimmerman 2013).
The two types of listed PE are the listed PE firm and the listed PE fund. Investors who
select listed PE firms are essentially provided an opportunity to benefit from the success
of the PE firm that manages PE funds. Profits are derived from management fees and
carried interest earned by the investment professionals and managers of the PE firm.
An example of a listed PE firm is the Blackstone Group (NYSE: BX) that listed on June
22, 2007. Investors who select listed PE funds can benefit from the profits earned from
the investments made by the listed fund. Such investments allow investors who would
otherwise be unable to invest in a traditional PE limited partnership to gain exposure to
a portfolio of PE investments. Examples of listed PE firms are KKR Private Equity Inves-
tors on the Euronext in Amsterdam (ENXTAM: KPE) on April 18, 2006, and Fortress
on February 7, 2009. Investors need to appreciate the difference between the two types
of listed PE.
Based on an analysis of 79 ordinary funds and 21 listed PE funds from 1992 to 2008,
Lahr and Kaserer (2010) find that listed PE funds start at an initial premium of 2.5
percent and adapt to the long-term average of -21 percent after two years. They note
that premia predict future returns and are explained by liquidity but not by investor
sentiment or the funds degree of investment. Instead, PE fund premia depend on credit
markets and systematic risk.
Jegadeesh, Krussl, and Pollet (2009) examine the performance of PE fund-of-funds
into unlisted PE funds and compare the performance to listed PE funds. Based on data
from 26 PE fund-of-funds and 129 listed PE funds from 1994 to 2008, the authors es-
timate the markets expectation of unlisted PE funds (via fund-of-funds) abnormal re-
turns (and net of their fees) to be 1 to 2 percent above the market accounting for risk,
while the markets expectation for listed PE abnormal returns is zero to slightly negative.
They find the betas of listed PE and unlisted PE (via fund-of-funds) to be close to one.
PE fund returns are positively correlated with gross domestic product (GDP) growth
and negatively correlated with credit spread.
Godineni and Megginson (2010) contend that listed PE became relatively more
popular in 2007 due to market changes and the need for fund managers to seek alterna-
tive ways of raising capital. They argue that listed PE affords insiders an alternative way
to cash out, subject to share price performance. Figure 23.1 shows Blackstones 2007
IPO performance is consistent with these rationales in its first year since its IPO, as
performance figures for one month, six months, and one year following the IPO were
18, 27, and 42 percent, respectively, compared to contemporaneous S&P returns of
3, 1, and 16 percent, respectively. Fortresss IPO shows similar poor performance
illustrated in Figure 23.1. The market downturn that started in August 2007 in part ac-
counts for these figures, particularly as Blackstones beta is 2.04 and Fortresss beta is
2.18. Both betas are sourced from Google finance in July 2014. As of July 2014, Black-
stones share price had recovered to its previous price in 2007, but Fortresss share price
has not recovered.
The performance of listed PE has markedly improved since the financial crisis of
20072008 leading up to 2014, despite the early poor performance of some newly listed
PE funds previously mentioned. The ability to raise capital continues to grow. As of
2014, Figures 23.2 and 23.3 show more than 200 listed PE funds worldwide.
L is t e d P riv at e E qu it y 421

Figure 23.1 Post Initial Public Offering Performance of Blackstone and


Fortress, 2007 to 2014 This figure shows the indexed performance of Blackstone
and Fortress from the time of their IPOs in early 2007 ( June 21, 2007, for Blackstone and
February 9, 2007, for Fortress) to July 31, 2014, relative to the S&P 500 Index.

Series1, Series1,
Asia Pacific, Rest of World, Series1,
5%, 5% 2%, 2% Europe,
30%, 30%

Series1, Series1,
North America, United Kingdom,
29%, 29% 34%, 34%

Figure 23.2 Worldwide Distribution of Listed Private Equity Companies,


December 2013 This figure shows the worldwide distribution of listed PE companies in
2013. Source: http://www.lpeq.com.

More recent performance statistics in Figure 23.4 show that listed PE does better
than the S&P 500 index and the MSCI Europe Index but is more volatile. Beta values
exceed 1 with the beta values greater than 2 for PE funds such as Blackstone.
Potentially important benefits of listed PE include providing access to the PE asset
class with improved liquidity and offering lower transaction costs. Institutions can
invest in both listed and limited partnership PE and can dynamically adjust exposure to
listed PE over time as their limited partnerships drawdown commitments. For smaller
422 t r e n d s i n p r i v a t e e q u i t y

Figure 23.3 Listed Private Equity Market Capitalization, 1993 to 2013 This
figure shows the market capitalization (vertical axis in billions of Euros) of listed PE
worldwide from 1993 to 2013. Source: http://www.lpeq.com.

Figure 23.4 Listed Private Equity Absolute Return Performance, 2004 to


2014 This chart shows the absolute return performance of listed PE from 2004 to 2014.
Source: http://www.lpeq.com.
L is t e d P riv at e E qu it y 423

institutions, listed PE enables them to meet diversification requirements with lower


transaction cost and liquidity. In view of the growth of popularity of listed PE, this chap-
ter seeks to discover who invests in listed PE and their rationale for doing so. This issue
is explored for the first time with a new data set, as presented in Cumming, Fleming and
Johan (2011).
The data are derived from a 2008 survey of 100 institutional investors in Europe, com-
pleted by the Listed Private Equity Association (LPEQ), the trade body for European
listed PE. This survey provides unique details about institutional investors allocations
into listed PE, as well as demographic characteristics such as size, investor type, location,
decision-making authority, and liquidity preference. This chapter explores three motiva-
tions to invest in listed PE based on search costs (size, investor type, and location), spe-
cific human capital (decision-making level), and liquidity-time preferences (the desire to
achieve full exposure to PE as soon as possible). The results show that listed PE is more
commonly used among institutions that are smaller, private (not public) pension institu-
tions, those that prefer liquidity and institutions that are based in the United Kingdom.
Institutions are also more likely to invest in listed PE when investment decision-making
is not empowered to a PE team, an alternative asset class team, or a board/investment
committee, but are more likely when decision-making is delegated to an equities team.
This chapter is organized as follows. In the first part, the related literature is discussed
and the testable hypotheses are developed. Thereafter, the chapter introduces the data
and provides summary statistics. Multivariate analyses of institutional investors deci-
sions to invest in listed PE are provided after describing the data. Concluding remarks
follow in the last section.

Literature and Hypotheses


This section explores the motivations of an institutional investor to invest in listed PE.
Hypotheses are based on the premise that differences in institutional investor alloca-
tions to listed PE are mainly a function of investor characteristics. Specifically, three pri-
mary motivations exist to invest are associated with five characteristics: (1) search costs
associated with investor size, type, and location, (2) specific human capital associated
with the investors decision-making structure, and (3) liquidity-time preferences asso-
ciated with the desire to achieve exposure to PE as soon as possible. The next section
outlines each hypothesis with relevant literature.

SEARCH COSTS
Empirical studies on PE and leverage buyout (LBO) risk and return show that return dis-
persion is relatively high, with mixed findings on the average and median return depend-
ing on controls for the extent of unexited investments. Some studies even suggest that
after controlling for unexited investments, the average and median manager provide net
after fee returns to investors less than public markets. For example, Woodward and Hall
(2003), Woodward (2004), Cochrane (2005), and Caselli, Gatti, and Perrini (2009)
examine venture capital risk and returns, Jones and Rhodes-Kropf (2003), Ljungqvist
and Richardson (2003), Kaplan and Schoar (2005), Wright, Weir, and Burrows (2007),
424 t r e n d s i n p r i v a t e e q u i t y

Diller and Kaserer (2009), and Martynova and Renneboog (2011) study buyout risk
and returns. Jegadeesh et al. (2009) examine the performance of PE fund-of-funds into
unlisted PE funds and compare their performance to that of listed PE funds. Based on
data from 26 PE fund-of-funds and 129 listed PE funds between 1994 and 2008, they
estimate the markets expectation of unlisted PE funds (via fund-of-funds) abnormal
returns (and net of their fees) to be 1 to 2 percent above the market accounting for risk,
while the markets expectation for listed PE abnormal returns is zero to slightly negative.
They also find the betas of listed PE and unlisted PE (via fund-of-funds) to be close to
one. PE fund returns are positively correlated with GDP growth and negatively cor-
related with credit spread. An important difference in the data set used in this chapter
differs from that used by Jegadeesh et al. (2009). They use the Venture Capital Trusts
(VCTs) data set. The data set in this research relies on LPEQ, which excludes VCTs as
part of the definition of listed PE. VCTs are tax-subsidized funds with major statutory
covenants that arguably lower their performance, and most investors would not invest
except for the tax subsidy. Cumming (2003) and Cumming and MacIntosh (2007) pro-
vide related work on analogous tax-subsidized listed PE funds in Canada.
Consistent with the work of Jegadeesh et al. (2009), empirical studies consistently
show PE investment is a specialized asset class involving high information and search
costs compared with listed equities. As a result, an investors willingness to incur search
costs (e.g., information collection and access to high-quality managers) to generate risk-
adjusted excess returns above those of public markets will be associated with the size
of the investor, type, and ease of access such as that facilitated by locational advantages.
First, regarding size, smaller institutions do not have time and possibly the experience
or skills to incur high search costs in identifying high-quality managers or to negotiate
limited partnership contracts (Cumming and Johan 2006). Smaller institutions may
also face limited access to more reputable PE funds with existing institutional investors
investing in multiple fund vintages within the same PE firm, which is a major disadvan-
tage in view of evidence of significant performance persistence among PE funds. The
idea of significant performance persistence among PE funds is important for limited
partnerships and listed PE funds alike, and shows that manager selection is critical.

Hypothesis 1A: Listed PE is a more attractive asset class for smaller institutional investors.

Search costs are a function of location. The data, described in the next section, comprise
institutional investors from the United Kingdom and continental Europe. The United
Kingdom has the most liquid stock market, and home bias is widely documented,
thereby leading to the expectation that greater institutional investor interest exists in
listed PE (Suh 2005). Additionally, London-based broker research analysts primarily
cover the listed PE sector and provide insightful research to their clients. Further, exist-
ing evidence is consistent with the view that English legal origin countries afford greater
protection to investors for publicly listed companies (La Porta, Lopez-De-Silanes, Shle-
ifer, and Vishny 1998).

Hypothesis 1B: Listed PE is more commonly considered by institutional investors based


in the United Kingdom than their counterparts in continental Europe.
L is t e d P riv at e E qu it y 425

Search costs should differ depending on institutional investor type. Unobserved char-
acteristics of the institutional investors in the data, as described in the next section, are
likely to differ depending on their type. Therefore, the empirical analyses control for
institutional investors type.

S P E C I F I C H U M A N C A P I TA L
Besides search costs, the investors stock of organizational human capital and
decision-making process can have a pronounced impact on an institutional investors
interest in investing in listed PE. As noted previously, the identification and manager
selection process in PE is relatively costly and successful manager selection can in-
crease the probability of generating excess returns. Indeed, differences exist in institu-
tional investors ability to select successful investment managers (Lerner, Schoar, and
Wongsunwai 2007).
An institutions human capital endowment (i.e., its investment team) can be organ-
ized in several ways. Early adopters of PE have built up specialized investment teams
over time, possessing the skills and industry networks to access private placements
with the best managers (Lerner et al. 2007). These private equity teams or alterna-
tive assets teams improve the ability of the institutional investor to implement its PE
program through unlisted vehicles. By contrast, PE selection and implementation could
also be handled by a more generalized equities team, as opposed to a PE team. Equities
teams are trained and experienced in stock picking, and thereby have less time, experi-
ence, and skill to carry out due diligence and negotiate contracts for limited partner-
ships. PE and alternative asset teams are comparatively more experienced in sourcing
and evaluating limited partnership deals and negotiating and writing limited partner-
ship contracts. While these teams focus on direct PE investment, they also use listed PE
as an investment tool. Additionally, these teams have more expertise than an equities
team to assess listed PE portfolios and manager selection processes.

Hypothesis 2: Listed PE is more likely to be considered by an institutional investor


whose decision-making is delegated to an equities team as compared to delegation to a
PE team or an alternative asset team.

Besides distinguishing between equities teams versus PE teams, the data set further
considers whether decisions are centralized and made by a board/investment commit-
tee. Investment preferences of board/investment committees are likely most directly
influenced by the experience of the committee members. While experience of com-
mittee members is not reported in the data, this research does observe institutional
investor investment committee structure and therefore controls for this aspect of the
decision-making process.
In the decision-making, consultants are often used in selecting PE funds (i.e., an
external influence on internal decision-making). They tend to restrict their advice on
unlisted funds to manager/GP selection and partnership terms. Often, the same GP/
manager will offer a listed vehicle that might suit clients who seek greater liquidity or
a smaller minimum commitment. This type of arrangement may become increasingly
common as defined benefit schemes, which can make sizable long-term commitments,
426 t r e n d s i n p r i v a t e e q u i t y

are replaced by defined contribution schemes for which providing alternative assets is
problematic unless listed vehicles are used. Consultants have reported to LPEQ that
choosing a listed PE vehicle involves an element of stock selection rather than solely
manager selection. Consultants rarely provide information on listed PE, although ev-
idence from the LPEQ survey shows many of their clients desire more information on
listed PE. The empirical analyses that follow consider the effect of consultants.

LIQUIDITY-TIME PREFERENCE
Most institutions with plans to diversify into alternative asset PE will have in place, with
their investment mandates, specific allocation amounts budgeted over a 2-to-5 and even
10-year horizon. These targets have to be met to ensure alignment among all other asset
class allocations. Unfortunately, PE investment through private placements has a dis-
tinct disadvantage of taking many years for an investor to achieve the desired exposure
level (i.e., capital invested). An initial issue is being invited by the right funds to make
a placement. However, in the postfinancial crisis period since July 2007, this issue is
less pressing. As unproven investors, a few vintages may be necessary before the PE
firms recognize an institution as a value-adding investor. Once reputation has been es-
tablished, the second issue is the capital funding duration. Investors commit capital to a
fund, but capital is rarely drawn down on completion of legal documentation. The fund-
ing obligation (the commitment) is drawn down (or called) by the PE firm when
required to complete new investments in companies over the investment period, which
is typically five years from the beginning of the fund.
A PE firm typically may call 75 to 80 percent of capital committed to its fund over the
first five years and reserve the remaining commitment to finance follow-on investments
in companies and management fees over the next five years of the fund. Therefore, in-
vestors are required to pay their commitment over a 10-year period, but it is called un-
evenly during these years.
As a result, the amount of capital committed to PE by an investor and the amount
invested in PE-backed companies differs, and investors can take many years to achieve
their desired level of exposure to private companies (Takahashi and Alexander 2002;
Cumming, Fleming, and Suchard 2005). Indeed, the modeling of drawdowns shows
that PE fund investment activities vary according to supply of investible opportunities,
competition for deals, and cost of financing especially for buyouts (Gompers and Lerner
1999; Ljungqvist and Richardson 2003). Similarly, distributions back to investors from
PE firms after the sale of a portfolio company are dependent on the state of public fi-
nance markets and the economy. As Takahashi and Alexander note, expectations about
drawdown and distribution rates influence the investors capital commitment decisions.
Listed PE provides two advantages to an investor, especially related to private place-
ments. First, an investor can achieve relatively rapid exposure to PE through listed ve-
hicles. Second, maintaining a listed PE exposure alongside private placements provides
a dynamic adjustment mechanism for an investors overall PE exposure. For example,
the investor could start at 100 percent listed PE and then reduce listed PE exposure as
limited partnership exposure increases. The use of listed PE in this way reveals an inves-
tors liquidity-time preferences: the willingness to trade-off exposure tomorrow from
private placement with exposure/flexibility today through listed vehicles.
L is t e d P riv at e E qu it y 427

Hypothesis 3: Listed PE enables institutional investors to achieve their target PE alloca-


tions quicker, and as such, institutions that invest in listed PE are more likely to adjust
their listed PE allocations over time in response to slower adjustments to limited part-
nership PE allocations.

Besides these institutional investor motivations for investing in listed PE, consider-
ation is given to other demographic factors provided by the LPEQ survey. The data set
and empirical analyses follow in the next sections.

Data and Summary Statistics


LPEQ, the trade body for European listed PE, commissioned the data in this study, sub-
sequently called the LPEQ data. Preqin, a leading source of information for the alterna-
tive assets industry, conducted telephone interviews in October 2008. LPEQ selected
Preqin to conduct the survey due to its extensive research among LPs. The sample com-
prises a representative array of 100 European institutional investors that do and do not
use listed PE. A pilot study of 21 LPs led to refining the questions and interviews. The
final sample of 100 represents the range of European countries, sizes, and types of in-
stitutions. Table 23.1 defines the variables in the LPEQ data with summary statistics
illustrating the breadth and depth of coverage of institutions. The confidential details
gathered in the data are not publicly available, and as such the data afford a unique and
invaluable look at institutional investors portfolio allocation decisions.
Among the 100 institutional investors, 43 have listed PE in their investment mandate.
Nineteen investors have a variable listed PE allocation over time. A variable listed PE allo-
cation enables adjustment of the institutional investors portfolio allocation over time as,
for example, drawdowns increase among limited partnership PE allocations. Nine investors
indicate that one of their limited partnership PE fund managers managing a listed PE fund
influenced their decision to invest in listed PE. For all investors, regardless of whether they
have listed PE in their investment mandates or have in fact invested in listed PE, the average
amount invested in listed PE relative to limited partnership PE is 2.78 percent. For inves-
tors who have listed PE in their investment mandate, the average allocation to listed PE rel-
ative to limited partnership is 7.58 percent. For all investors in the sample, the percentage
of investments into listed PE relative to total assets under management is 3.92 percent and
6.29 percent for the subset of investors with listed PE in their investment mandate.
The average institutional investor in the sample manages more than 57.7 billion,
with the median at 4.8 billion. Panel B of Table 23.1 shows that the sample of 100
institutional investors comprises 1 investment bank, 1 endowment, 9 family offices,
14 banks, 22 private pension funds, 23 insurance companies, and 30 public pension
funds. As Panel C of Table 23.1 shows, 26 investors are based in the United Kingdom
and the others are in continental Europe. Decision-making for listed PE investments
is carried out by a PE team among 24 institutional investors and by an alternative asset
team among 12 institutional investors as shown in Panel D of Table 23.1. The board/in-
vestment committee makes PE investment decisions for 38 institutional investors, and
the equities team by contrast makes PE investment decisions for 2 of the institutional
investors. Other institutional investors used an undefined or unspecified process. For
Table 23.1 Definitions and Summary Statistics
Definition Mean Median Standard Min Max
Deviation
Panel A Dependent Variables
Listed PE part A dummy variable equal to 1 if listed PE is part 0.43 0.00 0.50 0.00 1.00
of investment mandate of the institutional investors investment mandate,
and 0 otherwise
Listed PE is a variable A dummy variable is equal to 1 if the institutional 0.19 0.00 0.39 0.00 1.00
allocation over time investors target listed PE allocation is variable
over time, and 0 if it is a stable target
Listed PE investment A dummy variable equal to 1 if the institutional 0.09 0.00 0.29 0.00 1.00
decision depends on investors decision to consider listed PE is positively
whether fund also has affected by whether a fund manager in which they
428

a private PE fund invest also has a listed PE fund


Percentage of listed PE The percentage of investments in listed PE versus 2.78% 0.00% 10.43% 0.00% 60.00%
versus other PE other PE
Percentage of listed PE The percentage of investments into listed PE relative 3.92% 2.90% 4.95% 0.00% 30.00%
relative to total assets to total assets under management
Difference between The difference between the target PE allocation 0.32% 0.00% 1.02% 2.30% 5.00%
percentage of target and actual PE allocation
PE allocation and actual
PE allocation
Article I. Panel B. Institutional Investor Characteristics
Assets under management Assets under management (Euro million) 57,681.40 4,765.80 279.289.00 37.76 2,700,000.00
Investment bank A dummy variable equal to 1 for an investment 0.01 0.00 0.10 0.00 1.00
bank institutional investor, and 0 otherwise
Table 23.1continued

Definition Mean Median Standard Min Max


Deviation
Bank A dummy variable equal to 1 for an bank institutional 0.14 0.00 0.35 0.00 1.00
investor, and 0 otherwise
Family A dummy variable equal to 1 for a family investor, and 0 0.09 0.00 0.29 0.00 1.00
otherwise
Endowment A dummy variable equal to 1 for an endowment 0.01 0.00 0.10 0.00 1.00
institutional investor, and 0 otherwise
Public pension fund A dummy variable equal to 1 for a public pension fund 0.30 0.00 0.46 0.00 1.00
institutional investor, and 0 otherwise
Private pension fund A dummy variable equal to 1 for a private pension fund 0.22 0.00 0.42 0.00 1.00
429

institutional investor, and 0 otherwise


Insurance company A dummy variable equal to 1 for an insurance company 0.23 0.00 0.42 0.00 1.00
institutional investor, and 0 otherwise
Article II. Panel C. Location
United Kingdom A dummy variable equal to 1 for an institutional investor 0.26 0.00 0.44 0.00 1.00
based in the United Kingdom, and 0 otherwise
Switzerland A dummy variable equal to 1 for an institutional investor 0.14 0.00 0.35 0.00 1.00
based in Switzerland, and 0 otherwise
Denmark A dummy variable equal to 1 for an institutional investor 0.04 0.00 0.20 0.00 1.00
based in Denmark, and 0 otherwise
Netherlands A dummy variable equal to 1 for an institutional investor 0.12 0.00 0.33 0.00 1.00
based in The Netherlands, and 0 otherwise
continued
Table 23.1continued

Definition Mean Median Standard Min Max


Deviation
Finland A dummy variable equal to 1 for an institutional investor 0.08 0.00 0.27 0.00 1.00
based in Finland, and 0 otherwise
Germany A dummy variable equal to 1 for an institutional investor 0.12 0.00 0.33 0.00 1.00
based in Germany, and 0 otherwise
Liechtenstein A dummy variable equal to 1 for an institutional investor 0.01 0.00 0.10 0.00 1.00
based in Liechtenstein, and 0 otherwise
Sweden A dummy variable equal to 1 for an institutional investor 0.06 0.00 0.24 0.00 1.00
based in Sweden, and 0 otherwise
France A dummy variable equal to 1 for an institutional investor 0.02 0.00 0.14 0.00 1.00
based in France, and 0 otherwise
430

Italy A dummy variable equal to 1 for an institutional investor 0.02 0.00 0.14 0.00 1.00
based in Italy, and 0 otherwise
Austria A dummy variable equal to 1 for an institutional investor 0.03 0.00 0.17 0.00 1.00
based in Austria, and 0 otherwise
Belgium A dummy variable equal to 1 for an institutional investor 0.02 0.00 0.14 0.00 1.00
based in Belgium, and 0 otherwise
Article III. Panel D. Internal Investment Decision-making
The PE team A dummy variable equal to 1 if the institutional investors 0.24 0.00 0.43 0.00 1.00
decision-making for PE is done by the PE team.
Equities team A dummy variable equal to 1 if the institutional investors 0.02 0.00 0.14 0.00 1.00
decision-making for PE is done by the equities team.
Alternative asset team A dummy variable equal to 1 if the institutional investors 0.12 0.00 0.33 0.00 1.00
decision-making for PE is done by the alternative asset team.
Table 23.1continued

Definition Mean Median Standard Min Max


Deviation
Board / investment A dummy variable equal to 1 if the institutional 0.38 0.00 0.49 0.00 1.00
committee investors decision-making for PE is done by the
board / investment committee.
Consultant A dummy variable equal to 1 if a consultant presented 0.09 0.00 0.29 0.00 1.00
or advised of listed PE opportunities
Article IV. Panel E. Beliefs of Decision-Makers
Listed PE offers more A dummy variable equal to 1 if the decision-makers 0.84 1.00 0.37 0.00 1.00
liquidity than limited investing in PE believe listed PE offers more liquidity than
partnerships limited partnership PE
431

Listed PE allows access to A dummy variable equal to 1 if the decision-makers 0.69 1.00 0.46 0.00 1.00
PE immediately investing in PE believe listed PE enables access to PE
immediately
Listed PE companies A dummy variable equal to 1 if the decision-makers 0.54 1.00 0.50 0.00 1.00
are attractive to invest after investing in PE believe listed PE is an attractive way
the J-curve to invest after the J-curve (meaning low returns on
investment in initial periods with spiked returns later on).
Listed PE simplifies the A dummy variable equal to 1 if the decision-makers 0.73 1.00 0.45 0.00 1.00
administrative burden and investing in PE believe listed PE offers less administrative
cash-flow management costs and cash-flow management burden than limited
associated with PE partnership PE

Note: This table presents definitions and summary statistics from the Listed Private Equity (LPEQ 2008) survey. The sample comprises 100 institutional investors in
Europe in 2008.
432 t r e n d s i n p r i v a t e e q u i t y

these decision-makers, Table 23.1, Panel E, shows that 84 percent believe that listed PE
offered improved liquidity, 69 percent believe listed PE afforded access to PE without
any delay, 54 percent believe listed PE is an attractive way to invest after the J-curve.
The J-curve shows that PE shows lower returns at the start of a funds life due to
management fees and other costs, but higher returns later in a funds life as capital is
invested and investments are harvested. A total of 73 percent believe that listed PE
simplified administrative burdens and cash-flow management relative to limited part-
nership PE.
Table 23.2 presents comparison of means, medians, and proportions tests for the
different variables in the data set depending on whether listed PE is, or is not part of the
institutional investors investment mandate. The data in Panel A show a substantially
higher proportion of private pension funds invest in listed PE (33 percent) than those
that do not (14 percent), and this difference is statistically significant at the 5 percent
significance level. One explanation for this large difference is that in Europe, private pen-
sion funds often have much smaller investment teams than public pension funds, and
as such have less time and expertise to carry out due diligence and review and negotiate
limited partnership contracts. Not surprisingly, listed PE is a more attractive asset class
for private European pension funds, but a less attractive asset class for public European
pension funds. The attributes of public pension funds may not be the same across Eu-
ropean countries. For example, Dutch public pension funds tend to be generally larger
and better resourced than most of their U.K. counterparts. Several U.K. local authority
(public) pension funds invest in listed PE according to LPEQ. Smaller pension funds
may use listed PE as an easy way to gain diversification, either via a listed fund-of-funds
or via a portfolio of direct listed PE vehicles. Such pension funds do so because they
may not be large enough to run their own fund program. In effect, investing in listed PE
not only lowers due diligence and review costs but also improves diversification.
The comparison tests in Panel C further show that investment in listed PE is more
common for institutions where the decision to invest is made by the equities team
(4.65percent in listed PE versus 0 percent not invested), but less common when the
alternative asset team decides to invest (4.65 percent invested in listed PE versus 18
percent not invested). These differences are statistically significant at the 0.10 and 0.05
levels, respectively, and consistent with Hypothesis 2. Listed PE is much more common
when recommended by a consultant (19 percent versus 2 percent), and this difference
is significant at the 0.01 level. Finally, in terms of regional differences, the comparison
tests in Panel B indicate investments in listed PE are less common in Denmark, and this
difference is significant at the 0.10 level. This latter result is weakly consistent with our
Hypothesis 1B.
An examination of the correlation matrix of all the variables in the data set (available
on request) shows that the correlation matrix is consistent with the comparison tests
in Table 23.2 including the positive and significant correlation at the 0.05 level of 0.21
between specifying listed PE on the investment mandate and the investment decisions
being made by the equities team and private pension funds. Private pension funds are
less likely to have PE investment decision made by an equities team (correlation 0.10).
The size of assets under management is negatively correlated (0.19) with listed PE
being part of the investment mandate, which is significant at the 0.10 level and consist-
ent with Hypothesis 1A. Further, consistent with Hypothesis 2, listed PE is less likely
Table 23.2 Comparisons of Descriptive Statistics between Listed Private Equity as a Part and not a Part of Investment Mandate
Listed PE Part of Investment Mandate Listed PE Not Part of Investment Mandate Comparison of Means
(Medians) and Proportions
Article V. Mean (or Proportion Article VI. Article VII. Mean (or Proportion Article VIII.
for dummy variables
for Dummy Variables) Median for Dummy Variables) Median
Panel A. Institutional Investor Characteristics
Assets under 82,924.70 3,587.60 37,945.80 5,000.00 0.70(P <= 0.53 for
management comparison of medians)
Investment 0.00 0.00 0.02 0.00 0.87
Bank
Bank 0.116279 0.00 0.16 0.00 0.59
Family 6.98E-02 0.00 0.11 0.00 0.61
Endowment 2.33E-02 0.00 0.00 0.00 1.16
Public Pension Fund 0.23 0.00 0.35 0.00 1.28
Private Pension Fund 0.33 0.00 0.14 0.00 2.21**
Insurance Company 0.23 0.00 0.23 0.00 0.05
Article IX. Panel B. Location
United Kingdom 0.33 0.00 0.21 0.00 1.30
Switzerland 0.14 0.00 0.14 0.00 0.01
Denmark 0.00 0.00 7.02E-02 0.00 1.77*
Netherlands 0.16 0.00 8.77E-02 0.00 1.14
Finland 9.30E-02 0.00 7.02E-02 0.00 0.42

continued
Table 23.2continued

Listed PE Part of Investment Mandate Listed PE Not Part of Investment Mandate Comparison of Means
(Medians) and Proportions
Article V. Mean (or Proportion Article VI. Article VII. Mean (or Proportion Article VIII.
for dummy variables
for Dummy Variables) Median for Dummy Variables) Median
Germany 6.98E-02 0.00 0.157895 0.00 1.34
Liechtenstein 2.33E-02 0.00 0.00 0.00 1.16
Sweden 9.30E-02 0.00 3.51E-02 0.00 1.21
France 0.00 0.00 3.51E-02 0.00 1.24
Italy 0.00 0.00 3.51E-02 0.00 1.24
Austria 0.00 0.00 5.26E-02 0.00 1.53
434

Belgium 0.00 0.00 3.51E-02 0.00 1.24


Article X. Panel C. Internal Investment Decision-Making
PE team 0.23 0.00 0.25 0.00 0.15
Equities team 4.65E-02 0.00 0.00 0.00 1.65*
Alternative Asset Team 4.65E-02 0.00 0.18 0.00 1.96**
Board / Investment 0.35 0.00 0.40 0.00 0.56
Committee
Consultant 0.19 0.00 0.02 0.00 2.90***
Article XI. Panel D. Beliefs of Decision-Makers
Listed PE offers more 0.86 1.00 0.82 1.00 0.48
liquidity than limited
partnerships
Table 23.2continued

Listed PE Part of Investment Mandate Listed PE Not Part of Investment Mandate Comparison of Means
(Medians) and Proportions
Article V. Mean (or Proportion Article VI. Article VII. Mean (or Proportion Article VIII.
for dummy variables
for Dummy Variables) Median for Dummy Variables) Median
Listed PE allows access 0.70 1.00 0.68 1.00 0.14
to PE immediately
Listed PE companies 0.58 1.00 0.51 1.00 0.72
435

are attractive to invest


after the J-Curve
Listed PE simplifies the 0.74 1.00 0.72 1.00 0.28
administrative burden
and cash flow
management associated
with PE

Note: This table presents comparison of means, medians, and proportions tests for the different variables in the dataset depending on whether listed PE is or is not part of
the institutional investors investment mandate.
436 t r e n d s i n p r i v a t e e q u i t y

to be an attractive option where an alternative asset team is empowered with the deci-
sion to invest (correlation of 0.19 and significant at the 0.10 level).

Multivariate Analyses
This section first describes logit regressions that were carried out to explore who invests
in listed PE. The second part describes multivariate tests of the allocations to listed PE rel-
ative to limited partnership PE and total assets. For conciseness, the regression tables are
not explicitly presented, although they are available on request from the chapter authors.

LOGIT REGRESSIONS
Six logit models are considered for three different dependent variables. Models 1 and 2
are logit analyses of whether the institutions invest in listed PE. Models 3 and 4 consider
whether the institutional investors allocation to listed PE is variable over time. Finally,
Models 5 and 6 examine whether the institutional investors decision to invest in listed
PE is dependent on whether the fund manager also manages a limited partnership PE
fund. The general specification of the models has the following form:
Invest in Listed PE = f [constant, institutional investor characteristics (size, type),
location, decision-making (e.g., equities team and PE team), beliefs of decision-makers]
Four variables exist for the beliefs of the decision-makers: (1) whether the decision-
makers believe the greater liquidity, (2) whether there is quicker access to PE, (3)
whether the J-curve is avoided, and (4) whether the simplified administrative burden
of listed PE is useful. For each of the three dependent variables, two sets of explanatory
variables are used to assess robustness to different specifications. The first specification
includes assets under management in a linear specification with dummy variables for
private and public pension funds. The analyses exclude additional dummy variables for
other types of institutional investors to avoid multicollinearity. The second specification
uses the natural log of assets under management to account for a potential decreasing
importance of size on the decision to invest as size gets larger. Also, the interaction be-
tween type of institutional investor and size is analyzed to explore whether larger public
pension funds behave differently than larger private pension funds.
Models 1 and 2 both show strong results in support of Hypothesis 2. Where
decision-making is allocated to the PE team, an alternative asset team, or the board
or investment committee, investment in PE is approximately 28, 43, and 38 percent
less likely, respectively, relative to allocating decision-making to an equities team. The
dummy for the equities team is suppressed for reasons of multicollinearity. Alternative
investment teams are 13 percent less likely to use variable listed PE allocations (Models
3 and 4), at least relative to PE teams and board/investment committee and equities
teams. By contrast, consultants that advise listed PE opportunities increase the proba-
bility of listed PE investment by 53 percent (significant at the 0.01 level in Models 1 and
2), and increase the probability of listed PE investment by up to 65 percent in Model 3
(significant at the 0.01 level). This marginal effect is 45 percent in Model 4 and signifi-
cant at the 0.05 percent level.
L is t e d P riv at e E qu it y 437

As indicated previously, relative to public European pension funds, private European


pension funds have fewer individuals on the investment team, which makes the added
due diligence and costs and time with limited partnership PE much less attractive. The
data indicate listed private allocations are less likely to be variable for private pension
funds and larger private pension funds.
Models 1 and 2 do not provide support for Hypothesis 3 in terms of liquidity prefer-
ences and beliefs influencing the decision to invest in listed PE. However, in Models
3 and 4, institutional investors using listed PE to obtain access to PE immediately are
about 15 percent more likely to have variable allocations to listed PE over time, which
is consistent with Hypothesis 3. This finding suggests that as exposure to limited part-
nership PE varies over time with a slower drawdown rate, institutional investors make
adjustments to adjust to exposure to listed PE.
To complement the analyses in Models 1 through 4, Models 5 and 6 examine whether
the decision to invest in listed PE depends on whether a PE fund manager in which the
institution invests manages both a listed and limited partnership PE fund. The one var-
iable that is significant in Models 5 and 6 is the variable investment decisions made by
the board or investment committee, which shows that investment would be roughly 10
percent less likely for funds that manage both listed and limited partnership PE funds.
Some investors may want to diversify into a more liquid investment of a different man-
agers listed fund while waiting for the capital commitments to be called for its direct in-
vestment in an LP fund. One benefit of investing with the different manager of the listed
PE fund is to take advantage of a different maturity profile while managing cash flows
(i.e., liquidation events) to time commitments to an LP fund. Other investors having
conducted the due diligence on a manager may believe that that they would like to hold
some additional allocation to the same manager in a more liquid form and maintain ex-
posure to similar underlying investments.

R E G R E S S I O N A N A LY S E S O F P E R C E N TA G E A L L O C AT I O N S
The next set of regression tests considered evidence for percentage allocations into listed
PE. In these regressions, each of the dependent variables is in percentage terms. For the
residuals and estimates to have properties consistent with assumptions underlying or-
dinary least squares (OLS), the dependent variable is transformed so it is not bounded
between 0 and 100 percent, in a standard way of modeling fractions (Bierens 2003).
Models 7 and 8 explain the percentage of investment into listed PE relative to limited
partnership PE. The data indicate that larger institutional investors invest less in listed
PE relative to limited partnership PE. This effect is insignificant in Model 7 with a linear
specification and at the 0.10 level in Model 8 with a log specification. The economic sig-
nificance in Model 8 is such that a change in institutional investor size from 5 billion to
10 billion reduces the amount invested in listed PE relative to total equity by 0.5 per-
cent. This result is significant because the average amount invested in listed PE versus
limited partnership PE is 2.78 percent and the average institutional investor in the
sample manages more than 57 billion as shown in Table 23.1. The evidence in Models
9 and 10 is similar in respect of institutional investor size and the amount invested in
listed PE relative to total assets, which shows significant effects in both models at the
0.05 level. In Model 9, the economic significance is such that an increase in institutional
438 t r e n d s i n p r i v a t e e q u i t y

investor size by 5 billion reduces investment in listed PE by 2 percent. Overall, Models


7 through 10 provide strong support for Hypothesis 1A.
Models 7 through 10 do not provide support for Hypothesis 1B for location in the
United Kingdom versus continental Europe. However, Models 7 through 10 do provide
strong support for Hypothesis 2. Investment in listed PE relative to limited partner-
ship PE is significantly smaller for institutions with decision-making delegated to the PE
team and for the board/investment committee in Models 7 and 8. Similarly, in Models
9 and 10, investment in listed PE relative to total assets is significantly smaller for in-
stitutions that delegate decision-making to an alternative asset team and to the board/
investment committee.
The evidence in Models 7 and 8 is consistent with Hypothesis 3. Allocations to listed
PE versus limited partnership are about 7 percent higher for investors that believe listed
PE affords access to PE immediately. This effect is significant at the 0.05 level in both
Models 7 and 8. Hypothesis 3 is not supported in Models 9 and 10.
Models 11 and 12 provide complementary analyses of how far away the institutional
investor is from its target PE allocation. The strongest and robust results across Models
11 and 12 are seen from whether or not the investor uses listed PE, and if this allocation
changes over time. The results are consistent with expectations and are in line with Hy-
pothesis 3. Institutions that use listed PE are about 9 percent closer to their desired PE
allocation relative to institutions that do not invest in listed PE. By contrast, where insti-
tutions change their desired allocation to listed PE, they tend to deviate more from their
desired overall allocation to PE by approximately 19 percent. As the results of Model
11 show, allocations made by an equities team tend to be closer to their desired PE
allocation, consistent with Hypothesis 2 and equities teams using listed PE, although
this effect is marginally insignificant in Model 12. Similarly, larger public pension funds
tend to be farther away from their desired allocations to PE, consistent with the above
evidence respecting Hypothesis 1A and complementary evidence that public pensions
are less likely to use listed PE.

Summary and Conclusions


Recent empirical work by Jegadeesh et al. (2009) explores the public equity markets
expectations about current expected returns to investment in listed versus limited part-
nership PE. This evidence suggests that exploring factors affecting institutional investor
portfolio management and allocations to PE is worthwhile. It raises the question of how
institutional investors allocate capital within different segments of the PE market. This
chapter explores various factors that influence such capital allocation decisions to listed
PE versus limited partnership PE funds.
Based on data derived from a LPEQ (2008) survey of 100 institutional investors in
Europe, the results show that smaller institutional investors allocate capital to listed PE,
as do private (not public) pension funds, and those that prefer liquidity and are based
in the United Kingdom but not continental Europe. The data further indicate that the
empowerment of decision-making to an equities team has a pronounced impact for in-
vestment in listed PE. Where institutions allocate decision-making to a PE team, an alter-
native asset team, the board, or investment committee, investment in listed PE is much
L is t e d P riv at e E qu it y 439

less common. These findings are consistent with the hypotheses that institutions invest
in listed PE in order to reduce search costs associated with the asset class and improve
their ability to achieve a desired investment exposure in as timely a manner as possible.

Discussion Questions
1. Explain why listed PE might be more commonly considered by institutional inves-
tors based in the United Kingdom than their counterparts in continental Europe.
2. Discuss why listed PE is less likely to be considered by an institutional investor
whose decision-making is delegated to an equities team.
3. Discuss why listed PE enables institutional investors to achieve their target PE al-
locations quicker, and as such, institutions that invest in listed PE are less likely to
adjust their listed PE allocations over time in response to slower adjustments to lim-
ited partnership PE allocations.
4. Discuss how listed PE funds have performed over time since their first listing date.
5. Explain the values of the listed PE fund betas.
6. Discuss why listed PE is a more attractive asset class relative to limited partnership
PE investment for smaller institutional investors.

References
Bergmann, Bastion, Hans Christophers, Matthias Huss, and Heinz Zimmermann. 2010. Listed Pri-
vate Equity. In Douglas J. Cumming, ed., Companion to Private Equity, 5370. Hoboken, NJ:
John Wiley & Sons, Inc.
Bierens, Herman J. 2003. Modeling Fractions. Available at http://econ.la.psu.edu/~hbierens/
EasyRegTours/FRACTIONS.PDF.
Caselli, Stefano, Stefano Gatti, and Francesco Perrini. 2009. Are Venture Capitalists a Catalyst for
Innovation? European Financial Management 15:1, 92111.
Cochrane, John. 2005. The Risk and Return of Venture Capital. Journal of Financial Economics
75:1, 352.
Cumming, Douglas J. 2003. The Structure, Governance and Performance of UK Venture Capital
Trusts. Journal of Corporate Law Studies 3:2, 401427.
Cumming, Douglas J., Grant Fleming, and Sofia A. Johan. 2011. Institutional Investment in Listed
Private Equity. European Financial Management 17:3, 594618.
Cumming, Douglas J., Grant Fleming, and Joanne A. Suchard. 2005. Venture Capitalist Value-
Added Activities, Fundraising and Drawdowns. Journal of Banking and Finance 29:2, 295331.
Cumming, Douglas J., and Sofia A. Johan. 2006. Is It the Law or the Lawyers? Investment Cove-
nants around the World European Financial Management 12:4, 535574.
Cumming, Douglas J., and Jeffrey MacIntosh. 2007. Mutual Funds That Invest in Private Equity?
An Analysis of Labour Sponsored Investment Funds Cambridge Journal of Economics 31:3,
445487.
Diller, Christian, and Christoph Kaserer. 2009. What Drives Private Equity Returns? Fund Inflows,
Skilled GPs, and/or Risk? European Financial Management 15:3, 643675.
Godineni, Sridhar, and William L. Megginson. 2010. IPOs and Other Non-traditional Fundraising
Methods of Private Equity Firms. In Douglas J. Cumming, ed., Companion to Private Equity,
3151. Hoboken, NJ: John Wiley & Sons, Inc.
Gompers, Paul, and Josh Lerner. 1999. The Venture Capital Cycle. Cambridge, MA: MIT Press.
440 t r e n d s i n p r i v a t e e q u i t y

Huss, Matthias, and Heinz Zimmermann. 2013. Listed Private Equity: A Genuine Alternative for
an Alternative Asset Class. In Douglas J. Cumming, ed., Oxford Handbook of Private Equity,
579610. New York: Oxford University Press.
Jegadeesh, Narasimhan, Roman Krussl, and Joshua M. Pollet. 2009. Risk and Expected Returns of
Private Equity Investments: Evidence Based on Market Prices. Working Paper, Emory University.
Jones, Charles, and Matthew RhodesKropf. 2003. The Price of Diversifiable Risk in Venture Cap-
ital and Private Equity. Working Paper, Columbia University.
Kaplan, Steven, and Antoinette Schoar. 2005. Private Equity Performance: Returns, Persistence,
and Capital Flows. Journal of Finance 60:4, 17911823.
Lahr, Henry, and Christoph Kaserer. 2010. Net Asset Value Discounts in Listed Private Equity
Funds. Working Paper, Technische Universitt Mnchen.
La Porta, Rafael, Florencio Lopez-De-Silanes, Andrei Shleifer, and Robert W. Vishny. 1998. Law
and Finance. Journal of Political Economy 106:6, 11131155.
Lerner, Josh, Antoinette Schoar, and Wan Wongsunwai. 2007. Smart Institutions, Foolish Choices?
The Limited Partner Performance Puzzle. Journal of Finance 62:2, 731764.
Ljungqvist, Alexander, and Matthew Richardson. 2003. The Cash Flow, Return and Risk Charac-
teristics of Private Equity. NBER Working Paper No. 9454.
LPEQ. 2008. How Do Institutional Investors Regard Listed Private Equity: A Research Survey of
100 European LPs? LPEQ Listed Private Equity.
Martynova, Marina, and Luc Renneboog. 2011. The Performance of the European Market for Cor-
porate Control: Evidence from the Fifth Takeover Wave, European Financial Management,
17:2, 208259.
Suh, Jungwon. 2005. Home Bias among Institutional Investors: A Study of The Economist Quarterly
Portfolio Poll. Journal of the Japanese and International Economics 19:1, 7295.
Takahashi, Dean, and Seth Alexander. 2002. Illiquid Alternative Asset Fund Modeling. Journal of
Portfolio Management 28:2 (Winter), 90100.
Woodward, Susan E. 2004. Measuring Risk and Performance for Private Equity. Sand Hill Econo-
metrics. Available at http://www.sandhillecon.com/pdf/MeasuringRiskPerformance.pdf.
Woodward, Susan E., and Robert Hall. 2003. Benchmarking the Returns to Venture. NBER Work-
ing Paper No. 10202.
Wright, Mike, Charlie Weir, and Andrew Burrows. 2007. Irrevocable Commitments, Going Private
and Private Equity. European Financial Management 13:4, 757775.
24
Private Equity Growth in International
and Emerging Markets
ALEXANDER PETER GROH
Professor of Finance and Director of the Research Centre for Entrepreneurial Finance,
EMLYON Business School

Introduction
Data on international private equity (PE) activity since 2000 reveal different develop-
ment patterns across the countries examined. In general, after the historic high in 2000,
PE peaked again in 2007 in traditional PE markets, especially the United States. Tra-
ditional PE markets are those in which PE activity reached a substantial level no later
than in the 1980s and where investment volumes meanwhile reached remarkable levels
(e.g., the United States, Canada, the United Kingdom, Australia, Japan, and most of the
Western European economies). In emerging countries, no such peak occurred. In
Africa, Asia, and Latin America, PE investment volume increased gradually over time,
compensating for some of the lower volume in traditional markets. However, emerging
market PE transactions remain concentrated in a few countries, especially China and
India. Several other developing countries do not show the vibrant activity that might be
expected given their economic growth potential.
This chapter presents comparative data on international PE investments and dis-
cusses the conditions required to increase transaction volumes. The data allow bench-
marking analysis for specific countries including an analysis of strengths and weaknesses
in establishing vibrant PE markets. Although many emerging PE countries provide op-
portunities from a long-term perspective, investors should not neglect the traditional
markets.
The remainder of the chapter is organized as follows. First, data on global PE ac-
tivity are presented. Then the data are broken down into several geographic regions.
Afterwards, the main drivers of PE activity are discussed followed by a presentation of
the ongoing research project entitled the PE country attractiveness index. This index
serves as a basis for a five-year comparison to point to promising host countries and to
explain increasing PE activity in selected markets, notably the BRICs (Brazil, Russia,
India, and China), Indonesia, Mexico, the Philippines, and Turkey. The index also helps
to determine obstacles that deter additional PE investments. Next, the question about

441
442 t r e n d s i n p r i vat e e q u i t y

a minimum maturity level from which PE activity emerges is discussed and a correla-
tion between this index and actual PE returns is revealed. Finally, the chapter ends with
concluding remarks.

Private Equity Activity in the New Millennium


Several commercial providers of PE transaction data are available. All have different
coverage of international markets and, more important, various definitions of PE in-
cluding or excluding sub-categories such as venture capital or mezzanine financing. The
information on the volume and number of transactions therefore varies from one data
provider to another. A brief look at growing global PE activity in Figure 24.1 shows the
historic high in the year 2000 and the peak before the financial crisis of 20072008.
Activity in the United States, which is still by far the dominant market, mainly drove
the 2007 peak. Nevertheless, other regions have gained in importance, despite evolving
from relatively low activity levels. Figure 24.1 also reveals that 2013 was a weak year
from a global perspective. However, the breakdown by geographic region in Figure 24.2
shows several different development paths.
Figure 24.2 illustrates the trends and volatility of PE activity in different geographic
markets. The vertical axis is on a logarithmic scale presenting all distinct regions in one
diagram and directly comparing fluctuations. The growth trend over the last decade is
positive for most of the emerging regions, but flat for Europe and North America. The

300

250

200

150

100

50

0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

Figure 24.1 Global Private Equity Investment Volume This figure shows global
PE investment volume in $ billion in the new millennium. It reveals the all-time high in
2000 and the peak before the financial crisis of 20072008. Since then, the investment
volume has been volatile. Source: ThomsonOne.com.
PE Gr owt h in I n t e rn at ion al M arke t s 443

1,000.00

100.00

10.00

1.00

0.10

0.01
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

Africa Asia Australasia Europe


Middle East North America South America

Figure 24.2 Private Equity Investment Volume Differentiated by Region


This figure shows PE investment volume in $ billions on a logarithmic scale, differentiated
according to major geographic regions. Growth trends are positive for most of the emerging
regions while they are rather flat for Europe and North America. Source: ThomsonOne.com.

BRICs mainly drive emerging market growth while most other developing economies
have relatively low and erratic investment volumes.
Figure 24.3 isolates the BRICs activity and highlights their growing stake in the
global competition for international capital. It shows the rise in PE investments in the
BRICs, again using a logarithmic scale. The figure reveals that the financial crisis of
20072008 did not strongly affect these markets. Therefore, they provide suitable alter-
natives for the fall-off in activity in the traditional PE countries. Nevertheless, PE is an
illiquid asset class and investors need to allocate strategically as they cannot switch their
regime in the short term such as during times of financial crisis.
Future investment opportunities in the BRICs and in the emerging countries that
might follow the BRICs path are open to debate. One assumed reason the BRICs are
so successful in their PE market development is the combination of high population
and the catch-up potential during their transition to modern economies. The catch-up
potential is usually assessed by comparing gross domestic product (GDP) to capita fig-
ures of less and more highly developed countries. A countrys population is important
because it determines the expected economic growth in absolute terms. Larger coun-
tries require higher levels of financing and attract more investors. They might, therefore,
have greater access to capital, which is essential for economic development. Thus, the
search for the future BRICs points toward countries with large populations and strong
economic growth potential.
444 t r e n d s i n p r i vat e e q u i t y

100.00

10.00

1.00

0.10

0.01
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

Brazil Russia India China

Figure 24.3 Private Equity Investment Volume in Brazil, Russia, India, and
China This figure shows PE value in the BRIC countries in $ billions on a logarithmic
scale. Growth trends are positive for the BRICs and highlight their gain of global market
share. Furthermore, the BRICs were virtually unaffected by the debt market crisis and
provide alternatives for fall-offs in traditional countries. Source: ThomsonOne.com.

Several countries fulfill these two criteria at the time of publication including Ban-
gladesh, Egypt, Ethiopia, Indonesia, Nigeria, Mexico, Pakistan, the Philippines, Turkey,
and Vietnam. However, these countries are in different stages of economic development,
and most of them have minimal reported PE activity. In the end, size and economic
growth expectations are not the only relevant drivers for PE activity. More favorable
conditions are required to establish a vibrant PE market with the most obvious being
the prospect of appropriate deal flow. Deal flow requires a countrys development level
to be sufficiently elevated to support the PE transaction model. To date, several devel-
oping regions have not achieved this development level. In an ongoing research project,
Groh, Liechtenstein, and Lieser (2014) address country maturity and the ability to sus-
tain PE transactions.

Ability of Countries to Support the Private Equity


Business Model
Next, the chapter tackles the question of countries ability to create a PE deal-making
environment by using an index that benchmarks PE attractiveness from an investors
perspective for 118 countries of which 85 are emerging markets. The first step in this
search is to understand the criteria shaping national PE markets based on interviews of
PE Gr owt h in I n t e rn at ion al M arke t s 445

selected investors. A total of 1,079 institutional investors worldwide are surveyed and
their inputs reviewed using selected academic research (Groh and Liechtenstein 2009,
2011a, b, c; Groh, Liechtenstein, and Canela 2010a).
The index is built using a top-down approach by assessing the ways in which so-
cioeconomic criteria affect the deal-making environment. The methodology follows
a practical approach summarizing factors that make a country attractive for PE in-
vestment into a single composite measure. Therefore, six key drivers that shape
vibrant PE markets are differentiated: (1) economic activity, (2) depth of the cap-
ital market, (3) taxation, (4) investor protection and corporate governance, (5)
human and social environment, and (6) entrepreneurial culture and the resulting
deal opportunities.

ECONOMIC ACTIVITY
The state of a countrys economy affects its PE market. General prosperity, the vari-
ety of existing corporations, overall entrepreneurial activity, the size of the economy,
and employment levels affect the deal flow, and hence the number and volume of PE
transactions. However, the factor that most strongly affects the PE market is the ex-
pected economic growth (Gompers and Lerner 1998). Economic activity and growth
opportunities, in particular, also result from a combination of additional socioeconomic
criteria that are discussed later in the chapter.

D E P T H O F T H E C A P I TA L M A R K E T
Black and Gilson (1998) discuss the differences in two types of capital markets that
affect PE transactions: bank-centered and stock marketcentered. Well-developed stock
markets allow for IPO exits, which are essential for establishing PE activity. This devel-
opment is also a strong sign of merger and acquisition (M&A) market size, which offers
the possibility of deal sourcing and future divestment.
Conversely, bank-centered capital markets prove less efficient in fostering the in-
stitutional infrastructure that supports deal flow and deal-making. Besides lacking
a strong stock market, bank-centered capital markets cultivate a conservative ap-
proach to lending and investing. Social and financial incentives in these secondary
institutions reward entrepreneurs in a less lucrative manner while penalizing fail-
ure more harshly, which in turn compromises entrepreneurial activity. According
to Greene (1998), inaccessibility to debt financing hinders economic development
and can prove especially difficult for start-up activity. The ability to find willing, risk-
bearing backers is essential to success. Gompers and Lerner (2000) claim that this
risk capital is cultivated through deep, liquid stock markets. Cetorelli and Gambera
(2001) argue that bank concentration facilitates access to credit, in turn promoting
growth in capital-intensive sectors.
The quality of the PE deal-making infrastructure can be measured by the liquidity
of the M&As, banking, and public capital markets. Countries in which these aspects
are strong generally have the key professional institutions (e.g., investment banks,
accountants, lawyers, M&A boutiques, and consultants) required for successful PE
deals.
446 t r e n d s i n p r i vat e e q u i t y

TA X AT I O N
Bruce (2000, 2002) and Cullen and Gordon (2002) acknowledge the impact that tax
regimes have on general business activity including market entry and exit barriers.
Djankov, Ganser, McLiesh, Ramalho, and Shleifer (2008) confirm that taxes, both direct
and indirect, affect entrepreneurial activity. According to Bruce and Gurley (2005), the
gap between personal income and corporate tax rates incentivizes entrepreneurship and
corporate activity.
Such a link with PE investments is difficult to establish. In fact, PE activity is not
strongly correlated to tax brackets. Countries with relatively high corporate tax rates
may still have high numbers of PE transactions. Conversely, many emerging countries
with low corporate tax rates have little to no recorded PE investments. Although devel-
oped countries have higher tax rates, they also have higher PE investment in general.

I N V E S TO R P R OT E C T I O N A N D C O R P O R AT E G O V E R N A N C E
Legal structures and protecting property rights strongly influence activity in PE markets.
La Porta, Lopez-de-Silanes, Shleifer, and Vishny (1997, 1998) confirm that a countrys
legal environment determines the size and extent of its capital market and the ability
of local companies to get outside financing. Roe (2006) reiterates the importance of
strong shareholder protection through a comparison of the political determinants of
corporate governance legislation for the major economies. Glaeser, Johnson, and Shle-
ifer (2001) and Djankov, La Porta, Lopez-de-Silanes, and Shleifer (2003, 2008) suggest
that enforcing investors rights is easier in common law countries.
Cumming, Flemming, and Schwienbacher (2006) determine that PE-backed exits
are more strongly correlated to the quality of a countrys legal system than to stock
market characteristics. According to Cumming, Schmidt, and Walz (2010), differences
in legal activity and accounting practices between countries strongly affect PE invest-
ments. Fairness and property rights protection strongly affect growth and emerging new
enterprises (Desai, Gompers, and Lerner 2006). Along similar lines, La Porta, Lopez-
de-Silanes, Shleifer, and Vishny (2002) and Lerner and Schoar (2005) note the cost of
capital is typically lower in countries with better investor protection.
The quality of a countrys legal system is essential for its capital market and the cor-
porate environment. Without proper legal protection and enforcement, conducting
business can become costly. Since PE investments are built on long-term relations with
institutional investors, and given that investment source and host countries can be dis-
tant, investors must be confident that their claims will be well protected before they
allocate capital.

HUMAN AND SOCIAL ENVIRONMENT


Researchers have focused on the way in which differences in culture can lead to varying
entrepreneurial activity levels (Black and Gilson 1998; Lee and Peterson 2000; Baughn
and Neupert 2003). Besides, countries with more developed educational institutions
create an environment that supports and cultivates the market for risk capital (Meg-
ginson 2004).
PE Gr owt h in I n t e rn at ion al M arke t s 447

Imagine a company unable to keep up with changing market conditions due to re-
strictions imposed on its employment policies. All other things being equal, this com-
pany could not compete at the same level as another company in a country with flexible
labor market conditions and policies. The companies and the countries would have
different growth prospects, and therefore varying PE investments. Black and Gilson
(1998) believe, however, that labor market restrictions less strongly influence PE in-
vestment than the stock market.
Additionally, a society with corruption, crime, black markets, or substantial bureauc-
racy will have the highest barriers and therefore the highest costs to entry (Djankov,
La Porta, Lopez-de-Silanes, and Shleifer 2002). Some emerging countries have high
perceived corruption and crime levels that are particularly important in assessing their
attractiveness.

E N T R E P R E N E U R I A L C U LT U R E A N D D E A L O P P O R T U N I T I E S
A countrys innovative capacity and research output are important for PE investment.
Researchers show that investment in research and development (R&D) can act as an es-
timate of human capital endowment and is highly correlated with PE investment (Gom-
pers and Lerner 1998; Schertler 2003). Without innovation through R&D, building
and preserving the strong brand names and market positions that attract PE investors
would be difficult for businesses.
Despite the innovative output of a society, Djankov et al. (2002) and Baughn and
Neupert (2003) claim that bureaucracy (e.g., excessive rules, procedural requirements,
the need to get approval from various institutions, and cumbersome documentation
requirements) can deeply hinder entrepreneurial activity. These time-consuming and
costly burdens may discourage investment (Lee and Peterson 2000).

C O N C L U S I O N S O N T H E D E T E R M I N A N T S O F V I B R A N T P R I VAT E
EQUITY MARKETS
Identifying the most appropriate parameters for PE activity is challenging. No consen-
sus is available on ranking the criteria. Although some parameters receive more atten-
tion in the literature than others, no clear-cut interaction exists among them.
That being said, many criteria are highly correlated. Black and Gilson (1998) refer
to it as a chicken and egg problem so determining causality is impossible. One may
argue that modern, open, and educated societies develop legislation to protect inves-
tors claims, favoring the output of innovation and the development of a capital market,
which, in turn, stimulates economic growth and the required investments. Yet, one
could argue the reverse: economic growth fosters innovation and developing modern,
open, and educated societies. Others may believe that support for innovation, economic
growth, capital markets, and therefore PE activity, can only be developed through com-
petitive legal environments. Another possible option is that modern, open, and edu-
cated societies are created by attracting investors through low tax schemes.
Although each of these arguments is reasonable and validated by the economic de-
velopment of different countries over different historical periods, improving these fac-
tors is crucial to increase a countrys PE activity. For this reason, the index used in this
448 t r e n d s i n p r i vat e e q u i t y

chapter does not rely on selecting only a few parameters; instead, a country must rank
highly on each individual criterion for it to be ranked highly in the overall index.

The Current Situation in Terms of Private Equity


Country Attractiveness
The structure of the index used for analysis in this chapter identifies the six key drivers
previously discussed and confirms their importance for institutional investors. These
drivers are labeled the six latent key drivers because they cannot be measured directly.
They are instead driven by various other factors that are more easily measured through
proxies. At times, a lack of data availability on these proxies for a large portion of the
sample countries constrained analysis for this chapter. The six latent key drivers of PE
maturity comprise the factors discussed previously: (1) economic activity, (2) depth of
the capital market, (3) taxation, (4) investor protection and corporate governance, (5)
human and social environment, and (6) entrepreneurial culture and deal opportunities.
If the six key drivers sufficiently determine a countrys PE attractiveness, then finding
data to estimate these drivers is important. For example, the quality of a countrys deal-
making environment can ideally be expressed by the number of its investment banks,
M&A boutiques, law firms, accountants, and consultants. Yet, these data do not exist on
a global scale, and the only alternative is to gather more general information (e.g., the
debt level provided by the banking sector or estimates about the perceived sophistica-
tion of the financial system) shaping the latent key driver, depth of the capital market.
While they may not be perfect proxies, the more developed these criteria are, the deeper
is the countrys capital market. This, in turn, leads to the emergence of more deal-
supporting institutions to facilitate PE activity. This principle is maintained throughout
constructing the index: the ability to evaluate each latent key driver via many proxy
parameters. As described in detail in Groh, Liechtenstein, and Lieser (2010b, 2014), 51
data series and different weighting and aggregation techniques are used to condense the
information into a single composite score.

T H E P R I VAT E E Q U I T Y AT T R A C T I V E N E S S R A N K I N G
The results show the United States is the country with the best PE deal-making condi-
tions. Therefore, the United States is defined as the benchmark and its score is rescaled
to 100 to serve as a reference point for all other countries. Figures 24.4 and 24.5 present
rankings of all countries covered. The United States is followed by the other traditional
PE markets. The ranking of the emerging countries is led by Malaysia, ahead of China,
Chile, Taiwan, and India.
The index leaves room for debate. Some may think that particular countries are
ranked too high, while others are too low. The index ranking is the result of commonly
available, transparent, aggregated socioeconomic data, which are relevant according to
existing research on the topic. The results can be traced back to individual data series,
and hence, can be completely reconciled. The most important conclusion from the
ranking is not the exact ranking position of each country but rather its level compared
to the overall competition (e.g., among the top 20; or above, or below, the median of all
countries).
PE Gr owt h in I n t e rn at ion al M arke t s 449

0 10 20 30 40 50 60 70 80 90 100
United States
Canada
Singapore
United Kingdom
Hong Kong
Japan
Germany
Australia
Sweden
Switzerland
New Zealand
Norway
Malaysia
Netherlands
Belgium
Denmark
Korea, South
Finland
Israel
France
Austria
China
Chile
Taiwan
Ireland
Saudi Arabia
Spain
India
Poland
Turkey
Thailand
South Africa
Luxembourg
Italy
Portugal
Colombia
Czech Republic
United Arab Emirates
Mexico
Brazil
Russian Federation
Philippines
Lithuania
Oman
Hungary
Indonesia
Peru
Slovakia
Morocco
Slovenia
Estonia
Romania
Jordan
Iceland
Latvia
Bulgaria
Mauritius
Bahrain
Argentina

Figure 24.4 Country Private Equity Attractiveness Score: First Half of


Sample Countries This figure presents the PE attractiveness scores for the first half of the
sample. The United States is most attractive for PE investors, followed by Canada, Singapore,
and the United Kingdom. The highest ranked emerging country is Malaysia, ahead of China,
Chile, Taiwan, and India. Source: The Venture Capital and Private Equity Country Attractiveness Index,
http://blog.iese.edu/vcpeindex//.

The data used to construct this index are based on information available at the end
of 2013. The index shows the current attractiveness ranking and includes the economic
outlook for 2014.

M E D I U M - T E R M T R E N D S I N T H E P R I VAT E E Q U I T Y
AT T R A C T I V E N E S S R A N K I N G
The medium-term development of the sample countries might be even more rele-
vant than their current ranking because it allows us to focus on recent trends and the
450 t r e n d s i n p r i vat e e q u i t y

0 10 20 30 40 50 60 70 80 90 100
Tunisia
Pakistan
Greece
Ukraine
Croatia
Cyprus
Uruguay
Zambia
Vietnam
Egypt
Kazakhstan
Georgia
Kuwait
Ecuador
Ghana
Bosnia-Herzegovina
Nigeria
Bangladesh
Macedonia
Serbia
Jamaica
Botswana
Kenya
Namibia
Montenegro
Uganda
Armenia
Mongolia
Algeria
Cambodia
El Salvador
Tanzania
Belarus
Paraguay
Guatemala
Dominican Republic
Moldova
Cte d'Ivoire
Malawi
Rwanda
Nicaragua
Kyrgyzstan
Ethiopia
Cameroon
Madagascar
Senegal
Albania
Venezuela
Zimbabwe
Mali
Mozambique
Syria
Benin
Burkina Faso
Mauritania
Lesotho
Chad
Angola
Burundi

Figure 24.5 Country Private Equity Attractiveness Score: Second Half


of Sample Countries This figure presents the PE attractiveness scores for the second
half of the sample. The least attractive countries are Lesotho, Chad, Angola, and Burundi.
Accordingly, out of these countries and several more above their level of attractiveness, no
PE activity has been reported as of 2014. Source: The Venture Capital and Private Equity Country
Attractiveness Index, http://blog.iese.edu/vcpeindex/.

improvement/weakening in investment conditions. In Figure 24.6, the development of


the countries index rankings are tracked between 2009 and 2014.
Figure 24.6 shows the improvement or decline in PE attractiveness for the 118
countries between 2009 and 2014. The vertical axis shows the current rank and
the horizontal axis the changes in rank. For example, the Philippines increased its
PE Gr owt h in I n t e rn at ion al M arke t s 451

0
Hong Kong
Australia
New Zealand
Malaysia
Denmark
Decreasing Finland Highly
20 Attractiveness: France Attractive:
Observe Chile Invest

Turkey

Colombia
40 Brazil Mexico
Russia Philippines
Hungary Oman Lithuania
Indonesia Peru
Slovakia
Current Rank Position

Slovenia Morocco
Estonia
Jordan Iceland
Bahrain Mauritius
60 Tunisia
Greece Ukraine
Cyprus
Egypt Kazakhstan
Kuwait Georgia
Ghana Ecuador

80 Serbia

Montenegro
Armenia

Paraquay

100
Unattractive: Inreasingly
Avoid Attractive:
Be Prepared

120
25 20 15 10 5 0 5 10 15 20 25
Rank Changes [negative = loss; positive = gain] over the Last Five Years

Africa Asia Australasia Eastern Europe


Latin America Middle East North America Western Europe

Figure 24.6 Attractiveness Rank Changes This figure depicts the changes of
attractiveness ranks over five years. It also groups countries into different segments of
attractiveness for investors. The countries in the upper-right corner have the highest likelihood
to achieve additional investor attention and to support increased deal activity. Source: The
Venture Capital and Private Equity Country Attractiveness Index, http://blog.iese.edu/vcpeindex/.

attractiveness by 22 positions and now ranks 42. The upper-right corner of the graph
shows highly attractive investment opportunities. The countries in this corner are
sufficiently mature to sustain the PE business model and to secure deal flow as meas-
ured by the overall index rank, and they have strongly improved their attractiveness
during the observation period. The emerging and developed countries that appeal to
452 t r e n d s i n p r i vat e e q u i t y

investors, and in which rising PE activity can be expected, include Hong Kong (not a
country but a city-state), New Zealand, Malaysia, Finland, Chile, Turkey, Colombia,
Mexico, Russia, the Philippines, Lithuania, Oman, Indonesia, Peru, Morocco, and
Estonia.
The countries in the lower-right corner also increased their attractiveness. However,
their general maturity level is likely not yet sufficiently high to support PE deal-making.
Investors might therefore choose to assess these countries with special attention so as
not to miss the opportunity when PE deal flow materializes there. The lower-left corner
of the figure contains the countries with generally low rankings that deteriorated over
the last five years. These countries are unattractive to investors. Nevertheless, the losses
might not necessarily be due to the generally decreasing quality of investment condi-
tions. Some countries underperform relative to their peers in the competition to attract
investors. Competition is fierce and small changes in the attraction determinants may
cause volatile rankings, particularly for the lower ranked countries.

THE SPECIAL CASE OF BRAZIL, RUSSIA, INDIA, AND CHINA


As noted previously, the BRICs have received strong investor attention in recent years.
China is among the top countries for worldwide PE activity, followed closely by India.
All the BRICs have reasonable levels of PE attractiveness and, especially for China and
India, one might argue that the attractiveness measure understates the investment real-
ity. Figure 24.7 takes a closer look at the key drivers that shape the BRICs and at the data
granulation level below the six key driving forces.
Investors capitalize on the BRICs strong economic growth. However, the economic
situation is not the only factor that explains rising PE activity. In parallel to their ec-
onomic progress, the BRICs, and Russia to a lesser extent, have developed strong
capital markets. China, meanwhile, ranks second worldwide in terms of the market
capitalization of listed companies, and India has the largest number of domestic public
companies. Similarly, the public issuing activity and the number and volume of M&A
transactions are high in both countries. This dynamic ensures professional investment
environments, general financial market liquidity, and PE deal opportunities.
Nevertheless, Figure 24.7 also reveals that the key driving factors of PE attractive-
ness are skewed in these countries. They are generally characterized by strong economic
activity and capital markets, despite the fact that the key drivers related to investor pro-
tection, human and social environment, and entrepreneurial culture are rather poorly
developed. To gain further insight, the key drivers are disaggregated to the level of their
underlying data series and presented in Figure 24.8.
Figure 24.8 supports the argument about economic growth and the BRICs capital
markets. Yet, it exposes some concerns about these countries and more generally about
most other emerging markets. Corporate governance factors and investor protection
remain major obstacles to be tackled. These countries, especially Russia, also suffer from
high levels of perceived bribery and corruption, while innovations and corporate R&D
are relatively weak.
Figure 24.8 also reveals that economic activity is constructed using size, GDP growth
in real terms, and employment as proxies. China is, therefore, the only country to have
a higher score than the United States because of its exceptional real growth and similar
PE Gr owt h in I n t e rn at ion al M arke t s 453

1 Economic Activity
125
100
6 Entrepreneurial 75
2 Capital Market
Opportunities
50
25
0

5 Human and Social


3 Taxation
Environment

4 Investor Protection
and Corporate
Governance
Brazil (40) Russian Federation (41) India (28) China (22)

Figure 24.7 Key Drivers of Brazil, Russia, India, and Chinas Private
Equity Attractiveness This figure shows the key driving forces that shape the BRICs
attractiveness to PE investors. The development of these forces is skewed. Economic
size and growth and the capital markets are highly attractive for investors. By contrast,
investor protection, the human and social environment, and entrepreneurial culture based
on innovations are rather poorly developed. This finding is representative for all other
emerging markets. Source: The Venture Capital and Private Equity Country Attractiveness Index,
http://blog.iese.edu/vcpeindex/.

size to the United States. All other economies remain smaller than the United States and
have higher expected real growth rates except Russia. Nevertheless, on combining the
three sub-indicators, they all receive less than 100 points for economic activity.
Growth and development in emerging economies are not widespread, but rather
concentrated in particular hubs or regions. Further, compared to developed countries,
creating wealth is typically more directly allocated toward small elite groups and not to
larger parts of the population. This presents socioeconomic and political challenges in
these countries and also affects their PE activity. If such countries cannot transfer the
wealth effects of growth to a broader section of their population, they are unlikely to im-
prove the other key driving forces for PE attractiveness. If the pace of economic growth
were to slow down, these countries would see their attractiveness ranking deteriorate.

W I L L I N D O N E S I A , M E X I C O, T H E P H I L I P P I N E S , A N D T U R K E Y
F O L L O W B R A Z I L , R U S S I A , I N D I A , A N D C H I N A S PAT H ?
For different reasons, out of the previously mentioned emerging countries with large
populations, Indonesia, Mexico, the Philippines, and Turkey are arguably those with
the highest potential for growth in PE activity. Turkey and Mexico even rank ahead of
Brazil and Russia in the attractiveness index, and the Philippines and Indonesia follow
454 t r e n d s i n p r i vat e e q u i t y

0 20 40 60 80 100 120 140 160


1.1 Size of the Economy (GDP)

1.2 Expected Real GDP Growth

1.3 Unemployment

2.1 Size and Liquidity of the Stock Market

2.2 Total Trading Volume

2.3 IPOs and Public Issuing Activity

2.4 M&A Market

2.5 Debt and Credit Market

2.6 Bank Non Performing Loans to Toal Gross Loans

2.7 Financial Market Sophistication

3.1 Entrepreneurial Tax Incentives and Admin. Burden

4.1 Quality of Corporate Governance

4.2 Security of Property Rights

4.3 Quality of Legal Enforcement

5.1 Education and Human Capital

5.2 Labor Regulations

5.3 Bribing and Corruption

6.1 Innovation

6.2 Scientific and Technical Journal Articles

6.3 Burdens of Starting and Running a Business

6.4 Simplicity of Closing a Business

6.5 Corporate R&D

Brazil (40) Russian Federation


on (41) India (28) China (22)

Figure 24.8 More Highly Granulated Determinants of Brazil, Russia,


India, and Chinas Private Equity Attractiveness This figure breaks down
the BRICs key driving forces of PE attractiveness to more granulated data levels. It
reveals a typical pattern for emerging economies. Corporate governance and investor
protection remain major obstacles to be tackled. Perceived bribery and corruption, missing
innovations, and corporate R&D are further detriments of PE activity. Source: The Venture
Capital and Private Equity Country Attractiveness Index, http://blog.iese.edu/vcpeindex/.

closely. All four countries have moved up several attractiveness ranks between 2009 and
2014 and are therefore located in the upper-right corner of the matrix in Figure 24.6.
Focusing on their key driving forces of attractiveness in Figure 24.9, patterns similar to
the BRICs emerge.
Figure 24.9 shows that the key drivers of PE attractiveness are skewed in a similar
way for Indonesia, Mexico, the Philippines, and Turkey. Their economies are smaller
than the BRIC economies, but their real growth perspectives are promising, except
Mexico, as Figure 24.10 shows.
PE Gr owt h in I n t e rn at ion al M arke t s 455

1 Economic Activity
125
100
6 Entrepreneurial 75
2 Capital Market
Opportunities 50
25
0

5 Human and Social


3 Taxation
Environment

4 Investor Protection
and Corporate
Governance
Indonesia (46) Mexico (39) Philippines (42) Turkey (30)

Figure 24.9 Key Drivers of Private Equity Attractiveness for Indonesia,


Mexico, the Philippines, and Turkey This figure presents the key driving forces of
PE attractiveness for Indonesia, Mexico, the Philippines, and Turkey. It reveals, as for the
BRICs, the skewed development of the six key driving factors. Economic perspectives are
appealing to investors but other parameters merit improvement. Source: The Venture Capital
and Private Equity Country Attractiveness Index, http://blog.iese.edu/vcpeindex/.

Figure 24.10 presents the generally strong expected economic growth of the poten-
tial BRICs followers. The bar chart also illustrates the well-developed capital markets of
the countries in question. Still, labor market regulation, perceived corruption, and low
innovation capacity, in particular, the lack of R&D investment in Indonesia, are likely
to deter PE activity. Until now, only low and occasional PE activity has occurred in In-
donesia, Mexico, the Philippines, and Turkey. Local politicians may want to improve
investment conditions to attract foreign capital. In this case, the results provide some
guidance about the factors that need to be improved.

M I N I M U M M AT U R I T Y L E V E L F O R P R I VAT E E Q U I T Y A C T I V I T Y
The index ranks the PE attractiveness of countries for institutional investors based on
many socioeconomic data series. As noted previously, several sample countries might
not yet have a sufficiently developed PE infrastructure to create appropriate deal flow.
In fact, the activity in many emerging countries remains at very low and occasional
levels.
Finding a hurdle rate at which PE investment takes off is the next step. This is ac-
complished by comparing the index scores with the actual PE activity in the various
countries using regression analysis. The PE activity measure is the logarithm of an aver-
age of all the PE investments in a certain country over the previous three years. Again,
the logarithm is used to compensate for the large divergence in activity (e.g., activity
in the United States versus several emerging countries), and using a three-year aver-
age smoothes fluctuations over time. For some emerging countries in particular, annual
456 t r e n d s i n p r i vat e e q u i t y

0 20 40 60 80 100 120 140 160


1.1 Size of the Economy (GDP)
1.2 Expected Real GDP Growth
1.3 Unemployment
2.1 Size and Liquidity of the Stock Market
2.2 Total Trading Volume
2.3 IPOs and Public Issuing Activity
2.4 M&A Market
2.5 Debt and Credit Market
2.6 Bank Non Performing Loans to Toal Gross Loans
2.7 Financial Market Sophistication
3.1 Entrepreneurial Tax Incentives and Admin. Burden
4.1 Quality of Corporate Governance
4.2 Security of Property Rights
4.3 Quality of Legal Enforcement
5.1 Education and Human Capital
5.2 Labor Regulations
5.3 Bribing and Corruption
6.1 Innovation
6.2 Scientific and Technical Journal Articles
6.3 Burdens of Starting and Running a Business
6.4 Simplicity of Closing a Business
6.5 Corporate R&D

Indonesia (46) Mexico (39)) Philippines (42) Turkey (30)

Figure 24.10 More Highly Granulated Determinants of Private Equity


Attractiveness for Indonesia, Mexico, the Philippines, and Turkey This
figure breaks down the key driving forces of PE attractiveness to more granulated data
levels for Indonesia, Mexico, the Philippines, and Turkey. It shows the same principal
pattern as discussed for the BRICs. Corporate governance and investor protection
remain investment obstacles. Perceived bribery and corruption, missing innovations, and
corporate R&D deter additional PE activity. Source: The Venture Capital and Private Equity
Country Attractiveness Index, http://blog.iese.edu/vcpeindex/.

activity fluctuates strongly from peak levels to zero in following years. The scatter plot in
Figure 24.11 highlights the relationship between the index score and PE market activity.
Figure 24.11 shows a strong link between the attractiveness scores and actual PE
activity. Missing values marking no activity are filled with nulls and are excluded from
the regression. The regression analysis further shows that PE activity can be expected in
countries that have more than about 45 index points. This finding serves as the hurdle
rate for emerging PE activity. More PE activity depends on how many countries soon
pass this hurdle and on how the countries above the hurdle improve their scores.
PE Gr owt h in I n t e rn at ion al M arke t s 457

12

Log of Three-Year Average PE Investments in Million $


10

0
0 20 40 60 80 100
Current Index Scores

Figure 24.11 Attractiveness Scores and Private Equity Activity of 118


Countries This figure shows the relationship between the attractiveness index scores and
actual PE activity for the sample of 118 countries. The PE investment volumes are averaged
over the previous three years to smooth annual fluctuations. Further, the natural logarithm of
this average is calculated to consider the deviations of activity and missing values are filled with
nulls. The figure reveals a strong correlation of the two covariates. The regression line suggests
a cut-off value from which PE activity can be expected in a country. This cut-off value is about
45 and can be interpreted as a minimum maturity level that is required for PE activity to start.

P R I VAT E E Q U I T Y C O U N T R Y M AT U R I T Y A N D H I S TO R I C P R I VAT E
EQUITY RETURNS
Concurrent to the arguments about the required maturity to support PE activity, ana-
lyzing whether this maturity is also linked with the experienced performance from
investments in traditional and emerging markets is of particular interest. PE Investors
have the tendency to extrapolate historic results to the future. They prefer to allocate to
host countries in which they experienced higher returns. Unfortunately, performance
figures are still one of the best kept secrets in the PE industry, and the principle not
to disclose information on past returns is equally valid in both traditional and emerg-
ing markets. As pointed out previously, emerging PE markets are young with generally
low activity (despite some exceptions). Hence, very few transactions exist to calculate
achieved returns. Therefore, assessing PE performance is even more challenging for de-
veloping countries than for developed countries. Commercial data suppliers provide
only limited figures on the returns earned in particular deals. The only way to get reliable
performance data on a sufficient number of transactions for empirical analyzes is via an
extensive effort to collect private placement memoranda (PPM). A PPM is a document
458 t r e n d s i n p r i vat e e q u i t y

edited by a GP who raises a PE fund and solicits capital commitments from institutional
investors. It is a marketing document used for fundraising purposes. GPs provide in-
formation about their track records and the performance of individual transactions in
PPMs. The figures are audited and investors trust them. However, only successful GPs
raise a subsequent fund and edit a PPM. Therefore, academic researchers criticize the
use of performance figures from PPMs because these figures are upward biased. Nev-
ertheless, no reason exists to believe that this upward bias is different among particular
countries. This assumption means that benchmarking countries is feasible. Because the
countries are compared on a consistent relative basis, absolute terms are unimportant.
Using PPMs, Lopez-de-Silanes, Phalippou, and Gottschalg (2013) put together the
most comprehensive database on PE returns at the investment level, containing the per-
formance and characteristics of 7,453 investments, of which 1,694 were in emerging
countries. The first transaction considered was closed in 1971 and the last before 2006.
Ludovic Phalippou kindly provided data on aggregate country returns for the subse-
quent analyses. The returns are compiled as the mean average of gross internal rate of
return (IRR) of all transactions in a particular country. This average is certainly a rough
estimate, disregarding different fund vintage years, industries, deal structures, and de-
velopment cycles of the particular PE markets. Unfortunately, controlling for these
effects is impossible given the general deal data availability. Additionally, an IRR is a
capital- and time-weighted return measure that requires a reinvestment assumption and
that has aggregation issues as described in Phalippou (2009). However, the IRR pitfalls
are the same for all transactions and for all countries examined. Therefore, they should
not affect a cross-sectional country benchmarking approach.
With these aggregate performance measures, the link between PE country matu-
rity and past performance can be explored. The provided data can be matched with
the country maturity scores of 48 countries out of which 24 are emerging countries. At
least 10 observed IRRs are available for each country, thus providing a generally robust
estimate of the means. The correlation between the country maturity scores and a coun-
trys average gross IRR of historic PE transactions is 0.62. This relative high correlation
is depicted in Figure 24.12, which plots the average of the country returns on their index
scores.
Figure 24.12 suggests that the maturity score is not only a valid proxy for PE activ-
ity but also a good indicator for historic performance. The averages of historic gross
IRRs are larger in countries with higher maturity levels than in low-ranked countries.
The regression line has a slope of 0.55, signaling that a one point increase in the index
score comes with a 0.55 percent rise of average historic IRRs. Nevertheless, there are
outliers, meaning low-ranked countries with high returns and vice versa. Addition-
ally, strong dispersion of returns occurs within each particular country, driven by very
successful transactions and write-offs in any of them. Further, Lopez-de-Silanes et al.
(2013) calculate the IRRs gross of any fees. Fees are assumed to be higher for inves-
tors in immature markets with less competition among GPs. Therefore, allocations to
emerging countries are more costly for investors. This effect supports the result and
would increase the correlation if net returns to investors were considered.
This analysis allows the conclusion that countries can increase their attractiveness
to PE investors by improving the presented key driving forces. These forces contribute
to better deal-making conditions, which allow higher returns to investors. These higher
PE Gr owt h in I n t e rn at ion al M arke t s 459

60%

50%

40%
Historic Average IRR of Countries

30%

20%

10%

0%
0 10 20 30 40 50 60 70 80 90 100 110
10%

20%

30%

40%
Current Index Scores

Figure 24.12 Country Maturity Scores and Historic Private Equity


Returns This figure presents historic PE returns relative to the attractiveness scores for
48 countries of which 24 are emerging markets. The correlation between both covariates
is 0.62 and the regression line suggests that a one rank increase of the attractiveness score
comes with a 0.55 percent point increase of historic returns. Hence, despite the outliers
toward very high and low average PE returns, the attractiveness score is an indicator for the
success of PE transactions at the same time.

returns should direct more funds to these countries in the future because PE investors
tend to allocate to markets where they gained a positive experience.

Summary and Conclusions


This chapter presents international PE activity data since 2000, grouped into several
traditional and emerging regions. No clear pattern of global development for the PE
market exists. After the historic high in 2000, activity peaked again in 2007, and has
exhibited high volatility since 2007. While the trend in the traditional markets is rela-
tively flat, emerging economies are increasingly attracting investments. The number of
countries in which economic development is sufficiently mature to allow PE activity
has increased. Analysis reveals six key drivers of PE activity: (1) economic activity, (2)
depth of the capital market, (3) taxation, (4) investor protection and corporate gov-
ernance, (5) human and social environment, and (6) entrepreneurial culture and deal
opportunities.
460 t r e n d s i n p r i vat e e q u i t y

After breaking down these key driving forces into 51 data series, 118 traditional and
emerging PE markets are ranked. The United States ranks number one, followed by
the most developed economies. The BRICs receive relatively high rankings amid fierce
competition to attract international capital for PE investments, and several other emerg-
ing countries follow closely. However, the current ranking in absolute terms (or rather
the ranking quartile) is not the only important factor; medium-term development
provides additional information on the countrys PE attractiveness. Malaysia, Finland,
Chile, Turkey, Colombia, Mexico, Russia, the Philippines, Oman, Indonesia, Peru, Mo-
rocco, and the Baltics stand out for their PE attractiveness development between 2009
and 2014. Of these countries, Indonesia, Mexico, the Philippines, and Turkey appear
to have the greatest potential for increasing their PE activity due to the combination of
strong economic catch-up potential and the size of their populations.
In general, emerging markets provide investment opportunities and require substan-
tial financing for their expected economic growth, but this conclusion may be mislead-
ing. In many of these countries, accessing transactions is challenged by the relatively
immature nature of their institutional deal-supporting environment. Deals can be much
more cumbersome and costly. Further, weak investor protection may be coupled with
bureaucracy, bribery, or corruption, and can result in lower investment return. Inves-
tors should thoroughly examine both the advantages and disadvantages of emerging
market opportunities, as the advantages may come at a cost. Although the landscape
of investable emerging markets is enlarging, many countries are probably not yet suf-
ficiently mature to support the PE investment rationale. The index helps to assess the
maturity of countries for sustaining the PE business model. PE investments take off at
index scores above 45. Early entry in countries around this score may allow investors to
establish relationships with local partners and to gain local experience. Nevertheless,
too early of an entry is unadvisable.

Discussion Questions
1. Describe the global trend of the PE investment volume and the development in tra-
ditional versus emerging countries.
2. Discuss why PE investments are concentrated in particular economies.
3. Discuss the six key criteria that lead to deal flow and attract investors.
4. Explain why the BRICs attract increasing amounts of PE capital.
5. Discuss whether a threshold of economic development exists in which PE activity
emerges.

References
Baughn, Christopher, and Kent Neupert. 2003. Culture and National Conditions Facilitating En-
trepreneurial Start-ups. Journal of International Entrepreneurship 1:3, 313330.
Black, Bernard, and Ronald Gilson. 1998. Venture Capital and the Structure of Capital Markets:
Banks versus Stock Markets. Journal of Financial Economics 47:3, 243277.
PE Gr owt h in I n t e rn at ion al M arke t s 461

Bruce, Donald. 2000. Effects of the United States Tax System on Transition into Self-Employment.
Labor Economics 7:5, 545574.
Bruce, Donald. 2002. Taxes and Entrepreneurial Endurance: Evidence from the Self-Employed.
National Tax Journal 55:1, 524.
Bruce, Donald, and Tami Gurley. 2005. Taxes and Entrepreneurial Activity: An Empirical Investi-
gation Using Longitudinal Tax Return Data. Small Business Research Summary, 242.
Cetorelli, Nicola, and Michele Gambera. 2001. Banking Market Structure, Financial Dependence
and Growth: International Evidence from Industry Data. Journal of Finance 56:2, 617648.
Cullen, Julie, and Roger Gordon. 2002. Taxes and Entrepreneurial Activity: Theory and Evidence
for the U.S. NBER Working Paper, 9015.
Cumming, Douglas, Grant Flemming, and Armin Schwienbacher. 2006. Legality and Venture
Capital Exits. Journal of Corporate Finance 12:2, 214245.
Cumming, Douglas, Daniel Schmidt, and Uwe Walz. 2010. Legality and Venture Capital Govern-
ance around the World. Journal of Business Venturing 24:1, 5472.
Desai, Mihir, Paul Gompers, and Josh Lerner. 2006. Institutions and Entrepreneurial Firm Dynam-
ics: Evidence from Europe. Working Paper, Harvard NOM Research Paper, 0359.
Djankov, Simeon, Tim Ganser, Caralee McLiesh, Rita Ramalho, and Andrei Shleifer. 2008. The
Effect of Corporate Taxes on Investment and Entrepreneurship. NBER Working Paper 13756.
Djankov, Simeon, Rafael La Porta, Florencio Lopez-de-Silanes, and Andrei Shleifer. 2002. The
Regulation of Entry. Quarterly Journal of Economics 117:1, 137.
Djankov, Simeon, Rafael La Porta, Florencio Lopez-de-Silanes, and Andrei Shleifer. 2003. Courts.
Quarterly Journal of Economics 118:2, 453517.
Djankov, Simeon, Rafael La Porta, Florencio Lopez-de-Silanes, and Andrei Shleifer. 2008. The Law
and Economics of Self-dealing. Journal of Financial Economics 88:3, 430465.
Glaeser, Edward, Simon Johnson, and Andrei Shleifer. 2001. Coase vs. the Coasians. Quarterly
Journal of Economics 116:3, 853899.
Gompers Paul, and Josh Lerner. 1998. What Drives Venture Capital Fundraising? Brooking
Papers on Economic Activity, Microeconomics 1998, 149192.
Gompers Paul, and Josh Lerner. 2000. Money Chasing Deals? The Impact of Funds Inflows on the
Valuation of Private Equity Investments. Journal of Financial Economics 55:2, 281324.
Greene, Patricia. 1998. Dimensions of Perceived Entrepreneurial Obstacles. In Paul Reynolds, ed.,
Frontiers of Entrepreneurship Research, 4849. Babson Park, MA: Center for Entrepreneurial
Studies, Babson College.
Groh, Alexander, and Heinrich Liechtenstein. 2009. How Attractive Is CEE for Risk Capital Inves-
tors? Journal of International Money and Finance 28:4, 624647.
Groh, Alexander, and Heinrich Liechtenstein. 2011a. Determinants for Allocations to CEE Ven-
ture Capital and Private Equity Limited Partnerships. Venture Capital: An International Journal
of Entrepreneurial Finance 13:2, 175194.
Groh, Alexander, and Heinrich Liechtenstein. 2011b. International Allocation Determinants of In-
stitutional Investments in Venture Capital and Private Equity Limited Partnerships. Interna-
tional Journal of Banking, Accounting and Finance 3:23, 176206.
Groh, Alexander, and Heinrich Liechtenstein. 2011c. The First Step of the Capital Flow from Insti-
tutions to Entrepreneurs: The Criteria for Sorting Venture Capital Funds. European Journal of
Financial Management 17:3, 532559.
Groh, Alexander, Heinrich Liechtenstein, and Miguel Canela. 2010a. Limited Partners Percep-
tions of the Central Eastern European Venture Capital and Private Equity Market. Journal of
Alternative Investments 12:3, 96112.
Groh, Alexander, Heinrich Liechtenstein, and Karsten Lieser. 2010b. The European Venture Cap-
ital and Private Equity Country Attractiveness Indices. Journal of Corporate Finance 16:2,
205224.
Groh, Alexander, Heinrich Liechtenstein, and Karsten Lieser. 2014. The Global Venture Capital
and Private Equity Country Attractiveness Index. Available at http://blog.iese.edu/
vcpeindex/.
462 t r e n d s i n p r i vat e e q u i t y

La Porta, Rafael, Florencio Lopez-de-Silanes, Andrei Shleifer, and Robert Vishny. 1997. Legal De-
terminants of External Finance. Journal of Finance 52:3, 11311150.
La Porta, Rafael, Florencio Lopez-de-Silanes, Andrei Shleifer, and Robert Vishny. 1998. Law and
Finance. Journal of Political Economy 106:6, 11131155.
La Porta, Rafael, Florencio Lopez-de-Silanes, Andrei Shleifer, and Robert Vishny. 2002. Investor
Protection and Corporate Valuation. Journal of Finance 57:3, 11471170.
Lee, Sang, and Suzanne Peterson. 2000. Culture, Entrepreneurial Orientation and Global Com-
petitiveness. Journal of World Business 35:4, 401416.
Lerner, Josh, and Antoinette Schoar. 2005. Does Legal Enforcement Affect Financial Transactions?
The Contractual Channel in Private Equity. Quarterly Journal of Economics 120:1, 223246.
Lopez-de-Silanes, Florencio, Ludovic Phalippou, and Oliver Gottschalg. 2013. Giants at the Gate:
Investment Returns and Diseconomies of Scale in Private Equity. Journal of Financial and
Quantitative Analysis, forthcoming.
Megginson, William. 2004. Toward a Global Model of Venture Capital? Journal of Applied Corpo-
rate Finance 16:1, 89107.
Phalippou, Ludovic. 2009. The Hazards of Using IRR to Measure Performance: The Case of Pri-
vate Equity. Journal of Performance Measurement 12:4, 5566.
Roe, Mark. 2006. Political Determinants of Corporate Governance. Oxford: Oxford University
Press.
Schertler, Andrea. 2003. Driving Forces of Venture Capital Investments in Europe: A Dynamic
Panel Data Analysis. Working Paper, United Nations University, European Integration, Finan-
cial Systems and Corporate Performance (EIFC) Working Paper, 0327.
25
Diversification Benefits of Private
Equity Funds-of-Funds
AXEL BUCHNER
Assistant Professor of Finance, Passau University

MARKUS KUFFNER
MBA student, Passau University

Introduction
Since their first appearance in the late 1950s, private equity (PE) investments have ex-
perienced waves of excessive contractions and expansions but the overall trend shows
growth. Worldwide PE growth has spread to developing regions such as Asia and Africa.
Preqin (2012) finds a high level of growth, with about $3.0 trillion in assets under PE
management at the end of 2011. However, the growth has resulted in criticism of the
absolute level of PEs managerial compensation. The typical fee structure of PE funds of
a 2 percent management fee based on fund size and a 20 percent performance-related
fee comes increasingly under fire as PE grows in prominence. Fund sizes of $5 to $10
billion, which are no longer a rarity, lead to absolute management fees of $100 to $200
million regardless of whether the fund generates profits.
The principalagent problem increasingly becomes a major challenge because such
high payments to the fund sponsors do not lead to alignment of interest between the
fund managers and investors. The principalagent problem is defined as two parties
having different interests with the agent having more information than the principal,
thus the principal cannot be sure the agent always acts in the principals best interests.
This situation is especially true for PE funds-of-funds that charge an extra management
fee of around 0.5 percent and a performance-related fee of 5 to 10 percent on top of
the fees charged by the funds in the portfolio. Through investing in funds-of-funds, in-
vestors may benefit from internal cost savings, management know-how, and especially
from a better risk and return profile through diversification over hundreds of portfolio
companies.

463
464 t r e n d s i n p r i vat e e q u i t y

As Weidig and Mathonet (2004) show, funds-of-funds reduce the risk of losses. Yet,
the question remains of whether the benefits outweigh the costs given an investors
risk aversion. Unlike the hedge fund industry, only limited empirical evidence exists
to enable us to answer that question about PE. Brown, Goetzmann, and Liang (2004)
find that funds-of-hedge-funds offer poor value to investors. In contrast, Ang, Rhodes-
Kropf, and Zhao (2005) change the benchmark from the pool of single hedge funds
reported in databases to those hedge funds an investor can select without recourse to
funds-of-funds. They find funds-of-hedge-funds justify their fees on fees which can,
among other reasons, be attributed to offering access to top quartile funds and to de-
creasing volatility.
This chapter expands on the previous risk and return profile studies of PE funds-
of-funds by calculating certainty equivalents of final wealth and showing how much an
investor with a given risk aversion is willing to pay for the convenience that funds-of-
funds offer. The chapter proceeds as follows. The next section defines essential terms
and presents the characteristics of the asset class followed by a discussion of the specif-
ics of PE funds and funds-of-funds. These topics are followed by a presentation of the
fee structures for PE funds and funds-of-funds. The next section offers insights about
the underlying data of this chapters study and explains the simulation process of funds
and funds-of-funds. The chapter ends with a summary and conclusions.

Defining Private Equity, Private Equity Funds,


and Funds-of-Funds
PE investments are part of the alternative asset class similar to hedge funds and real estate
investments. When investing in PE, investors usually provide equity for privately held
companies by buying common stock, convertible securities, and warrants or other equity-
related securities. The goal is to sell these investments to get a capital gain. PE is some-
times split into two distinct asset classes: venture capital (VC) or buyout (BO). The stage
of a target companys position in its life-cycle determines its classification as VC or BO.

D E F I N I N G V E N T U R E C A P I TA L
According to the National Venture Capital Association (NVCA) (2014), a VC inves-
tor participates as a co-investor by providing capital, sector knowledge, and business
contacts. Jesch (2004) describes VC as intelligent equity invested with a hands-on
approach. VC investments focus on young target companies with high growth potential
(Gompers and Lerner 2001), often in the medicinal, biotechnological, and information
technological sectors. In contrast to a BO, a VC fund invests in the early stage of a com-
panys life cycle. The early stage generally can be divided into a seed, start-up, and ex-
pansion phase. During the seed phase, products and market concepts are created. In the
start-up phase, the products are developed up to the point of industrial application. In
the expansion phase, VC is used to expand investments targeting additional marketing
channels. To mitigate the high risk of early stage investments, capital allocation occurs
in so-called financing rounds dependent on achieving prior defined milestones often
through syndication.
B e n e fit s of F u n ds -of-F u n ds 465

DEFINING A BUYOUT
Buyout investments generally apply to already established companies in the later stage of
their life cycle. BOs increase in value by buying a majority stake in a company and then
controlling, managing, and restructuring it. PE financing via BOs is not restricted to
private companies because investors can also invest in public companies. In a so-called
public-to-private transaction, a PE company buys a majority stake in a listed target and
later delists it from the stock exchange. Most BOs are leveraged. In a leveraged buyout
(LBO), the PE company typically invests in a target that has already achieved positive
cash flows. Kaplan and Strmberg (2008) find that between 60 and 90 percent debt is
used to finance the purchase. The high debt burden usually gets secured by the target
firms assets and is serviced from its positive cash flows.
The generic term BO includes all subcategories of this distinct asset class. Some-
times, the terms BO and PE are used interchangeably in U.S. focused studies such as
Kaplan and Strmberg (2008), whereas in Europe the term PE is often used to describe
the industry as a whole, showing both VC and BO. One reason for this difference is the
European VC market is smaller and less developed than the U.S. VC market. In this
chapter, the term PE refers to both VC and BO investments.

D E F I N I N G P R I VAT E E Q U I T Y F U N D S
Metrick and Yasuda (2010b) define a private equity fund as a financial intermediary be-
tween investors and target companies that takes an active role in managing its portfolio
companies. To accomplish this, a PE fund pools capital from its investors and negotiates
an equity capital participation in target companies. Therefore, a PE fund is classified as
a pooled investment vehicle typically having legal status as a limited partnership. The
founder of the fund (i.e., the PE company), usually has the legal status of a limited li-
ability corporation. To manage the fund, a PE company establishes a special purpose
entity (SPE). This SPE acts as the general partner (GP) and investors assume the role as
limited partners (LPs). To benefit from tax deductions, borrowing costs that arise when
the purchase price of a target company is partly financed by debt, another SPE, called
the NewCo, is founded. The NewCo raises the required capital to acquire a target
company from the fund, the fund managers, and debt providers. Banks often serve as
the debt providers making up between 60 and 90 percent of leverage financing for BO
deals. Afterwards, the NewCo is merged with the target company in a debt-push-down-
merger so the debt burden leads to tax deductions on the books of the target (Wei
and Fischer 2010).

D E F I N I N G P R I VAT E E Q U I T Y F U N D S - O F - F U N D S
According to the European Private Equity and Venture Capital Association (EVCA)
(2013), PE funds-of-funds accounted for more than 20 percent of the capital raised by
European PE funds in 2012. By the end of 2012, 165 funds-of-funds were seeking capi-
tal in the market worldwide, targeting nearly $40 billion (Preqin 2013). Funds-of-funds
face challenging fundraising conditions because investors are becoming more sophis-
ticated. These investors are switching their investment focus directly to funds allowing
466 t r e n d s i n p r i vat e e q u i t y

them to avoid the double layer of fees. However, funds-of-funds still play an important
role in PE investing.
A fund-of-funds can be described as a managed portfolio of investments in many
funds. Therefore, a fund-of-funds is a financial intermediary between investors and funds
and functions as an LP in every fund investment. For most of the investors in funds-of-
funds, diversification benefits are the main motivation for their investments. Someone
investing in a typical fund-of-funds invests in about 20 funds with different manage-
ment styles, resulting in investments in roughly 400 companies. PE through funds-of-
funds offers a means of access to PE with two alternatives: make or buy (Sovran 2006).

Proprietary Funds-of-Funds
Two possibilities exist through the make alternative. Investors can build up a fund-of-
funds by investing in many PE funds on their own. This strategy comes with the advan-
tages of building long-term relationships with GPs, gaining industry knowledge, and
having full control over the investment strategy. As the second alternative, investors can
employ an external fund-of-funds management team that customizes the fund-of-funds
exposure. The asset class of self-made funds-of-funds is called proprietary funds-of-
funds allowing investors to meet individual strategic, diversification, and tax targets.
Proprietary funds-of-funds have some disadvantages. Building a professional in-
house PE investment division is expensive requiring a longer time to attract high-
quality managers. Gaining a foothold in PE funds is challenging for new investors be-
cause of the time and funding required to establish a good reputation and to gain access
to top-quartile funds. Compared to direct investments in capital-seeking target compa-
nies, investors have to pay fees when investing via a proprietary fund-of-funds. This fee
structure includes annual management fees charged by the funds in the fund-of-funds. In
case of an external advisory mandate, management fees exist in addition to these fees. The
committed capital also must exceed a high minimum amount over the investment period
to ensure the fund-of-funds can sufficiently diversify over several vintage years of under-
lying funds. In general, PE investments account for a small part of the overall investment
portfolio of institutional investors ( Jugel 2008). Therefore, the smaller, newer, and inex-
perienced institutional and private investors who want to invest in funds-of-funds cannot
or are unwilling to build up an in-house PE investment division. They are also unable to
employ an external fund management team, requiring them to buy instead.

Pooled Funds-of-Funds
Pooled funds-of-funds collect and pool capital from several investors to invest in PE
funds. This widespread standardized concept usually has the legal structure of a stand-
ard limited partnership with the investors as LPs and the providers as GPs. These GPs in
turn act as LPs in their PE fund investments.
Compared to a single fund investment, pooled funds-of-funds offer some distinct
advantages to investors. As Weidig and Mathonet (2004) show, pooled funds-of-funds
substantially reduce the risk due to diversification. For VC investments, their study re-
veals that when a loss occurs, it is on average 29 percent of the investment for funds and
just 4 percent for funds-of-funds. European BO funds reveal that when a loss occurs,
it averages 23 percent, but when a loss occurs for funds-of-funds, it averages only 1
percent of the initial investments. These results show a clear diversification benefit for
B e n e fit s of F u n ds -of-F u n ds 467

pooled funds-of-funds. Additionally, the standard deviation of the pooled funds-of-


funds is about 25 percent of the risk of a single fund investment. Further, investors of
pooled funds-of-funds benefit from internal cost savings because they do not have to
negotiate and monitor several single investments. Managers of pooled funds-of-funds
who have good relationships with GPs with strong, historical track records are given
access to otherwise closed funds. Moreover, pooled funds-of-funds only require a small
amount of committed capital compared to single fund investments and therefore are
suitable for new investors who want to test the waters. Pooled funds-of-funds further
offer the advantages of professional management teams and due diligence services, as
well as access to information that would otherwise be expensive or unobtainable.
The main drawback of pooled funds-of-funds is the higher cost caused by an ad-
ditional layer of fees. Besides the fund-of-funds fees, investors are responsible for all
fees from the underlying portfolio funds. Due to the previously discussed advantages of
funds-of-funds versus single fund investments, an investor might expect that funds-of-
funds perform better compensating for the fee-on-fees because of superior knowledge.
Existing PE studies rarely analyze this circumstance, but it is an accepted opinion. To
shed light on whether the benefits of funds-of-funds justify the fees-on-fees, this chap-
ter takes a closer look at the existing fee structure in funds and funds-of-funds.

Fee Structure of Private Equity Funds


and Funds-of-Funds
Fees exist at the level of individual portfolio funds. In practice, remunerating PE fund
GPs is structured differently and may be adapted to the life cycle of the fund, depending
on the agreement in the limited partnership agreement (LPA) (Robinson and Sensoy
2013). However, the different models can be divided into a fixed and variable component.
The fixed component is the management fee, which is used to compensate manage-
ment and cover operating costs. The LPA contains the parameters needed to calculate
the fixed rate fee, fee percentage, and fee basis. These parameters can vary across time.
Gompers and Lerner (1996) and Prowse (1998) state that contractual fee percentages
typically range from 1 to 3 percent. Using data on 837 PE funds with nearly $600 billion
in committed capital between 1984 and 2009, Robinson and Sensoy (2013) find that
92 percent of funds have committed capital as the initial fee basis and 75 percent have a
fee percentage between 1.5 and 2.5 percent. The median initial fee is 2.5 percent for VC
funds and 2.0 percent for BO funds.
Besides the previously widespread practice of calculating fees as a constant per-
centage of 2 percent of committed capital throughout the funds lifetime, Metrick and
Yasuda (2010a) describe three other methods of determining management fees. First,
some GPs set up a management fee that is based on committed capital but with a de-
clining percentage after the investment period (e.g., a 3 percent management fee during
the investment period with the fee dropping annually by 50 basis points afterwards).
The second method keeps the percentage amount of the management fee stable but
changes the fee basis from committed capital during the investment period to net in-
vested capital during the divestment period. Third, some GPs offer a combination of
both methodsa change in both fee percentage and fee basis.
468 t r e n d s i n p r i vat e e q u i t y

To better align the interests of GPs and LPs, GPs earn carried interest. Carried interest
or carry is the variable component of the GPs remuneration and is its share in the capital
gains of the fund or fund-of-funds. Thus, GPs are further compensated for the risk they
take with their ownership stake, which is usually 1 percent of total funds. The standard
carried interest of 20 percent is often linked to a preferred return or hurdle rate. GPs
only receive carry fees when LPs have at least received the initial investments including
management fees plus a preferred return of usually 8 percent for each year. The hurdle
rate is measured by the internal rate of return (IRR) on all previous cash flows, which
contains drawdowns, management fees, and distributions.
Most LPAs establish a catch-up period. During this period, GPs that reach the hurdle
rate criterion receive a pre-specified percentage of the following net returns resulting in
an arrangement that appears as though net returns have always been distributed in the
agreed carried interest ratio between GPs and LPs. After the catch-up, all further pro-
ceeds above the hurdle rate are distributed based on this ratio.
Fees also exist on the level of the pooled fund-of-funds. Weidig and Mathonet
(2004) identify an annual fixed management fee of 0.5 percent and a carry fee of 5 to
10 percent. Sovran (2006) finds management fees in the range of 0.5 to 1.5 percent for
funds-of-funds. The data used in the analysis contained in this chapter do not include
information about management fees, carry rates, or hurdle rates. Therefore, the general
220 rule with 2 percent management fees, 20 percent carried interest, and 8 percent
hurdle rate is adopted.
Table 25.1 shows the double layer of fees with a simple numerical example. Assume
a PE fund-of-funds has $100 million in committed capital, annual management fees of
1 percent based on committed capital, a carried interest rate of 10 percent, a hurdle rate
of 8 percent, and a fixed lifetime of 10 years. In this example, only one underlying fund
exists in the portfolio and fees due will be paid on top of the committed capital. The
underlying portfolio fund thus also has $100 million in committed capital. This fund
has an annual management fee of 2 percent, a carried interest rate of 20 percent, and a
hurdle rate of 8 percent and a total lifetime of 10 years. Moreover, the committed capital
is fully drawn in the first four years. At the beginning of year five, the underlying fund
begins to liquidate some of its portfolio companies and continues to do so until the last
target company is sold at the end of year 10. The funds proceeds are immediately dis-
tributed to the fund-of-funds which then pass the proceeds to the investor.
Table 25.1 shows that the fund made profits of $150 million net of investment and
gross of fees at the end of its lifetime. The GPs for the PE fund earn $2 million a year in
management fees, whereas the GPs for the fund-of-funds earn $1 million in manage-
ment fees every year. The net cash flow based calculations of IRR show that the GPs of
the PE fund receive carried interest in year eight for the first time, as the hurdle rate cri-
terion of 8 percent is first met in year seven. The fund-of-funds GPs get carried interest
in year seven for the first time. Generally, the carried interest for GPs depends on the
existence of a catch-up clause in the LPA.
Assuming there is no catch-up clause in the LPA on the fund level, GPs receive 20
percent of capital distributions at the beginning of year eight continuing to the end of
year 10 to such a degree as the IRR is always above the hurdle rate. In this case, GPs re-
ceive $3.60 million in carry after year eight, $7.60 million after year nine, and $7.20 mil-
lion after year 10. In total, GPs earn $18.40 million in carried interest. In terms of total
Table 25.1 Calculation of Fees

Year 1 2 3 4 5 6 7 8 9 10 Total
Fund
CF 50 30 10 10 60 85 7 20 40 38 150
Fees 2 2 2 2 2 2 2 2 2 2 20
CF(net 52 32 12 12 58 83 5.00 18.00 38.00 36.00 130
of fees)
IRR 1 1 1 1 0.19 0.07 0.08 0.11 0.15 0.17 0.17
Carry (no 0 0 0 0 0 0 0 3.60 7.60 7.20 18.40
catch-up)
Carry 0 0 0 0 0 0 0 11.20 7.60 7.20 26
(catch-up)
Fund-of- Fund and fund-of-funds without catch-up
funds
CF 50 30 10 10 58 83 5 14.40 30.40 28.80 119.60
Fees 1 1 1 1 1 1 1 1 1 1 10
CF(net 51 31 11 11 57 82 4 13.40 29.40 27.80 109.60
of fees)
IRR 1 1 1 1 0.18 0.08 0.09 0.11 0.14 0.16 0.16
Carry 0 0 0 0 0 0 0.40 1.34 2.94 2.78 7.46
Fund-of- Fund and fund-of-funds with catch-up
funds
CF 50 30 10 10 58 83 5 6.80 30.40 28.80 112
Fees 1 1 1 1 1 1 1 1 1 1 10
CF (net 51 31 11 11 57 82 4 5.80 29.40 28.80 102
of fees)
IRR 1 1 1 1 0.18 0.08 0.09 0.10 0.13 0.15 0.15
Carry 3.90 0.58 2.94 2.78 10.20
Minimum 3 3 3 3 3 3 3.40 7.94 13.54 12.98 55.86
total fees
Maximum 3 3 3 3 3 3 6.90 14.78 13.54 12.98 66.20
total fees

Note: This table shows fees of a fund and a fund-of-funds consisting of only one fund based on cash
flows (CF). The fees of the fund are $46 and the fees of the fund-of-funds are $20.20, leading to total
fees of $66.20, assuming a catch-up period is agreed in both LPAs. In the case of no catch-up period,
total fees are $55.86. Assuming values are in millions of U.S. dollars.
470 t r e n d s i n p r i vat e e q u i t y

net profits this corresponds to 14 percent. On the fund-of-funds level, GPs are entitled
to receive carried interest in year seven. If the LPA between the fund-of-funds and its
investors contains no catch-up clause, fund-of-funds GPs earn $0.40 million after year
seven and $1.34, $2.94, and $2.78 million the three following years. Total earnings are
$7.46 million or about 7 percent in terms of total net profits. On the fund level, as well
as the fund-of-funds level, carried interest amounts to far less than the agreed rate of
20 percent for fund GPs and 10 percent for fund-of-funds GPs. In total, fund GPs will
receive $7.60 million less than anticipated, whereas fund-of-funds GPs receive $3.5 mil-
lion less than the agreed proportion of 10 percent of net profits.
A catch-up clause in the LPA enables GPs to receive the exact level of the agreed upon
carry rate of all net profits, as long as net profits exceed the hurdle rate. As assumed in
Table 25.1, the 100 percent catch-up provision allows fund managers to earn $11.20 mil-
lion after the eighth year to catch up to their share of 20 percent in net profits. Besides the
$7.60 million and $7.20 million in the end of years nine and 10, respectively, total carry
is $26 million, which is exactly the carried interest level of 20 percent times net profits of
$130 million. If the fund-of-funds GPs also have a catch-up clause in their LPA, then they
are entitled to receive $3.90 million after year seven plus another $0.58, $2.94 and $2.78
million in carry the following three years. Fund-of-funds GPs thus receive $10.20 million
in carry, which is exactly the carry level of 10 percent times net profits of $102 million.
The performance of PE funds and funds-of-funds is often measured by the multiple,
which is the capital distributed to LPs proportional to the capital paid-in by LPs. For
example, if a fund sells all of its portfolio companies and receives $250 million and the
investor invested $100 million, then the multiple is 2.50 ($250 million / $100 million).
Table 25.2 offers a closer look at the multiples of the fund and the fund-of-funds and
reveals the effect of the double layer of fees.
An investor in the PE fund receives a multiple of 2.50 gross of fees. The net of fees mul-
tiple is 2.03 without a catch-up clause and 1.96 with one. If the investor chooses a pooled
fund-of-funds vehicle, the multiples clearly decline. In the case of nocatch-up-clauses at
both the fund and fund-of-funds level, the net of fees multiple is 1.84. However, a catch-
up period is most commonly employed when the LPA contains an agreed upon preferred
return. Therefore, the net of fees fund-of-funds multiple with a catch-up clause in LPA of
both the fund and the fund-of-funds is 1.75. The difference between a single fund invest-
ment with a catch-up and a fund-of-funds investment with a catch-up is 0.21 for net of fees
multiples. Are the advantages of pooled funds-of-funds worth the additional fees, which in
this example account for $21million in costs (21 percent of $100 million)?

Table 25.2 Multiples

Fund Fund-of-funds Fund-of-funds


(Fund No Catch-up) (Fund Catch-up)
Gross of fees 2.50 2.03 1.96
Net of fees (no catch-up) 2.03 1.84 1.78
Net of fees (catch-up) 1.96 1.81 1.75

Note: This table shows the resulting multiples of the example in Table 25.1 on a gross and net of
fees basis.
B e n e fit s of F u n ds -of-F u n ds 471

Data Set
The following analysis uses the Preqin database, which is one of the leading commer-
cial data set providers for the PE industry. The data set contains data on vintage years,
liquidation status, fund sizes, regional focuses, residual values to paid-in (RVPI), mul-
tiples of invested capital, and IRR for 6,223 funds and funds-of-funds with vintage years
spanning from 1969 to 2013. To address the question of whether the double layer of
fees in funds-of-funds is acceptable, the certainty equivalent levels of final wealth must
be computed to compare how much an investor with a given risk aversion has to invest
in a fund (fund-of-funds) to reach the same utility of wealth upon liquidation of an al-
ternative investment in a fund-of-funds (fund). Only multiples of liquidated funds and
funds-of-funds, which contain the whole cash flows, provide the complete truth of the
performance. Given the data set contains only 31 liquidated funds-of-funds with mul-
tiples, funds-of-funds are simulated based on liquidated VC and BO funds. The detailed
subdivision of PE fund types in the Preqin data set first requires an aggregation of sub-
categories under the main categories venture capital and buyout. In this study, the
generic term venture capital stands for Preqins sub-categories Early Stage, Early
Stage: Seed, Early Stage: Start-Up, Expansion/Late Stage, and Venture (General).
The sub-categories Buyout, Growth, Special Situations, and Turnaround are sub-
sumed under the generic term buyout funds.

D E S C R I P T I V E A N A LY S I S
Table 25.3 reports basic descriptive statistics for liquidated BO and VC funds as well
as funds-of-funds. After the aggregation, 375 liquidated BO funds, 508 liquidated VC
funds, and 31 liquidated funds-of-funds remain. The BO (VC) funds have an average
multiple of 2.24 (2.75), whereas the median multiple is 1.87 (1.90). The BO (VC) fund
multiples range from 0.00 to 17.32 (0.01 to 42.45). VC fund multiples vary more, with
a standard deviation of 3.63 compared to BO fund multiples with a standard deviation
of 1.78. Furthermore, the loss frequency (i.e., the frequency of funds with a multiple less
than one) is 13 percent for BO funds and 18 percent for VC funds.
The figures clearly show that PE fund-investing is not risk-free because the average
loss given a loss (i.e., a multiple less than one) is 37 percent for BO funds and 49 percent
for VC funds. The risk of a total loss is virtually nonexistent since only two BO funds
have a multiple of zero. The risk-return ratio, which gives the return per unit risk, is cal-
culated as the average multiple minus one divided by the standard deviation resulting in
values of 0.69 for BO funds and 0.48 for VC funds. These findings reveal higher risk of
VC funds compared to BO funds.
The 31 liquidated funds-of-funds consist of a number of factors, which may suggest
that risk-averse investors are better off with a fund-of-funds investment instead of a
single fund investment. The minimum multiple of 1.01 highlights investors did not face
losses. With a risk-return ratio of above one, funds-of-funds seem to fulfill all attributed
advantages, but the maximum multiple of 7.85 shows the disadvantage of the additional
diversification in funds-of-funds. Besides the fees-on-fees, investors pay for the benefits
of diversification and a lower downward risk, with reduced chances of achieving high
returns.
472 t r e n d s i n p r i vat e e q u i t y

Table 25.3 Sample 1: Liquidated Funds


Sample I BO Funds VC Funds Funds-of-funds
N 375 508 31
Minimum multiple 0.00 0.01 1.01
Median multiple 1.87 1.90 2.27
Average multiple 2.24 2.75 2.49
Maximum multiple 17.32 42.45 7.85
Standard deviation 1.78 3.63 1.31
Probability of a loss 0.13 0.18 0.00
Average loss given a loss 0.37 0.49 0.00
Probability of a total loss < 0.01 0.00 0.00
Risk-return ratio 0.69 0.48 1.14
Regional focus
Africa 1 3 0
Americas 9 5 0
Asia 21 36 1
Australasia 6 7 0
Diversified Multi-Regional 1 0 0
Europe 102 70 12
Middle East and Israel 1 10 0
United States 234 377 18

Note: This table provides statistical information of the first subset of Preqins database. The regional
focus of the funds is not surprising with about 69 percent in the United States. Another 19.5 percent of
funds focus on Europe and the remaining 11.5 percent on the rest of the world.

In sum, funds-of-funds reduce the downward risk compared to single funds. Con-
versely, the odds of achieving exceptional returns are clearly limited due to diversifica-
tion of more than 10 to 20 funds.
The kernel density estimations (i.e., estimations of the probability density function
of a random variable) in Figure 25.1 illustrate what the numerical findings of median,
average, minimum, and maximum multiples already suggest. The mean always exceeds
the median, implying that distributions of the multiples are positively skewed. The dis-
tributions have a longer tail to the right as a result of a few high multiples or outliers.
For BO funds, five multiples and for VC funds, 15 multiples are higher than 10, which
is 1 percent and 3 percent of the sample, respectively, in terms of all multiples of the
particular asset class. Moreover, the density estimations show the higher risk for VC
funds compared to BO funds, as more mass under the curve is concentrated in the lower
portion of the distribution. Further, the upward potential of funds-of-funds is clearly
limited compared to single fund investments.
B e n e fit s of F u n ds -of-F u n ds 473

0.5
Liquidated buyout funds
Liquidated venture capital funds
0.4 Liquidated funds-of-funds

0.3
Probability

0.2

0.1
10 and
more
0
0 2 4 6 8 10
Multiple

Figure 25.1 Kernel Density Estimates for Liquidated Funds and Funds-of-
Funds Note: In this figure, the kernel density estimates illustrate the numerical findings.
Liquidated VC funds are positively skewed with a value of 5.97 and have a kurtosis of
50.77. Liquidated BO funds are also positively skewed with a value of 4.12 and a kurtosis of
about 25, but are not as leptokurtic as VC funds. Liquidated funds-of-funds come closest
to a normal distribution, with a positive skewness of 2.4 and a kurtosis of 8.6.

Restricting the analysis to liquidated funds could limit the results as funds with more
recent vintage years would systematically be omitted. To mitigate this problem, Kaplan
and Schoar (2003) increase their data by adding largely liquidated funds to their sample.
This chapter also follows this common approach and adds non-liquidated BO and VC
funds to the sample if their RVPI is not higher than 10 percent. Treating the current net
asset value (NAV) at the end of the observation period as a final cash flow will have a neg-
ligible effect on the results. All funds that are not liquidated by the end of 2006 and satisfy
this condition are added to the liquidated funds to form the extended second sample in
Table 25.4 called mature funds. This sample consists of 555 BO and 651 VC funds.
In the second sample, median multiples fell to 1.82 for BO funds and 1.67 for VC
funds. This change provides some indication of conservative NAVs. It may also result
from worse exit conditions during the stock market crash from 2001 to 2003, and/or to
high target company prices in the course of the stock market rally from the late 1990s
to 2001. Further, the risk of BO funds, measured as the standard deviation, fell to 1.55
and to 3.32 for VC funds, respectively. However, the risk-return ratio stayed the same
for BO funds at 0.70, as the average multiple also fell to 2.09. For VC funds, the average
multiple is 2.47, which results in a slightly lower risk-return ratio of 0.44.
A closer look at the geographical distribution of the funds reveals that the United
States still has the biggest market for PE investing, with 76 percent of VC funds, and 62
percent of BO funds focusing on this region. Figure 25.2 shows the kernel density esti-
mates for the extended sample. A closer look at the skewness and kurtosis of the samples
shows that the second sample achieves higher values in both measures.
Figure 25.3 summarizes the differences in funds per vintage year between the two
samples. By adding largely liquidated funds to the second sample, the study does a better
job considering funds with more recent vintage years (i.e., the late 1990s and early 2000s).
474 t r e n d s i n p r i vat e e q u i t y

Table 25.4 Sample 2: Mature Funds


Sample 2 BO Funds VC Funds
N 555 651
Minimum multiple 0.00 0.01
Median multiple 1.82 1.67
Average multiple 2.09 2.47
Maximum multiple 17.32 42.45
Standard deviation 1.55 3.32
Probability of a loss 0.14 0.25
Average loss given a loss 0.35 0.48
Probability of a total loss 0.00 0.00
Risk-return ratio 0.70 0.44
Regional focus
Africa 3 3
Americas 12 7
Asia 28 38
Australasia 10 8
Diversified Multi-Regional 1 0
Europe 154 86
Middle East and Israel 1 13
United States 346 496

Note: This table shows the second sample of both liquidated and not yet liquidated funds that have
an RVPI of not higher than 10 percent. Thus, this sample also includes more recent funds from the late
1990s and early 2000s.

S I M U L AT I N G F U N D S - O F - F U N D S
A bootstrap technique is used to simulate funds-of-funds. The simulated funds-of-funds
invest in 20 funds, which is a reasonable number of investments for a fund-of-funds
(Weidig and Mathonet 2004). The investments are spread over an investment period of
five years, with four investments per year. The simulation is conducted in two steps. In
the first step, the simulation randomly chooses the fund-of-funds vintage year within the
range of underlying fund vintage years. Each vintage year is equally weighted. Second,
the simulation randomly chooses four fund multiples per investment year starting with
the vintage year and stopping after five consecutive years of investing. It then assembles
them to form a fund-of-funds. In the last step, the sum of these multiples is divided by
the number of underlying funds, yielding the multiple achievable by investing in a fund-
of-funds containing the chosen funds. To compare the results of funds and funds-of-
funds, a similar bootstrap technique is applied to simulate 10,000 funds.
0.5
Mature buyout funds
Liquidated buyout funds
0.4 Mature venture capital funds
Liquidated venture capital funds
Probability

0.3

0.2

0.1 10 and
more
0
1 0 1 2 3 4 5 6 7 8 9 10 11
Multiple

Figure 25.2 Comparing Kernel Density Estimates for Both Samples


Note: This figure shows that kernel density estimates for the second sample have
consistently higher values for skewness and kurtosis leading to even fatter tails than for the
sample of liquidated funds. The sample of mature VC funds has a skewness of 6.32 and a
kurtosis of 58.96. The skewness of mature BO funds is 4.45 and their kurtosis is 31.73.

60
Black: Liquidated VC funds
50 White: Mature VC funds

40

30

20

10

0
1969 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006

70
Black: Liquidated BO funds
60 White: Mature BO funds
50

40

30

20

10

0
1977 1980 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007

Figure 25.3 Funds per Vintage Year Note: This figure shows that extending the
sample of liquidated funds with funds that have an RVPI not higher than 10 percent takes
better account of funds with more recent vintage years. Furthermore, this helps to mitigate
an upward bias in multiples given that liquidated BO (VC) funds have an average multiple
of 2.24 (2.75), whereas the added funds have an average multiple of just 1.79 (1.46).
476 t r e n d s i n p r i vat e e q u i t y

For Sample 1, the average multiple of BO (VC) funds-of-funds is just 1.91 (2.17)
compared to 2.43 (2.52) of simulated funds. The maximum multiple of 4.20 (7.18)
is only about one-third of the maximum multiple of BO funds, and one-sixth of the
maximal possible VC funds multiple. The minimum multiples, which are 0.79 for BO
funds-of-funds and 0.51 for VC funds-of-funds, show the average loss given a loss must
be much lower for funds-of-funds than for funds. If an investor of a synthetic BO fund-
of-funds faces any loss, occurring in 1 percent of all investments in this case, an average
loss of 7 percent of committed (invested) capital is observed.
As Table 25.5 shows, an investor of a simulated VC fund-of-funds has a 3 percent
probability of a loss. If a loss occurs, it is on average about 16 percent. This is compared
to a 21 percent probability of enduring a loss for single funds, in which case the expected
loss averages 47 percent. Moreover, the risk of BO (VC) fell to 0.57 (0.84) leading to a
risk-return ratio of 1.60 (1.38).
Comparing the figures of the extended sample of mature funds and funds-of-funds
(Sample 2) in Table 25.6 with Sample 1 reveals that adding non-liquidated funds with
more recent vintage years worsens most performance figures. However, the risk also de-
creases making investors in Sample 2 less willing to pay the extra fees of funds-of-funds,
than investors in the first sample.
For both samples, the expected advantages and disadvantages of funds-of-funds
versus funds hold true. While for both samples much lower risks, higher risk-return
ratios, and lower probabilities of any loss are observed, lower maximum and average
multiples occur due to the diversification over 20 funds.

Table 25.5 Liquidated Funds and Funds-of-Funds


Sample 1 BO Funds BO Funds-of- VC Funds VC Funds-
Funds of-Funds
N 10,000 10,000 10,000 10,000
Minimum multiple 0.00 0.79 0.01 0.51
Median multiple 2.09 1.80 1.82 1.96
Average multiple 2.43 1.91 2.52 2.17
Maximum multiple 13.52 4.20 42.45 7.18
Standard deviation 1.78 0.57 3.07 0.84
Probability of a loss 0.11 0.01 0.21 0.03
Average loss given a loss 0.39 0.07 0.47 0.16
Probability of a total loss < 0.01 0.00 0.00 0.00
Risk-return ratio 0.80 1.60 0.49 1.38

Note: This table shows the numerical findings for 10,000 simulated funds and funds-of-funds
out of the sample of liquidated funds. As not every vintage year satisfies the condition of at least four
liquidated funds, the figures below represent the vintage years 1984 through 2002 for BO funds and
1980 through 2002 for VC funds and funds-of-funds.
B e n e fit s of F u n ds -of-F u n ds 477

Table 25.6 Mature Funds and Funds-of-Funds

Sample 2 BO Funds BO Funds- VC Funds VC Funds-


of-Funds of-Funds
N 10,000 10,000 10,000 10,000
Minimum multiple 0.00 0.73 0.01 0.44
Median multiple 2.01 1.71 1.74 1.93
Average multiple 2.33 1.85 2.39 2.08
Maximum multiple 13.52 4.30 42.45 6.70
Standard deviation 1.70 0.53 2.87 0.80
Probability of a loss 0.10 < 0.01 0.22 0.05
Average loss given a loss 0.35 0.06 0.43 0.18
Probability of a total loss <0.01 0.00 0.00 0.00
Risk-return ratio 0.78 1.59 0.48 1.35

Note: This table represents the numerical findings for the sample of mature funds and funds-of-
funds. The vintage years range from 1984 to 2005 for BO funds and from 1980 to 2002 for VC funds
and funds-of-funds.

Next, the study determines how much an individual investor should be pre-
pared to pay when investing in a fund-of-funds using the certainty equivalent of
final wealth concept. Investors are assumed to have power utility, displaying con-
stant relative risk aversion (CRRA). To calculate the investors average utility
of funds and funds-of-funds investments, the utility function in Equation 25.1 is used:

w(1 )
u( w) = , (25.1)
1
where w is the final wealth achieved and is the coefficient of risk aversion. By solving
for CE, Equation 25.2 emerges:

u(CE) = E[u( w)]. (25.2)

Table 25.7 displays the certainty equivalents of final wealth (CE).


Inserting these CEs into Equation 25.3:

=[1(CE fund/CE fund-of-funds)]100 (25.3)

yields the answer to the question of how much an investor with a given risk aversion
is prepared to pay for the benefits of funds-of-funds. The variable is the percent-
age amount of wealth a fund-of-funds investor could give up today and still be indiffer-
ent between investing in funds or funds-of-funds. Table 25.8 summarizes the levels of
indifference.
478 t r e n d s i n p r i vat e e q u i t y

Table 25.7 Certainty Equivalents


Risk Aversion Levels of 0.5 0.8 1.5
Venture capital
Liquidated funds 2.05 1.81 1.24
Liquidated funds-of-funds 2.09 2.05 1.97
Mature funds 1.97 1.76 1.23
Mature funds-of-funds 2.01 1.97 1.88
Buyout
Liquidated funds 2.19 2.04 0.75
Liquidated funds-of-funds 1.87 1.85 1.80
Mature funds 2.11 1.98 1.12
Mature funds-of-funds 1.82 1.79 1.75

Note: This table reports certainty equivalent multiples for three different investors with risk
aversion levels of equal 0.5, 0.8, and 1.5. Certainty equivalents represent the amount of money an
investor would accept today in order to avoid the risk of an investment.

R E S U LT S
Assume an investor with $100 of initial wealth and a risk aversion of = 1.5 decides to
invest in a VC fund-of-funds instead of a fund. In the liquidated sample, the investor
could give up 36.72 percent or $36.72 of wealth today, and invest $63.28 in the fund-
of-funds and still end with the same level of utility similar to investing $100 in a VC
fund. The figures in Table 25.8 are presented gross of fund-of-funds fees. Considering
an additional fund-of-funds management fee of 1 percent, a riskless rate of r = 0 percent,
and a lifetime of funds-of-funds of 10 years, an extra 10 percent in fees are recorded.
For simplicity, carried interest, which depends on the agreements in the LPA and could
result in another 10 percent in additional fees, is not considered. In the final step, the
net present value (NPV) of the extra management fees is compared to the amount an

Table 25.8 Levels of Indifference


Risk Aversion Levels of 0.5 0.8 1.5
Liquidated VC funds and funds-of-funds 2.01 11.88 36.72
Mature VC funds and funds-of-funds 1.82 10.75 34.52
Liquidated BO funds and funds-of-funds 16.91 10.40 58.07
Mature BO funds and funds-of-funds 16.33 10.48 36.12

Note: This table reports the gross of fees values in the percentage amounts of wealth a fund-of-funds
investor with a given risk aversion could give up today and still be indifferent between a fund and a
fund-of-funds investment. In the case of a positive value, the investor would be better off with investing
in funds-of-funds.
B e n e fit s of F u n ds -of-F u n ds 479

investor could give up today and still be indifferent between an investment in a fund or a
fund-of-funds. In the current example, the investor will still invest in the fund-of-funds
because $36.72 is higher than the NPV of the additional fees of $10. Investors do not
mind the additional fees in funds-of-funds if the relationship in Equation 25.4 holds:

NPV additional fees ( initial wealth). (25.4)

The corresponding is 0.74 for liquidated VC, 0.77 for mature VC, 1.23 for liquidated
BO, and 1.31 for mature BO. The sample of liquidated VC investments shows that in-
vesting in a fund-of-funds and accepting the double layer of fees adds value, if the inves-
tors coefficient of risk aversion is at least 0.74 or higher. The figures clearly reflect the
higher risk of VC investments. While investors in mature VC only need to have a risk
aversion coefficient of 0.77 to accept the double layer of fees, investors in mature BOs
only are prepared to pay them, if their is at least 1.31.
By using the same population of underlying funds and randomly selecting them to
build a fund-of-funds, the ability of selecting funds is assumed to be equal between an
unskilled investor and a fund-of-funds manager. Now assume that funds-of-funds man-
agers add value by their ability to invest only in the best 75 percent of funds of each vin-
tage year. To simulate these funds-of-funds, the simulation process excludes the worst
25 percent of funds of every vintage year.
Table 25.9 reports the simulation results. The minimum multiples of BO funds-of-
funds are 1.21 for liquidated funds-of-funds and 1.20 for mature funds-of-funds. Both
samples of BO funds-of-funds have lower average multiples compared to their corre-
sponding funds-samples, but have risk-return ratios that are roughly twice as high, which
can be attributed to much lower risks of 0.57 and 0.55. VC funds-of-funds even show
higher average multiples than the underlying single funds. Their standard deviations de-
crease from 3.07 for liquidated funds to 1.00 for liquidated funds-of-funds, and from
2.87 for mature funds to 0.94 for mature funds-of-funds, yielding risk-return ratios of
1.65 and 1.62, which is more than three times the ratios achieved by investing in funds.
Table 25.10 establishes the investors marginal risk aversion coefficients for two
cases: funds-of-funds managers with and without selection ability. In case the funds-of-
funds managers do not have any selection ability, investors in the sample of liquidated
(mature) VC funds-of-funds need a minimum risk aversion coefficient of 0.74 (0.77).
Liquidated (mature) BO funds-of-funds breakeven only if investors have a risk aversion
of at least 1.23 (1.31) due to the lower risk of BO funds in the data set. If funds-of-funds
managers have the ability to only invest in the best 75 percent of funds of every vintage
year, the marginal coefficients of risk aversion are 0.16 for both samples of VC funds-of-
funds and 0.99 (1.07) for liquidated (mature) BO funds-of-funds. These findings show
the advantages of a professional fund-of-funds manager, especially for the high-risk
asset class of VC investments.

Summary and Conclusions


This chapter examines the value proposition of the double layer of fees in funds-of-
funds. The methods used in this chapter include a bootstrap technique to simulate funds
and funds-of-funds, consider the individual investors degree of risk aversion, and apply
480 t r e n d s i n p r i vat e e q u i t y

Table 25.9 Selection Ability


Funds-of-funds Liquidated Mature Liquidated Mature
BO BO VC VC
N 10,000 10,000 10,000 10,000
Minimum multiple 1.21 1.20 0.66 0.63
Median multiple 2.00 1.93 2.36 2.30
Average multiple 2.13 2.08 2.64 2.52
Maximum multiple 4.68 4.52 8.86 8.71
Standard deviation 0.57 0.55 1.00 0.94
Probability of a loss 0.00 0.00 < 0.01 0.02
Average loss given a loss 0.00 0.00 0.09 0.11
Probability of a total loss 0.00 0.00 0.00 0.00
Risk-return ratio 1.98 1.97 1.65 1.62
Funds Liquidated Mature Liquidated Mature
BO BO VC VC
N 10,000 10,000 10,000 10,000
Minimum multiple 0.00 0.00 0.01 0.01
Median multiple 2.02 1.97 1.82 1.74
Average multiple 2.38 2.32 2.52 2.39
Maximum multiple 13.52 13.52 42.45 42.45
Standard deviation 1.79 1.70 3.07 2.87
Probability of a loss 0.11 0.12 0.21 0.22
Average loss given a loss 0.36 0.34 0.47 0.43
Probability of a total loss 0.01 < 0.01 0.00 0.00
Risk-return ratio 0.77 0.77 0.49 0.48

Note: This table shows the numerical findings for the simulated funds and funds-of-funds assuming
that funds-of-funds managers add value by selecting only the best 75 percent of funds. The numbers
represent the vintage years 1986 through 2001 for liquidated BO funds, 1986 through 2003 for mature
BO funds, and 1980 through 2002 for liquidated and mature VC funds.

Table 25.10 Marginal Coefficients of Risk Aversion


Asset Class without Selection Ability with Selection Ability
Liquidated VC 0.74 0.16
Mature VC 0.77 0.16
Liquidated BO 1.23 0.99
Mature BO 1.31 1.07
B e n e fit s of F u n ds -of-F u n ds 481

a certainty equivalent level of final wealth concept. To calculate marginal coefficients of


individual risk aversion in which an investor is indifferent between investing in a fund
or a fund-of-funds, the investor is assumed to have power utility. According to the anal-
ysis, VC funds-of-funds simulated of liquidated funds pay off if the investor has a risk
aversion coefficient of at least 0.74. This coefficient drops to 0.16 if the management can
detect the worst performing 25 percent of funds of each vintage year and avoids invest-
ing in these funds. The same figure applies to the sample of mature VC funds-of-funds.
Investors who must choose whether to invest in the sample of liquidated (mature) BO
funds or funds-of-funds find the latter attractive if their risk aversion coefficient is not
less than 1.23 (1.31). But if the fund-of-funds management adds value by investing only
in the best 75 percent of funds for every underlying vintage year, investors in liquidated
(mature) BO choose funds-of-funds, given is at least 0.99 (1.07). Friend and Blume
(1975) find average risk aversion coefficients to be at least 2.0. Applying this figure
leads to the conclusions that PE funds-of-funds even add value, if their management
randomly chooses funds and especially are worth the fees-on-fees assuming manage-
ment has selection ability.

Discussion Questions
1. Discuss the two typical forms of PE funds-of-funds.
2. Discuss an alternative structure to the general 220 framework for managerial
compensation.
3. Describe the typical fees of a PE fund and fund-of-funds.
4. Explain how to calculate carried interest for a fund using the following data and as-
suming a $100 million fund. Assume a 2 percent management fee, 20 percent car-
ried interest rate, and 100 percent catch-up period.

Year 1 2 3 4 5 6 7 8 9 10 Total
CF 50 30 10 10 60 85 7 20 40 38 150

References
Ang, Andrew, Matthew Rhodes-Kropf, and Rui Zhao. 2005. Do Funds-of-Funds Deserve Their
Fees-on-Fees? Working Paper, Columbia Business School.
Brown, Stephen J., William N. Goetzmann, and Bing Liang. 2004. Fees on Fees in Funds of Funds.
Working Paper, Yale International Center for Finance, University of Massachusetts, Amherst.
European Private Equity and Venture Capital Association. 2013. 2012 Pan-European Private
Equity and Venture Capital Activity. EVCA Yearbook 2012.
Friend, Irwin, and Marshall E. Blume. 1975. The Demand for Risky Assets. American Economic
Review 65:5, 900922.
Gompers, Paul, and Josh Lerner. 1996. The Use of Covenants: An Empirical Analysis of Venture
Partnership Agreements. Journal of Law and Economics 39:2, 463498.
Gompers, Paul, and Josh Lerner. 2001. The Venture Capital Revolution. Journal of Economic Per-
spectives 15:2, 145168.
482 t r e n d s i n p r i vat e e q u i t y

Jesch, Thomas A. 2004. Private-Equity-Beteiligungen: Wirtschaftliche, rechtliche und steuerliche Rah-


menbedingungen aus Investorensicht. Wiesbaden: Betriebswirtschaftlicher Verlag Dr. Th. Gabler,
GWV Fachverlage GmbH.
Jugel, Stefan. 2008. Private Equitydie Rolle nicht brsennotierter Beteiligungen der deutschen
Versicherungswirtschaft. In Stefan Jugel, ed., Private Equity Investments: Praxis des Beteili-
gungsmanagements, 314. Wiesbaden: Betriebswirtschaftlicher Verlag Dr. Th. Gabler, GWV
Fachverlage GmbH.
Kaplan, Steven N., and Antoinette Schoar. 2003. Private Equity Performance: Returns, Persistence
and Capital Flows. Journal of Finance 60:4, 17911823.
Kaplan, Steven N., and Per Strmberg. 2008. Leveraged Buyouts and Private Equity. Working
Paper, National Bureau of Economic Research, Cambridge.
Metrick, Andrew, and Ayako Yasuda. 2010a. The Economics of Private Equity Funds. Review of
Financial Studies 23:6, 23032341.
Metrick, Andrew, and Ayako Yasuda. 2010b. Venture Capital and Other Private Equity: A Survey.
Working Paper, National Bureau of Economic Research, Cambridge.
National Venture Capital Association. 2014. Available at: http://www.nvca.org/index.php?option=
com_content&view=article&id=141&Itemid=589.
Preqin. 2012. The Preqin Quarterly: Private Equity, Q3 2012. Available at https://www.preqin.
com/docs/quarterly/PE/Private_Equity_Quarterly_Q3_2012.pdf.
Preqin. 2013. The 2013 Preqin Private Equity Fund of Funds Review: A Comprehensive Guide to
Multi-manager Private Equity Funds. Available at https://www.preqin.com/docs/samples/
The_2013_Preqin_Private_Equity_Fund_of_Funds_Review_Sample_Pages.pdf?rnd=1.
Prowse, Stephen D. 1998. The Economics of the Private Equity Market. Federal Reserve Bank of
Dallas, Economic Review, Third Quarter, 2134.
Robinson, David T., and Berk A. Sensoy. 2013. Do Private Equity Fund Managers Earn Their Fees?
Compensation, Ownership, and Cash Flow Performance. Working Paper, Charles A. Dice
Center for Research in Financial Economics, The Ohio State University, Fisher College of
Business.
Sovran, Joe. 2006. Private Equity Funds of Funds. In Iggy Bassi and Jeremy Grant, eds., Structuring
European Private Equity, 8796. London: Euromoney Institutional Investor Plc.
Weidig, Tom, and Pierre-Yves Mathonet. 2004. The Risk Profiles of Private Equity. Working Paper,
QuantExperts and European Investment Fund. Available at http://ssrn.com/abstract=495482.
Wei, Andreas, and Matthias Fischer. 2010. Akquisitionsdarlehen und ihre steuerliche Abzugs-
fhigkeit. Research Note, mlfischer.com, Center for Financial Research, Georg-Simon-Ohm
Hochschule Nrnberg.
26
Publicly Traded Private Equity
HENRY LAHR
Lecturer in Finance, The Open University

Introduction
Publicly traded private equity (PTPE) is an asset class comprised of vehicles that offer
investors an opportunity to take part in private equity (PE) investments. These vehicles
follow a PE strategy (e.g., venture, buyout, and mezzanine) and are committed to the PE
investment process. They select portfolio companies, structure transactions, monitor
the portfolio firms operations, and divest to generate capital gains for their sharehold-
ers. Portfolio companies are private firms, as opposed to the vehicle itself, which are
listed on a stock exchange. The literature offers several synonyms for this asset class,
including listed private equity (LPE), publicly traded private equity, or, more rarely, quoted
private equity and liquid private equity. Although some academic studies and finan-
cial products use the term publicly traded private equity, listed private equity is the most
common name for this asset class.
LPE vehicles offer several advantages compared to non-listed PE funds. These ad-
vantages are mainly associated with investors being able to continuously trade shares on
regulated public markets. For an investor, PTPE broadens the feasible investment set
by providing access to PE management companies that are similar to the general part-
ner (GP) in conventional GP/limited partner (LP) structures. PTPE requires no min-
imum investment and can readily be sold in most circumstances. It provides liquidity
where secondary markets in conventional PE offer little or expensive liquidity. Investors
can use PTPE for their cash management by parking cash committed to conventional
PE funds or other temporarily uninvested capital. Performance measurement is much
easier due to the availability of market prices. Finally, most traded vehicles are subject to
shareholder protection laws that reduce monitoring costs for shareholders.
An important disadvantage for newly listed vehicles is the necessity to invest the
proceeds of an initial public offering (IPO) immediately as excess cash may reduce
performance and dilute the portfolios of early investors. Listing fees and costly legal
requirements for public firms can similarly reduce performance. In general, investors
have less control over their exposure to PE investments because listed funds often invest
in instruments other than private portfolio companies. PTPE vehicles are often much
smaller than non-listed, billion-dollar PE funds. As a result, their liquidity often remains

483
484 t r e n d s i n p r i vat e e q u i t y

poor with higher bid-ask spreads and lower trading volume compared to blue-chip
stocks.
The purpose of this chapter is to introduce PTPE, outline its structure, and provide
an overview of its unique pricing characteristics. The rest of the chapter is organized as
follows. The next section briefly describes the current and historical market for PTPE,
followed by an introduction to the main organizational forms of such vehicles. Sections
on net asset value (NAV) discounts and risk-return characteristics highlight risks and
opportunities in this market. The chapter concludes with a summary from an investors
point of view.

Market Trends
LPE as a recognized asset class is a relatively recent phenomenon. First listings of firms
that now might be called PTPE vehicles, such as Eurazeo or Capital Southwest, date
back to the 1960s and 1970s. Many of these firms started as small business investment
companies (SBICs) or business development companies (BDCs) in the United States
or, later, venture capital trusts (VCTs) in the United Kingdom. The sectors trade asso-
ciation LPEQ (http://www.lpeq.com) was founded in 2006. Figure 26.1 reveals two
peaks of listing activity. Many firms went public during the dot-com boom, but the in-
dustry was growing most quickly between 2005 and 2007. In 2000, 53 companies listed
their stock on an exchange, while the most recent peak occurred in 2006 with the same
number of firms going public. Similar to waves of mergers and acquisitions (Martynova
and Renneboog 2008), new listings follow the business cycle. The financial crisis of
20072008, however, had a long-lasting negative effect on the market for new issues
including PE vehicles. After a sharp drop in 2008, 2009 saw the lowest number of new
listings since the early 1990s. The industrys growth quickly turned negative with 44
delistings in 2009 alone and several more in recent years. About one-third of all iden-
tifiable PTPE vehicles delisted over the entire observation period. The recent market
consolidation left its mark on the total number of existing vehicles, reaching a peak in
2007 and slightly decreasing since then.
LPE vehicles are mostly located in Europe. Out of 606 vehicles, 35 percent are head-
quartered in the United Kingdom, while 16 percent are located in the United States
and 6 percent in Germany. The next largest markets are Canada, Australia, Israel, and
Sweden. Their size distribution is heavily positively skewed. At the end of 2013, aver-
age market capitalization was $1,177 million, but the median size was just $35 million.
During the dot-com boom, the largest vehicles were incubators such as Softbank and
CMGI. Despite a few large firms, many listed vehicles tend to be small and illiquid,
which can be a problem if investors wanted to use them build a portfolio that closely
tracks the PE market. Figure 26.1 shows the total market capitalization was $337 billion
at the end of 2013, slightly more than the previous peak in early 2000. PTPE market
values are volatile, showing large swings over equity market movements, which is re-
flected by the large market betas of listed PE described.
Several indexes track the PTPE market and help investors evaluate the sectors
risk and return characteristics. The first indexes appeared in 2004, when LPX GmbH
began developing a family of indexes consisting of global indexes varying in scope
Publ icl y Trade d P riv at e E qu it y 485

Panel A. Number of new listings


60

50

40

30

20

10

0
1980 1985 1990 1995 2000 2005 2010

Panel B. Market capitalization

500 8
Market cap. (left scale) Market cap. index (righ scale)
400 6

Log index, base = 1980


300 4
USD bn

200 2

100 0

0 2
1980 1985 1990 1995 2000 2005 2010

Figure 26.1 Listings and Market Capitalization of Publicly Traded


Private Equity The column for 1980 includes the listings of PTPE vehicles in Panel A
before and during 1980. New listings for recent years probably underestimate the actual
number of new PTPE vehicles because such vehicles may not have been discovered yet.
Panel B shows the market capitalization in USD billions and as the logarithm of an index
with base year 1980. Source: Authors calculation based on data from Thomson Datastream.

(LPX composite, LPX50, and LPX Major Market), regional reach (LPX Europe, LPX
UK, and LPX America) as well as in investment style (LPX Buyout, LPX Venture,
LPX Direct, LPX Indirect, and LPX Mezzanine). Red Rocks publishes an index sim-
ilar to the one constructed by LPX GmbH. Their Global Listed Private Equity Index
(GLPEI) comprises the 40 to 60 most liquid vehicles worldwide. The S&Ps Listed
Private Equity Index is constructed from a global portfolio of 30 large, liquid LPE
companies. The Socit Gnrale Private Equity Index (PRIVEX) includes the 25
most representative stocks from the buyout, venture capital, and growth capital sub-
sectors. The STOXX Europe Private Equity 20 index is constructed to reflect the per-
formance of the 20 largest PTPE firms in Europe. Many intermediaries, including
ALPS Fund Services, BlackRock Advisors, Deutsche Bank, Invesco, Merrill Lynch,
RBS, or UBS offer exchange-traded funds (ETFs), certificates, and mutual funds
tracking these indexes.
486 t r e n d s i n p r i vat e e q u i t y

Organizational Forms of PTPE


PTPE vehicles show large variability in their different organizational characteristics
and underlying economic structures (Lahr and Herschke 2009). As opposed to con-
ventional PE in which most entities are structured as limited partnerships, the PTPE
universe is diverse about the vehicles legal and economic structures. For investors, a
distinction of organizational forms is essential to understand the underlying value gen-
eration processes, which in turn affect risk and expected return.
Many PTPE vehicles resemble the conventional PE structure, but others combine in one
firm functions that different legal entities in the GP/LP structure perform. Lahr and Her-
schke (2009) distinguish between funds that invest in non-listed companies directly and
management companies that provide investment management for funds. Figure 26.2 shows
organizational relationships between these basic entities. Indirect exposure but higher de-
grees of diversification can be achieved through a third organizational form, namely listed
funds-of-funds, which invest in conventional PE funds. A fourth category is investment
companies that invest directly in private companies and manage their portfolio internally.
Vehicles can be classified along two dimensions: (1) internally or externally managed
companies and (2) structures consisting of single or multiple funds. Investment man-
dates often do not prescribe exclusive investment in private companies but also allow
funds to be invested in other assets such as AIM (formerly the Alternative Investment
Market) quoted companies or third-party funds. The organizational forms described in
the upcoming sections have large overlap that forces close inspection of portfolios and
legal structures when assigning vehicles to the different categories.

FUNDS
LPE funds are externally managed vehicles that invest directly in private companies (e.g.,
Altamir-Amboise, Electra Private Equity, KKR Private Equity Investors, and HgCapital

Management Company

Fund-of -Funds
Investors
Fund
Portfolio
Portfolio
Portfolio Company
Companies
Company
(Private)
Investment Company

Figure 26.2 Organizational Forms of Listed Private Equity This figure


shows ownership and control relationships in PTPE. Investors can own shares or units
in four basic types of entities that directly or indirectly own portfolio companies: (1)
amanagement company, (2) fund-of-funds, (3), fund, and (4) investment company.
Source: Adapted from Lahr (2010).
Publ icl y Trade d P riv at e E qu it y 487

Investment
General Partner Limited
Common Professionals
Shares
Management Fee
GP Interest
Carried Interest
Shares,
Units Issuer Limited Partnership Common Shares
Investors
(Fund)
GP Interest/
Management
LP Interest

U.S. Fund L.P. European Fund S..r.l.

European Portfolio
U.S. Portfolio Companies
Companies

Figure 26.3 Legal Structure of a Publicly Traded Private Equity


Fund This figure shows the structure of a typical listed fund. Similar to conventional
PE funds, the fund is managed by a GP, which is owned by investment professionals, and
has the usual management agreement and fee structure in place. Source: Adapted from Lahr
(2010), based on AP Alternative Assets (2006) and KKR Private Equity Investors (2006).

Trust). Economically, they resemble conventional PE funds as Figure 26.3 shows. These
vehicles are balance sheet investing (i.e., they collect funds from shareholders with the
purpose of earning capital gains from investments in private companies). Similar to
conventional PE limited partnerships, a management company provides investment
management. However, a management company is not owned by the fund but often
affiliated with a PE group that also manages more conventional PE funds. Besides a flat
management fee, managers are paid performance fees based most often on NAV returns
and sometimes on stock returns. Funds typically invest directly or alongside their con-
ventional GPs other funds.

FUNDS-OF-FUNDS
Another publicly listed, externally managed investment vehicle is a fund-of-funds. It in-
vests in conventional PE funds as limited partners such as Castle Private Equity, Pantheon
International Participations, and Scandinavian Private Equity. A PTPE fund-of-funds di-
versifies its portfolio at the fund level with financing stages, GPs (PE groups), underly-
ing fund vintage years (i.e., the year in which capital is first invested), and geographical
region. This diversification has a big impact on performance and risk characteristics, such
as adjusting the funds systematic risk closer to the average. As in conventional PE, fund
management is contracted out to a management company as Figure 26.4 shows. The GP
typically receives management and performance fees. These fees may be seen as form-
ing a second fee layer between the investor and underlying portfolio investments in pri-
vate companies. Investors have to balance this added cost against the potential benefit
of greater diversification and specialist skills in selecting investments, due diligence ex-
pertise, or syndicated deals with other PE fund managers in long-standing relationships.
488 t r e n d s i n p r i vat e e q u i t y

Investment
Investors Professionals

Common Shares Common Shares

Issuer PLC
(Fund-of-Funds) Investment Manager Ltd.

Common Shares Management Fee, Performance Fee Common Shares

Investing Fund Ltd. Investing Fund Ltd.


(Cayman Islands) (Home Country)

Limited Partnerships Limited Partnerships

Portfolio Companies Portfolio Companies

Figure 26.4 Legal Structure of a Private Traded Public Equity Fund-of-


Funds This figure shows a typical PTPE fund-of-funds that invests in two (or more)
conventional PE funds in its home country and abroad. Source: Adapted from Lahr (2010),
based on shaPE Capital (2001) and Castle Private Equity AG (2000).

M A N A G E M E N T C O M PA N I E S
A listed management company is an internally managed vehicle that corresponds
to the GP in a conventional PE group. Examples include The Blackstone Group,
3i, Candover Investments, DEA Capital, GIMV, and Intermediate Capital Group.
Listed managed companies provide an opportunity for investors to participate in
cash flows from management activities, which is impossible in conventional PE
structures.
Figure 26.5 shows that management companies often hold GP interests in the PE
funds they manage. Cash flow is generated mostly through management agreements
with PE funds and performance fees (carried interest). A challenge for investors is to
distinguish PE equity managers from general asset management firms, which some-
times manage alternative asset portfolios in addition to a much larger portfolio com-
prised of other asset classes. Management companies often have small minority stakes
in the funds they manage.

I N V E S T M E N T C O M PA N I E S
An investment company is an internally managed vehicle similar to a fund, but does
not rely on external management and employs its own investment professionals instead.
Figure 26.6 shows this much simpler structure compared to GP/LP structures. Invest-
ment companies combine functions in a single entity that are separated in non-listed
Publ icl y Trade d P riv at e E qu it y 489

100 % Investment
Investors Special Voting Unit L.L.C. Professionals
Limited Voting No Economic
100 %
Common Rights Rights
Units GP Interest General Partner L.L.C.
Issuer Limited Partnership

GP Interest GP Interest GP Interest


Limited Partnership I Limited Partnership II Limited Partnership III

Figure 26.5 Legal Structure of a Publicly Traded Management


Company This figure shows the organizational structure of a typical management
company in which investors own units with limited voting rights since a GP, which is
owned by investment professionals, makes investment decisions. A standard fee structure
used in conventional PE funds applies between the issuing partnership and the funds it
manages. Source: Adapted from Lahr (2010), based on The Blackstone Group (2007).

PE structures. Examples include Ratos AB, Wendel, and Compagnie Du Bois Sauvage.
Investment companies come in a wide range of local flavors. Many jurisdictions regulate
special legal statuses for investment companies such as BDCs in the United States and
VCTs in the United Kingdom. Regulations often stipulate tax benefits tied to lock-up
periods or investment restrictions.
An investment company typically holds a portfolio of direct investments in private
companies, but some stacked holding structures may exist within an investment com-
pany. Except for its commitment to the PE business model, an investment company
often lacks features that distinguish it from an ordinary holding company. It usually
prepares consolidated financial statements but also reports a detailed portfolio com-
position, which is similar to reporting fair values of portfolio investments by a listed
fund. Portfolios held by an investment company are comparable to those of listed funds.
However, contrary to a listed fund, which is restricted by its statutes or partnership
agreement, an investment company can take on more debt on its balance sheet, which
can influence risk and return characteristics.

Common
Shares
Investors Investment Company PLC

Portfolio Company I Portfolio Company II Portfolio Company III

Figure 26.6 Legal Structure of a Publicly Traded Investment


Company This figure shows the structure of an investment company that owns several
subsidiaries in a holding structure. Contrary to conglomerate firms, investors hold portfolio
companies for financial gain and managed according to the PE business model. Source:
Adapted from Lahr (2010), based on Ratos AB (2012).
490 t r e n d s i n p r i vat e e q u i t y

S P E C I A L L E G A L F O R M S F O R P U B L I C I N V E S T M E N T S I N P R I VAT E
EQUITY
Besides the usual legal forms of listed firms, several legal structures facilitate access to
finance by small entrepreneurial firms. These structures, which have emerged in the
United Kingdom and the United States, usually offer tax benefits or access to govern-
ment funds. In the terminology of the previous section, they can be seen as funds or
investment companies. Their special legal status can have a large impact on their risk-
return profile.

Small Business Investment Companies


Many of the first PTPE vehicles are SBICs in the United States that were created follow-
ing the passage of the Small Business Investment Act of 1958. The Act allowed the Small
Business Administration (SBA) to license private SBICs to help provide funding and
management expertise to small entrepreneurial businesses in the United States. SBA
funds can be used to leverage privately raised investment funds. Some of the first stud-
ies of PTPE examined SBICs, such as Martin and Pettys (1983) analysis of 11 traded
firms, of which 8 are SBICs.

Business Development Companies


BDCs are publicly traded closed-end companies that are regulated under the U.S. Invest-
ment Company Act of 1940 and seek to invest in small and mid-sized private companies.
They are required by law to provide support and substantial managerial assistance to their
portfolio companies. To qualify as a BDC, a company must elect to be registered in com-
pliance with the Investment Company Act and must hold at least 70 percent of its total
assets in shares of private companies or securities for which there is no ready market, cash
equivalents, U.S. government securities, or short-term high-quality debt securities. BDCs
are usually taxed as a regulated investment company (RIC), which requires them to dis-
tribute at least 90 percent of their taxable income to shareholders each year.

Investment Trusts
Investment trusts are special U.K. closed-end investment companies under the Income
and Corporations Taxes Act 1988. They are exempt from paying income or capital gains
tax within the fund and can deduct management charges. The companys income must
be derived wholly or mainly from shares or securities, and no holding in a company
other than an investment trust represents more than 15 percent of the trusts assets. The
companys memorandum or articles of association must prohibit distributing capital
gains as dividends, and the trust must not retain more than 15 percent of the income it
derives from shares and securities each year. The ability to leverage their equity distin-
guishes them from other collective investment schemes such as unit trusts.

Venture Capital Trusts


VCTs are closed-end investment companies in the United Kingdom first introduced by
the Finance Act in 1995. They typically invest in small, potentially high-growth, private
companies and new shares of companies that are traded on the AIM. VCTs are exempt
from capital gains tax and income tax on dividends and include income tax relief on the
Publ icl y Trade d P riv at e E qu it y 491

initial investment when subscribing to new VCT share issues. To qualify for income tax
relief on subscription, investors must hold VCTs for a minimum of five years. These rules
governing the tax benefits of VCTs have changed several times since their initial adoption.
At least 70 percent of their investments must be in shares of small private U.K. companies
with pre-money valuations of less than 7 million (15 million before April 2006). Be-
cause of tax benefits and due to their special statutory governance mechanisms, VCTs are
sometimes believed to underperform the market on a share price basis (Cumming 2003).

Split Capital Trusts


Split capital trusts were introduced in the United Kingdom in 1965 and originally had a
limited life with a fixed wind-up date and two classes of shares: income shares and capital
shares. The different share classes are designed to meet different investors needs by offer-
ing income generated from investments (income shares) or the capital value of the com-
pany at wind-up (capital shares). Over time, several more share types have been developed
such as zero dividend preference shares. At least one share class within a split capital trust
is likely to have a limited life (usually between 5 and 10 years) with a fixed wind-up date.

Special Purpose Acquisition Companies


Special purpose acquisition companies (SPACs) are shell or blank-check companies in
the United States that raise capital to acquire a yet unidentified business in the future
(Hale 2007). Davidoff (2008) provides a legal description of SPACs and Berger (2008)
offers an economic discussion from a practitioners viewpoint. Since 2007, a few SPACs
have launched in Europe. They provide investors with potential access to acquisitions of
private companies. SPAC shares are issued as units consisting of common stock and one
or two separate warrants that typically can be exercised only if the SPAC completes an
acquisition. SPAC charters usually require an acquisition within 18 to 24 months after
the effective date of the offering. If a suitable target cannot be found within this specified
period, the SPAC is liquidated and its assets are returned to shareholders. SPACs resem-
ble a risk-free asset in the early stages of their life cycle, yet many become single-trans-
action buyout funds if successful. After an acquisition has been consummated, SPACs
are similar to normal holding companies. Therefore, PE indexes usually exclude SPACs.

NAV Discounts
Pricing PE funds and estimating their risk and return characteristics can be challenging
because of the lack of observable market data. PTPE funds offer an opportunity to gain
insights into these pricing problems by viewing their market prices and other publicly
reported fundamental data. One practical application of observable market prices is to
estimate the relationship between a funds market price and its underlying NAV. Lahr
and Kaserer (2010) use a sample of publicly traded funds to examine the determinants
of this pricing relationship in PE funds. This section summarizes their findings and de-
scribes economic causes for a divergence of prices and NAVs. The difference between a
funds NAV and its market price is often called the NAV discount because of funds usually
trading at a fraction of their NAV. For an econometric analysis, dividing market value by
the funds NAV and referring to its logarithm as the funds premium is more convenient.
492 t r e n d s i n p r i vat e e q u i t y

If premiums can be successfully explained, this has implications for pricing conven-
tional PE funds and secondary fund transactions and may also shed light on the so-called
closed-end fund puzzle because a similar phenomenon can be observed in closed-end
equity funds. The volume of transactions involving the transfer of fund shares between
investors (secondary transactions) surged since 2004, and selling pressure intensified
during the financial crisis of 20072008. As a result, specialized secondary market funds
in the multi-billion dollar range were raised in 2008 and 2009. Secondary transaction
data collected by Cogent Partners (2010) show a strong similarity between premiums
in conventional PE funds and their listed counterparts. From a peak at 108 percent for
non-listed funds and 92 percent for listed funds in 2007, NAV premiums dropped to
39 percent and 47 percent, respectively, in the first half of 2009 before the gap between
market values and NAV narrowed to 72 percent and 67 percent in late 2009.
PTPE funds bridge the gap between closed-end funds, non-listed PE funds, and
listed holding companies. Therefore, identifying which theories apply to price these
funds is not immediately clear. Since PTPE funds emerged as an asset class in the 1990s,
these vehicles have constantly traded at a discount to their respective NAV on average.
If listed PE funds are similar to holding companies, they should trade at a premium on
average. Historically, market-to-book ratios in stocks have been well above one (Kothari
and Shanken 1997; Loughran 1997). One can also view listed PE funds as traded port-
folios of private companies. If this perspective is correct, these funds would be more
similar to closed-end funds, which also represent a portfolio, albeit of securities instead
of non-listed shares. The cross-sectional average NAV premiums in closed-end funds
are almost always negative in a range between 5 and 10 percent (Dimson and Minio-
Kozerski 1999; Cherkes, Sagi, and Stanton 2009). Figure 26.7 shows that during the

0.5

0.5

1.5

2.5

1992 1994 1996 1998 2000 2002 2004 2006 2008 2010

Figure 26.7 Net Asset Value Premium in Calendar Time This figure shows NAV
premiums for a sample of 75 listed funds and 22 listed funds-of-funds. The dots represent
2,010 premium observations (log (Price/NAV)) and the solid line is an equally weighted
average of NAV premiums. Source: Adapted from Lahr and Kaserer (2010).
Publ icl y Trade d P riv at e E qu it y 493

same time period, traded PE funds had a much larger negative premium, trading at an
average discount to NAV of about 26 percent.
Researchers propose several theories to explain the difference between NAV and the
market price of a listed fund. Two areas of research are of particular relevance to PTPE
funds. The first area is the current literature focusing on closed-end funds investing in
securities. A second source, PE-specific explanations, takes account of PE funds invest-
ing in non-listed companies.

THE CLOSED-END FUND PUZZLE


Closed-end equity funds trading at a premium over NAV present a puzzle, but such
premiums should be eliminated by arbitrage in perfect markets. Closed-end invest-
ment funds are usually issued with a premium of up to 10 percent. Within a few months
after the IPO, they trade at a discount. If a fund is converted into an open-end fund,
merged with an open-end fund, liquidated, or if the funds portfolio is sold as a whole,
the fund price rises and the discount disappears (Brauer 1984; Kadapakkam, Misra,
and Yildirim 2005).
Several theories try to explain the closed-end fund puzzle and may also apply to
listed PE. Several scholars propose theories in which management fees correlate with
NAV premiums (Boudreaux 1973; Kumar and Noronha 1992). Empirical findings
of such correlations are controversial (Malkiel 1977; Lee, Shleifer, and Thaler 1991;
Dimson and Minio-Kozerski 1999). Additionally, the small part of total variance ex-
plained in regression analysis suggests that other explanatory factors may be present.
Because management fees in PE funds are usually determined as a fixed percentage of
net assets, all arguments put forward to explain premiums in closed-end mutual fund
should therefore carry over to PE funds.
The theory of managerial ability proposes that many closed-end funds have higher
costs than the expertise of investment managers could justify. Shares of those funds
should therefore trade at discounts while offering comparably low returns. However,
Malkiel (1977) and Thompson (1978) find no evidence for this hypothesis in their
analysis. According to Chay and Trzcinka (1999), a significant correlation exists be-
tween high discounts and low future NAV returns. Lee, Shleifer, and Thaler (1990) and
Pontiff (1995) do not find this correlation in their analyses but show the opposite: large
NAV discounts lead to better future NAV returns. Berk and Stanton (2007) contend
that discounts change over time as investors alter beliefs about the managers ability.
If managers cannot be fired, poor managerial performance leads to discounts. Positive
premiums are short-lived because managers learn about their above-average perfor-
mance and negotiate a pay increase.
Because these theories cannot sufficiently explain NAV premiums, the rationality of
the market is sometimes questioned, leading to several arguments that suggest inves-
tor sentiment as a mechanism explaining fund premiums. Zweig (1973) contends that
premiums in closed-end funds mirror the expectations of private investors. De Long,
Shleifer, Summers, and Waldmann (1990) explore the impact of these investors and
the possibility of using arbitrage to reduce the premiums. They suggest the existence
of two different groups of investors: rational and irrational investors. Rational inves-
tors have unbiased expectations of future returns, but irrational investors predictions
494 t r e n d s i n p r i vat e e q u i t y

are systematically biased in either direction. Because of their unpredictable behavior,


irrational investors prevent rational investors from eliminating premiums by arbitrage. A
more indirect reason for discounts is the risk induced by noise traders (Lee et al. 1991).
Noise traders induce discounts because of a non-predictable risk of stochastically acting
irrational investors. Investors wanting to sell their investments in finite time have to be
compensated for this noise trader risk resulting in NAV discounts. Empirical studies,
however, find conflicting results (Weiss 1989; Ammer 1990; Lee et al. 1991).
Lee et al. (1990) argue the liquidity of the securities the fund holds cannot be re-
sponsible for the NAV discount in investment funds because only a few funds have il-
liquid securities in their portfolio. Yet, Cherkes et al. (2009) find that closed-end funds
holding illiquid securities trade at higher premiums because of the additional liquidity
these funds offer their investors compared to direct investments in these portfolio com-
panies. In listed PE, a funds stock is listed on an exchange, but shares in the portfolio
companies are highly illiquid. At the fund level, the ability to float portfolio firms on
a public market is highly critical to realize capital gains from investments (Lahr and
Mina 2014). Since the illiquidity of an investment is not considered when calculating
the NAV, investors in such a fund might pay an additional liquidity premium. Note that
noise trader risk would imply that prices should be closer to NAV if the liquidity of the
funds underlying portfolio is high and arbitrage strategies are easier to implement.
Private benefits enjoyed by fund managers could be responsible for discounts (i.e.,
negative premiums) because managers owning shares of the fund could resist open-
ending proposals due to poor returns. Barclay, Holderness, and Pontiff (1993) find the
greater the managerial stock ownership in closed-end funds, the larger are discounts to
NAV. The situation in listed PE funds is more complicated. While VCTs allow the funds
dissolution by shareholder resolution, shareholder (or unitholder) rights differ in funds
with a partnership structure. Limited partners in these funds typically have no right to
terminate and dissolve the fund. Therefore, a small or no effect of ownership structures
would be expected.
In his tax timing hypothesis, Constantinides (1984) says the capital gains tax can
be reduced by skillfully timing purchases and sales of shares. According to this theory,
investors forgo their chance to minimize taxes through managing their portfolio by in-
vesting in externally managed funds (Brickley, Manaster, and Schallheim 1991). Tax
timing, however, cannot explain why funds occasionally trade at a positive premium.
Contrary to investment funds, PTPE funds invest in illiquid assets. The replication ar-
gument rarely holds because precise control of tax payments is difficult to implement
in practice. Because of this illiquidity, the tax timing hypothesis turns into a liquidity
hypothesis.

P R I VAT E E Q U I T Y S P E C I F I C E X P L A N AT I O N S
Besides explanations focusing on mutual funds and the closed-end fund puzzle, cer-
tain pricing characteristics are special to PE funds. A well-established empirical phe-
nomenon is the J-shaped relationship between a funds age and its lifetime NAV return
(Kaserer and Diller 2004; Kaplan and Schoar 2005; Phalippou and Gottschalg 2009).
Average reported NAVs of most funds drop during the first few years and then grow
steadily until the end of the funds lifetime. Management fees, initial investment costs,
Publ icl y Trade d P riv at e E qu it y 495

and aggressive depreciation of unsuccessful investments can explain this. These effects
also apply to listed PE except for management fees, for which the size of the J-curve is
supposedly lower than in conventional PE funds. More than half of the European insti-
tutional investors surveyed by LPEQ believe that listed PE offers smaller management
fees compared to limited partnership PE. This represents an attractive way to invest in
PE after the J-curve, avoiding low returns on investment in initial periods (Cumming,
Fleming, and Johan 2011).
PE transactions and mainly those in the buyout market typically involve large amounts
of debt and, therefore, depend on credit markets. Kaplan and Strmberg (2009) con-
tend that PE funds may take advantage of systematic mispricings in the debt and equity
markets by overleveraging accordingly. Baker and Wurgler (2000) put forward a similar
argument for public companies. Axelson, Strmberg, and Weisbach (2009) propose a
different hypothesis based on the observation that PE firms aim for large transactions
relative to their fund sizes. PE firms might be constrained in equity they can invest in
a given deal. Therefore, they must use leverage to fund their investments. One possi-
ble conclusion is that PTPE funds might undertake more transactions if interest levels
are unusually low. Fund investors could anticipate excess value creation by funds during
these periods, which in turn increase NAV premiums.
Investors usually get information about portfolio values only through financial re-
ports issued by the fund. A lack of disclosure by the fund and being uninformed can
therefore form a risk for investors. For example, La Porta, Lopez-de Silanes, and Shleifer
(2006) find that both extensive disclosure requirements and standards of liability are
associated with larger stock markets. Lang, Lins, and Maffett (2012) show that greater
transparency is related to higher liquidity, which in turn is associated with higher To-
bins Q. Therefore, a countrys regulatory environment and the availability of public in-
formation might affect PTPE valuations. The quality of NAVs strongly depends on the
choice of valuation model parameters and the method used for their calculation. There-
fore, a fund can act with relatively high flexibility in pricing those portfolio companies
for which no market price is available (Anson 2002). Many management companies are
reluctant to change valuations without value-determining events such as a change of
ownership, which can lead to NAVs containing less and less current information (stale
pricing). Anson finds a lag between NAV returns and stock index returns. Rather than
biasing the average premium, stale pricing induces autocorrelation in premiums, which
may affect estimating NAV premiums across time.

M A R K E T T I M I N G A N D I N V E S TO R S E N T I M E N T
PTPE funds start trading at an average premium of almost zero. Since no fund went
public on a reporting date, the observation closest to day zero is recorded for one fund
two days after its IPO. Figure 26.8 shows a premium of 0.4 percent for this day. If
extrapolated linearly from the first observations, the IPO premium is indistinguisha-
ble from zero. This premium is considerably less than those reported for U.K. and U.S.
closed-end funds. The average investment trust issue is quoted at an effective premium
of 5.7 percent above its NAV at the end of the first day of trading (Levis and Thomas
1995). U.S. stocks still trade at a 4.8 percent premium five weeks after the IPO (Weiss
1989). As Weiss reports, an initial return of 6.5 percent over the first 30 days suggests
496 t r e n d s i n p r i vat e e q u i t y

0.5

0.5

1.5
0 1 2 3 4 5 6 7 8 9 10

Figure 26.8 Premium in Event Time, Years from Initial Public OfferThis
figure presents premium data in event time relative to the funds IPO. The solid line
represents the average premium (log (Price/NAV)) estimated by locally weighted
regression with bandwidth 0.15. The dotted lines are confidence bands that are two
standard deviations from the estimated mean. Source: Adapted from Lahr and Kaserer (2010).

that U.S. stocks sell at an even higher premium. According to Lee et al. (1991), positive
premiums on the first trading day are generally attributed to investor sentiment, which
issuing firms can use to their advantage. If first day premiums occur in hot issue markets
due to positive investor sentiment and issuers market timing, less irrational behavior is
observed in listed PE funds. Rather than the level of premiums, the large drop in premi-
ums after a funds IPO may show investor sentiment.
Premiums in PTPE funds approach the long-run average of about 26 percent less
quickly than those in other closed-end funds. Figure 26.9 shows the drop in premiums
is linear and reaches its bottom after two to two and a half years. This period is similar to
investment periods of conventional PE funds. Weiss (1989) finds that within 24 weeks
of trading, closed-end equity funds in the United States trade at a significant average dis-
count of 10 percent. Levis and Thomas (1995) find that after 200 trading days, equity
funds in the United Kingdom fall in value by 5 percent.
The qualitative behavior of premiums is thus similar to closed-end funds but dis-
plays a time pattern as in non-listed PE funds where returns measured by the internal
rates of return (IRR) usually turn around after two to three years and break even
after five to six years. However, the pattern is reversed compared to what would be
expected if the J-curve phenomenon was driving premiums. The J-curve effect refers
to the shape of cumulative book returns on a portfolio company. They are at first neg-
ative due to costs associated with the transaction or restructuring and turn positive
after a few years. If NAV returns were low over the first quarters but shares earned
some risk-adjusted return, premiums should rise first and then remain at an equilib-
rium point.
Publ icl y Trade d P riv at e E qu it y 497

0.2
Buyout Venture Fund-of-Funds
0.1

0.1

0.2

0.3

0.4
0 1 2 3 4 5 6 7 8 9 10

Figure 26.9 Premium in Event Time, Years from Initial Public Offering
by Fund Type This figure shows estimated mean premiums in event time by fund type.
The bold lines show estimates of locally weighted regressions that are performed with
bandwidth 0.15. Thin lines are confidence bands that are two standard deviations from the
estimated mean. NAV premiums are calculated as (log (Price/NAV)). Source: Adapted from
Lahr and Kaserer (2010).

Weidig, Kemmerer, Lutoborski, and Wahrenburg (2007) find that PEF managers
value their investments conservatively. Such a valuation according to the PE industrys
guidelines would lead to premiums greater than zero. To the contrary, evidence shows
that either the management reports inflate NAVs or other factors drive down the funds
share price. The time dynamics of NAV premiums strongly suggest the latter explana-
tion in favor of investor sentiment, market timing, and liquidity as the main factors in-
fluencing premiums.
An explanation for discounts in line with Berk and Stantons (2007) management
ability hypothesis is the market needs some time to learn about the managements qual-
ity. This argument works for PE funds because much more time is needed to invest the
IPO proceeds in PE funds than in closed-end mutual funds. If, for example, manage-
ment ability can be assessed based on the acquisitions during this initial investment
period rather than the portfolios subsequent performance, the largest changes in premi-
ums should be observed in this early post-IPO period. Consistent with this explanation,
Figure 26.9 shows that premiums in venture capital funds take about four years to settle,
whereas premiums in buyout funds and funds-of-funds reach their long-term average
two years after their IPO. Initial changes in fund premiums could therefore be an ex-
pression of managerial ability.
In a multivariate setting, Lahr and Kaserer (2010) identify many similarities between
premiums in PTPE funds and closed-end funds but also several differences. Premiums
can be explained mainly by investor sentiment and to some extent by liquidity when
measured by the funds bid-ask spread, but not by managerial ability or fees. Sensitivity to
small-cap indexes and proxies for hot marketssuch as IPO volume, fund commitments,
498 t r e n d s i n p r i vat e e q u i t y

and consumer sentimentlends support to the investor sentiment hypothesis. The drop
in NAV premiums during the first two years cannot be explained directly by economic
causes in regressions of premiums levels, but has only small explanatory power.
According to Berk and Stanton (2007), a possible reason for the decreasing pre-
miums is a market mechanism when participants learn about management ability. A
funds cash holdings can be used to proxy for its investment degree, which should be
related to the information available about successful or unsuccessful portfolio acqui-
sitions. Because the funds investment degree has no significant effect on premiums,
this casts doubt on managerial ability being the only explanation for declining premi-
ums. Besides, an age effect cannot be found in regressions of differenced premiums and
stock returns. Interestingly, initial changes in NAVs are negative, which would imply
increasing premiums after an IPO if share prices were stable. To the contrary, share
prices drop even more than NAVs, causing NAV premiums to fall after the funds IPO.
The finding that fund IPOs cluster in times following positive market sentiment sup-
ports this result.
Both individual and aggregate liquidity variables explain fund premiums. A positive
relationship exists between the funds bid-ask spread and premiums as well as a depend-
ency on Pstor and Stambaughs (2003) measure of aggregate liquidity. Surprisingly,
infrequently traded funds have high premiums. This effect disappears if robustness tests
against data errors are applied to trading volume. Premiums are also higher in times of
low market volatility, which supports the sentiment theory but not the liquidity hypoth-
esis by Cherkes et al. (2009).

Market Risk
Some researchers believe that PE investments sometimes offer higher risk-adjusted re-
turns than other asset classes due to a low correlation with the market portfolio (Bilo,
Christophers, Degosciu, and Zimmermann 2005). Even if positive excess returns for PE
do not occur, investors may gain diversification benefits from a low exposure to market
risk. For example, Bance (2004) estimates historical correlation coefficients to equity
markets in the range 0.5 to 0.6 and 0.1 to bond markets. However, academic research-
ers widely discuss and question these measures (Phalippou 2007; Diller and Kaserer
2009; Phalippou and Gottschalg 2009). While the extent of co-movements with es-
tablished asset classes is difficult to measure in conventional non-listed PE, the PTPE
segment naturally lends itself to studying risk and return patterns. Listed PE vehicles
provide the opportunity to estimate model parameters in a straightforward way using
readily available market prices.
Since business models and organizational forms of PTPE vehicles bear a strong sim-
ilarity to conventional PE funds and management companies, results obtained from
examining listed vehicles may allow inferences about non-listed PE. Moreover, non-
listed PE returns are highly correlated with those of listed PE. Pooled periodic IRRs
for the period 19942009 show correlations with listed PE, measured by LPX50 index
returns, of 0.76 using quarterly returns and 0.87 using yearly returns. Their correlation
with MSCI World index returns is 0.72 quarterly and 0.77 when using yearly returns
(Lahr 2010).
Publ icl y Trade d P riv at e E qu it y 499

Market risk not only depends on a PTPE vehicles underlying assets but also varies
across organizational forms. Lahr and Herschke (2009) estimate market risk of PTPE
indexes constructed from 274 liquid PTPE vehicles for entities for funds, management
companies, funds-of-funds, and investment companies. The authors find substantially
lower exposure to systematic risk in externally managed vehicles (i.e., funds and funds-
of-funds) than in internally managed ones. Beta estimates for value-weighted indexes
of investment companies and firms are 2.0 and 1.5, respectively, while funds and funds-
of-funds have betas of 1.3 and 0.8. Equally weighted, spread-adjusted indexes show a
similar ranking of market risk. Market risk for the whole PTPE market is 1.8 for a value-
weighted index and 1.2 for an equally weighted one. Excess returns in CAPM regres-
sions augmented by currency factors are generally negligible.
Management activities in management companies and investment companies are
the most likely cause of different risk characteristics. Both types of vehicles get cash
flows from fund management operations in the form of management fees or carried
interest. In investment companies, these cash flows remain within the same firm. The
interesting part is the carried interest. Management companies as well as investment
companies may be entitled to a share in capital gains larger than their share in the fund.
Additionally, clawback provisions may require the GP to return some of the carried in-
terest in situations where such distributions exceed a fixed percentage over the life of
the fund. These provisions protect investors by allowing them to claw back any excess
carry paid on successful investments if later investments by the fund turn out less suc-
cessful. Conversely, funds gain proportionately less from positive market trends and
suffer less because of clawback provisions. This asymmetry in cash flows between inter-
nally and externally managed vehicles creates differences in market risk.
In addition to variation depending on organization forms, market risk changes over
time. Kaserer, Lahr, Liebhart, and Mettler (2010) show that individual betas are highly
unstable, while the market risk ranking individual vehicles changes slightly less. They
speculate that factors unique to this sector affect market risk of PE including acquisi-
tions and divestments that constantly rebalance portfolios, scarcity of information
about portfolio companies, and rapid changes within portfolio companies. Unstable
market risk seems to be a fundamental characteristic of PE assets, which must be incor-
porated in the valuation process casting doubt on diversification benefits of PE in times
of crisis. An important aspect of conventional PE funds is that investors usually hold
fund shares to maturity, which can be up to 10 years. Unpredictable changes in market
risk pose a challenge for portfolio allocation because investors would be buying assets
that behave differently from what they were supposed to when first included in the in-
vestors portfolio.

Summary and Conclusions


PTPE has made a strong comeback after the financial crisis of 20072008, attracting re-
newed investor attention after losing more than half of its market capitalization in 2008.
NAV discounts of listed funds are narrowing while the sector is outperforming major
equity market indexes. These developments reflect economic features of traded PE ve-
hicles that were the subject of research over the past few years. Public firms that invest in
500 t r e n d s i n p r i vat e e q u i t y

private firms to improve their operating businesses and sell them at a profit often show
much higher systematic risk than the market portfolio, which improves returns in the
post-crisis era.
Investors in listed PE can benefit from the possibility of buying and selling shares in
listed vehicles at lower costs than equivalent transactions in conventional non-listed PE
funds. Low transaction costs can be especially important for small investors who want
to gain exposure to PE assets. Market price availability enables straightforward risk-
return analyses and helps to evaluate portfolio strategies in non-listed PE by providing
a public return benchmark. In sum, PTPE offers several advantages for investors who
require liquidity or lack the resources to invest in screening and monitoring non-listed
PE funds or direct investments in private companies.
Organizational structures of traded PE vehicles can be as intricate and complex as
their non-listed counterparts. Investors need to be aware that indexes or exchange-
traded funds covering this market often combine highly diverse firms in their portfolios.
While externally managed vehicles, such as listed funds and funds-of-funds, closely re-
semble conventional PE fund structures; listed management companies and investment
offer access to incomes from asset management. In this sense, PTPE firms broaden the
investible universe and may offer diversification benefits. From a strategic portfolio al-
location perspective, these organizational forms present different systematic risk and
uneven exposure to geographic regions. PE firms are still mainly quoted on European
exchanges, but the sector is expanding into other geographic markets, increasingly offer-
ing access to private companies in East Asia and South America.

Discussion Questions
1. Discuss the advantages and disadvantages of PTPE compared with conventional PE
from an investors viewpoint.
2. Describe the organizational forms of PTPE and explain the similarities and differ-
ences compared with non-listed PE.
3. Explain the reasons for discounts to NAVs in listed PE funds.
4. Discuss how different organizational structures can explain variation in market risk
across PTPE firms.

References
Ammer, John M. 1990. Expenses, Yields, and Excess Returns: New Evidence on Closed-End Fund
Discounts from the UK. Financial Markets Group Discussion Paper No. 108, London School
of Economics.
Anson, Mark J. P. 2002. Managed Pricing and the Rule of Conservatism in Private Equity Portfo-
lios. Journal of Private Equity 5:2, 1830.
AP Alternative Assets. 2006. Listing Prospectus.
Axelson, Ulf, Per Strmberg, and Michael S. Weisbach. 2009. Why Are Buyouts Levered? The Fi-
nancial Structure of Private Equity Funds. Journal of Finance 64:4, 15491582.
Baker, Malcolm, and Jeffrey Wurgler. 2000. The Equity Share in New Issues and Aggregate Stock
Returns. Journal of Finance 55:5, 22192257.
Publ icl y Trade d P riv at e E qu it y 501

Bance, Alex. 2004. Why and How to Invest in Private Equity. EVCA Special Paper, Zaventem.
Barclay, Michael J., Clifford G. Holderness, and Jeffrey Pontiff. 1993. Private Benefits from Block
Ownership and Discounts on Closed-End Funds. Journal of Financial Economics 33:3, 263291.
Berger, Robert. 2008. SPACS: An Alternative Way to Access the Public Markets. Journal of Applied
Corporate Finance 20:3, 6875.
Berk, Jonathan B., and Richard Stanton. 2007. Managerial Ability, Compensation, and the Closed-
End Fund Discount. Journal of Finance 62:2, 529556.
Bilo, Stphanie, Hans Christophers, Michl Degosciu, and Heinz Zimmermann. 2005. Risk, Returns,
and Biases of Listed Private Equity Portfolios. WWZ Working Paper No. 105, University of Basel.
The Blackstone Group. 2007. Listing Prospectus.
Boudreaux, Kenneth J. 1973. Discounts and Premiums on Closed-End Mutual Funds: A Study in
Valuation. Journal of Finance 28:2, 515522.
Brauer, Greggory A. 1984. Open-Ending Closed-End Funds. Journal of Financial Economics 13:4,
491507.
Brickley, James, Steven Manaster, and James Schallheim. 1991. The Tax-Timing Option and the
Discounts on Closed-End Investment Companies. Journal of Business 64:3, 287312.
Castle Private Equity AG. 2000. Listing Memorandum.
Chay, J. B., and Charles A. Trzcinka. 1999. Managerial Performance and the Cross-Sectional Pric-
ing of Closed-End Funds. Journal of Financial Economics 52:3, 379408.
Cherkes, Martin, Jacob Sagi, and Richard Stanton. 2009. A Liquidity-Based Theory of Closed-End
Funds. Review of Financial Studies 22:1, 257297.
Cogent Partners. 2010. Secondary Pricing Trends & Analysis, Report. January. Available at http://
ilpa.org/wp-content/uploads/2010/02/Cogent-January-2010-Pricing-Paper.pdf.
Constantinides, George M. 1984. Optimal Stock Trading with Personal Taxes: Implications for
Prices and the Abnormal January Returns. Journal of Financial Economics 13:1, 6590.
Cumming, Douglas J. 2003. The Structure, Governance and Performance of UK Venture Capital
Trusts. Journal of Corporate Law Studies 3:2, 191217.
Cumming, Douglas J., Grant Fleming, and Sofia Johan. 2011. Institutional Investment in Listed
Private Equity. European Financial Management 17:3, 594618.
Davidoff, Steven M. 2008. Black Market Capital. Columbia Business Law Review 2008:1, 172268.
De Long, J. Bradford, Andrei Shleifer, Lawrence H. Summers, and Robert J. Waldmann. 1990.
Noise Trader Risk in Financial Markets. Journal of Political Economy 98:4, 703738.
Diller, Christian, and Christoph Kaserer. 2009. What Drives European Private Equity Returns?
Fund Inflows, Skilled GPs, and/or Risk? European Financial Management 15:3, 302335.
Dimson, Elroy, and Carolina Minio-Kozerski. 1999. Closed-End Funds: A Survey. Financial Mar-
kets, Institutions and Instruments 8:2, 141.
Hale, Lola Miranda. 2007. SPAC: A Financing Tool with Something for Everyone. Journal of Cor-
porate Accounting & Finance 18:2, 6774.
Kadapakkam, Palani-Rajan, Lalatendu Misra, and Sinan Yildirim. 2005. Open-Ending of Closed-
End Funds: An Analysis of Discount Reduction. Working Paper, University of Texas.
Kaplan, Steven N., and Antoinette Schoar. 2005. Private Equity Performance: Returns, Persistence
and Capital Flows. Journal of Finance 60:4, 17911823.
Kaplan, Steven N., and Per Strmberg. 2009. Leveraged Buyouts and Private Equity. Journal of Ec-
onomic Perspectives 23:1, 121146.
Kaserer, Christoph, and Christian Diller. 2004. European Private Equity FundsA Cash Flow
Based Performance Analysis. In European Private Equity and Venture Capital Association,
Performance Measurement and Asset Allocation of European Private Equity Funds, 2768. EVCA
Research Paper, Brussels.
Kaserer, Christoph, Henry Lahr, Valentin Liebhart, and Alfred Mettler. 2010. The Time-Varying
Risk of Listed Private Equity. Journal of Financial Transformation 28, 8793.
KKR Private Equity Investors. 2006. Listing Prospectus.
Kothari, S. P., and Jay Shanken. 1997. Book-to-Market, Dividend Yield, and Expected Market Re-
turns: A Time-Series Analysis. Journal of Financial Economics 44:2, 169203.
502 t r e n d s i n p r i vat e e q u i t y

Kumar, Raman, and Gregory M. Noronha. 1992. A Re-examination of the Relationship between
Closed-End Fund Discounts and Expenses. Journal of Financial Research 15:2, 139147.
Lahr, Henry. 2010. Pricing of Listed Private EquityRisk and Return, Net Asset Values, and Loss Aver-
sion. PhD dissertation, Technische Universitt Mnchen.
Lahr, Henry, and Florian T. Herschke. 2009. Organizational Forms and Risk in Listed Private
Equity. Journal of Private Equity 13:1, 8999.
Lahr, Henry, and Christoph Kaserer. 2010. Net Asset Value Discounts in Listed Private Equity Funds.
CEFS Working Paper No. 20092012. Available at http://dx.doi.org/10.2139/ssrn.1494246.
Lahr, Henry, and Andrea Mina. 2014. Liquidity, Technological Opportunities and the Stage Distri-
bution of Venture Capital Investments. Financial Management 43:2, 291325.
Lang, Mark H., Karl V. Lins, and Mark G. Maffett. 2012. Transparency, Liquidity, and Valuation:
International Evidence on When Transparency Matters Most. Journal of Accounting Research
50:3, 729774.
La Porta, Rafael, Florencio Lopez-de Silanes, and Andrei Shleifer. 2006. What Works in Securities
Laws? Journal of Finance 61:1, 132.
Lee, Charles M. C., Andrei Shleifer, and Richard H. Thaler. 1990. Anomalies: Closed-End Mutual
Funds. Journal of Economic Perspectives 4:4, 153164.
Lee, Charles M. C., Andrei Shleifer, and Richard H. Thaler. 1991. Investor Sentiment and the
Closed-End Fund Puzzle. Journal of Finance 46:1, 75109.
Levis, Mario, and Dylan C. Thomas. 1995. Investment Trust IPOs: Issuing Behavior and Price Per-
formance. Evidence from the London Stock Exchange. Journal of Banking and Finance 19:8,
14371458.
Loughran, Tim. 1997. Book-to-Market across Firm Size, Exchange, and Seasonality: Is There an
Effect? Journal of Financial and Quantitative Analysis 32:3, 249268.
Malkiel, Burton G. 1977. The Valuation of Closed-End Investment-Company Shares. Journal of
Finance 32:3, 847859.
Martin, John D., and J. William Petty. 1983. An Analysis of the Performance of Publicly Traded Ven-
ture Capital Companies. Journal of Financial and Quantitative Analysis 18:3, 401498.
Martynova, Marina, and Luc Renneboog. 2008. A Century of Corporate Takeovers: What Have
We Learned and Where Do We Stand? Journal of Banking and Finance 32:10, 21482177.
Pstor, ubos, and Robert F. Stambaugh. 2003. Liquidity Risk and Expected Stock Returns. Jour-
nal of Political Economy 111:3, 642685.
Phalippou, Ludovic. 2007. Investing in Private Equity Funds: A Survey. CFA Institute Research
Foundation Literature Reviews 2:2, 122.
Phalippou, Ludovic, and Oliver Gottschalg. 2009. The Performance of Private Equity Funds.
Review of Financial Studies 22:4, 17471776.
Pontiff, Jeffrey. 1995. Closed-End Fund Premia and ReturnsImplications for Financial Market
Equilibrium. Journal of Financial Economics 37:3, 341370.
Ratos AB. 2012. Annual Report.
shaPE Capital. 2001. Information Memorandum.
Thompson, Rex. 1978. The Information Content of Discounts and Premiums on Closed-End Fund
Shares. Journal of Financial Economics 6:23, 151186.
Weidig, Tom, Andreas Kemmerer, Tadeusz Lutoborski, and Mark Wahrenburg. 2007. Private
Equity Fund Managers Do Not Overvalue Their Company Investments. In Greg N. Grego-
riou, Maher Kooli, and Roman Krussl, eds., Venture Capital in Europe, 157170. Oxford:
Butterworth-Heinemann.
Weiss, Kathleen. 1989. The Post-Offering Price Performance of Closed-End Funds. Financial Man-
agement 18:3, 5767.
Zweig, Martin E. 1973. An Investor Expectations Stock Price Predictive Model Using Closed-End
Fund Premiums. Journal of Finance 28:1, 6778.
27
The Future of Private Equity
A Global Perspective

DIANNA C. PREECE
Professor of Finance, University of Louisville

Introduction
Private equity (PE) is a term widely used by financial professionals and average inves-
tors alike. In the mid-1950s, PE firms were known as buyout firms. These firms evolved
and experienced rapid growth in the early 1980s. Fueled by debt, the 1980s saw a boom
in mergers and acquisitions. During this period, PE firms were rebranded and became
known as leveraged buyout (LBO) firms. According to Alison Mass, co-head of the
Financial Sponsors Group in the Investment Banking Division at Goldman Sachs,only
six PE firms existed in the 1980s with greater than $1 billion in assets under manage-
ment (AUM) (Mass 2013). Today, more than 500 firms meet the same criterion.
The PE industry is undergoing major changes. Uncertainty exists about changes in
regulation, the tax treatment of both PE firms and their investors, investor demands
as firms engage in fundraising, and the operational activities of firms. The landscape is
also changing as the proportion of PE investments increase in Asia and the developing
world. As such, PE firms must adapt to an ever-evolving landscape and manage the de-
mands of investors and regulators.
Little peer-reviewed academic work exists about the future of PE. First, as noted
previously, the industry is relatively young. Second, academic studies often rely on his-
torical data for analysis. Thus, studies are generally backward looking, focusing more
on the past than on the future. In contrast, the works cited in this chapter are a mix of
recent academic studies on foundational ideas in PE combined with popular press ar-
ticles and ideas about where the industry is headed on the domestic and global fronts.
Additionally, this chapter references governmental and regulatory websites on regu-
latory changes that will affect the industry. Finally, the chapter includes content from
PE firm websites and industry associations and interviews with leaders in the field that
explore expected changes relevant to the PE industry.
The purpose of this chapter is to provide insights about the future of PE from a global per-
spective. The remainder of this chapter includes a brief discussion of PE associations and the
role they play in the industry. This topic is followed by a discussion of trends in globalization

503
504 t r e n d s i n p r i vat e e q u i t y

in the PE industry, including a look at where investors have and are likely to deploy funds.
Next, the chapter considers regulatory changes that will affect PE, both enacted and in the
discussion phase. A discussion of investor trends and how changes in investor behavior affect
the PE industry follows this topic. The final section offers a summary and conclusions.

Private Equity Associations


Numerous interests and concerns affect the PE industry. Since the early 2000s, several
PE associations have been established to address many issues of interest to PE manag-
ers and investors. General partners (GPs), who manage PE firms, and limited partners
(LPs), who invest in PE firms, are concerned about potential regulations that may affect
flexibility, reporting, taxation, and returns. PE firms rely on the advocacy and lobby-
ing activities of associations to communicate these concerns to Washington policy and
lawmakers. PE associations conduct and report research on relevant topics to GPs and
investors. Moreover, some associations have been established to meet the needs of spe-
cific geographic investors. These associations provide members with research on the
trends and challenges of foreign investments, particularly emerging markets. Some of
the key PE associations are discussed below.
One PE association is the Private Equity Growth Capital Council (PEGCC). The
council, established in 2007 and located in Washington, DC, is an advocacy, commun-
ications and research organization and resource center established to develop, analyze
and distribute information about the private equity and growth capital investment
industry and its contributions to the national and global economy (Private Equity
Growth Capital Council 2014, p. 1). PEGCC members include some of largest and
best-known PE firms such as The Carlyle Group, The Blackstone Group, and Kohlberg,
Kravis and Roberts (KKR). The council provides research on relevant topics to mem-
bers and nonmembers on its website. The PEGCC works to promote and preserve poli-
cies that encourage investment in innovation, jobs, and competitiveness. The website
also provides members with a forum to discuss trends and challenges in the industry.
Another prominent PE association is the Emerging Markets Private Equity Associa-
tion (EMPEA), which was established in 2004 and supports PE investment in emerging
markets. Like the PEGCC, the EMPEA provides research and data on topics of interest
to PE investors, with a specific focus on emerging markets. The group seeks to increase
capital flows and spur economic growth in emerging markets. For example, EMPEA
unveiled Key Elements of Legal and Tax Regimes Optimal for the Development of Private
Equity in 2012. The report is available at the EMPEA website (Emerging Markets Pri-
vate Equity Association 2014a). EMPEA Guidelines identify elements of legal and tax
regimes that experience from other markets has demonstrated will help attract robust
international and local PE and venture capital (VC) investment. The Guidelines are
intended to serve as a general framework for constructive dialogue among all stake-
holderspolicymakers, regulators, and investorsto strike a balance between effec-
tive protections and excessive restrictive regulatory approaches that stymie investment
flows. Reports such as the Key Elements of Legal and Tax Regimes Optimal for the Devel-
opment of Private Equity improve capital flows to emerging markets by reducing uncer-
tainty for investors.
Th e F u t u re of P riv at e E qu it y 505

Besides these two associations, many regional associations and online platforms are
devoted to teaching, researching, and connecting PE market participants from around
the world. For example, the European Private Equity and Venture Capital Association
(EVCA) (2014) is an advocacy group located in Brussels, Belgium that has more than
700 members. The group describes itself as one that explains PE to the public and helps
shape public policy around PE and VC issues. The Latin American Private Equity and
Venture Capital Association (LAVCA) (2014) has a similar goal focusing on the op-
portunities and challenges of Latin American investments.

Globalization
The largest PE firms include The Carlyle Group, TPG Capital, The Blackstone Group,
Apollo Management, Bain Capital, and Oaktree Capital Management. According to
Preqin (2013a), a PE data and intelligence services company, AUM reached a record $3.3
trillion in 2012, calculated as uncalled commitments (about 35 percent) plus the market
value of portfolio companies (the remaining 65 percent). Uncalled commitments, the
money that funds have raised but have not yet invested, are called dry powder in the PE
industry. The PE industry had only $100 billion in AUM as recently as 1994 (Fenn and
Liang 1998) and has undergone exponential growth during the past 20 years. According
to Mass (2013), PE firms began expanding into Europe in the 1990s. The 2000s saw the
expansion of U.S. PE firms into Asia and in the last few years funds have begun entering
Latin American markets. While these firms continue to expand globally, new firms are
likely to surface to fill local funding needs in the next decade, according to industry leaders
such as David Rubenstein, co-founder and chairman of the Carlyle Group (Garten 2013).

EUROPE
Investors entered Europe in search of private investing opportunities in the early 1990s.
The European PE market remains robust despite the economic downturn in 20072008.
Strmberg (2009) surveys academic literature on PE in Europe, but removes studies
conducted by PE firms for consistency and to remove bias. He notes that much of the
relevant research centers on U.S. firms because the European PE market is young and
data are scarce. He focuses on several aspects of PE including the impact of PE on eco-
nomic growth in Europe. Based on the relevant research (Meyer 2006; Kaplan, Sensoy,
and Strmberg 2008; Strmberg 2008; Davis, Haltiwanger, Jarmin, Lerner, and Mi-
randa 2011), Strmberg (2009) concludes that a positive relationship exists between
economic growth and PE investments. He notes, however, that no rigorous study has
examined whether PE affects growth in gross domestic product (GDP) because con-
trolling for the reverse causality explanation (i.e., that growth causes PE investment,
rather than PE investment causes growth) is difficult.
Strmberg (2009) finds strong support for the argument that venture capitalists
(VCs) provide substantial post-investment value-added support in both Europe and
the United States. Bottazzi, Da Rin, and Hellmann (2008) present evidence indicating
that VCs assist portfolio companies with fundraising, recruiting, business and market-
ing plan development, and mergers and divestment activities.
506 t r e n d s i n p r i vat e e q u i t y

Regarding PE and innovation, Strmberg (2009) asserts that the latest research
using U.S. PE data suggests that new firm creation and the number of patent counts
increase with increases in VC investment. He notes that studies using European data
are less definitive. Some studies show higher patent activity in countries with more VC
investments but other studies indicate that the companies receiving VC funds are ini-
tially more innovative and the innovation cannot be attributed to VC investment (Engel
and Keilbach 2007; Mollica and Zingales 2007; Caselli, Gatti, and Perrini 2008; Popov
and Roosenboom 2009).
More recently, Davis, Haltiwanger, Handley, Jarmin, Lerner, and Miranda (2013)
examine U.S. buyouts from 1980 to 2005 to find out if leveraged buyouts (LBOs) bring
massive job losses and little operational gains, as is often implied in popular media
sources. The authors control for industry, size, age, and prior growth. They report that
relative to the control firms, employment at target firms falls by 3 percent over the two
years following the LBO. Employment also falls by 6 percent over the five years follow-
ing the LBO. However, target firms also create jobs at new firms and thus the authors
estimate the net job loss is a mere 1 percent over the two years following the LBO. They
also find evidence of productivity gains in target firms that result from a focused reallo-
cation of jobs. Productivity, job loss, and innovation are important as the PE industry
engages in policy debates on the risks and rewards of PE to society as a whole. These
issues are part of the current regulatory debate in the discussion that follows.
Since 1998, Berlin has played host to a PE conference called the Super Return
German Private Equity Summit. At the 2014 Summit held in late February, a new trend
emerged involving investing in Southern Europe again (Schuetze 2014). Investors have
previously shunned the region. Financial markets in countries such as Greece, Spain,
and Italy suffered dramatically as a result of the European debt crisis that took hold in
early 2009. Due to the debt crisis, PE firms avoided investments in these countries but
now see opportunities as government reforms take hold. Also, the region is still cheap
relative to northern Europe and investors are finally more confident the region will not
relapse. According to research group Preqin (2013b), the PE industry had $1.074 tril-
lion of uninvested capital at the end of December 2013. As Schuetze points out, U.S.
firms such as Apollo Management and The Carlyle Group suggested at the Berlin con-
ference that Spain and Italy were potential target areas for investment as their econo-
mies have turned around. Moreover, both nations avoided government bailouts, unlike
their Eastern European neighbors Greece and Portugal.
One concern that still plagues European venture investors and U.S. PE funds is
how to exit pre-financial crisis investments (i.e., boom year deals). Little academic re-
search exists on this topic due to the short time horizon following the financial crisis
of 20072008 to study the issue. However, the subject is a widely discussed phenom-
enon in the popular PE literature and in PE industry reports (Altius Associates 2012;
Deloitte Center for Financial Services 2012). First, many funds are heavily invested in
pre-financial crisis deals at inflated valuations. The five-year mark (i.e., the time when
PE investors expect to begin seeing the return of capital) has passed for many of these
deals, and GPs feel pressure as LPs expect a return. Initial public offering (IPO) markets
have remained relatively tight, although volume increased substantially in 2013 (Ogg
2013) and the stock price performance of many IPOs was strong. According to Dezem-
ber (2013), PE firms returned record amounts of capital to investors in 2013, taking
Th e F u t u re of P riv at e E qu it y 507

advantage of a strong buyout market that provided opportunities for exits in IPO mar-
kets and debt deals that paid investors large dividends. PE returned about $120 billion
to investors in 2013.
Macroeconomic uncertainties remain for many European economies. Funds will
likely continue to look at alternative channels to exit boom year deals to return capital
to LPs. For example, industry leaders expect to see PE firms sell more portfolio compa-
nies to other PE firms in the coming decade. Many PE firms are also selling to public
investors and corporations. Companies have record amounts of cash. According to the
Deloitte Center for Financial Services (2012), U.S. firms held an estimated $2 trillion
of cash and Japanese firms held $2.5 trillion at the end of 2012. Additionally, many PE
firms offer a broader mix of asset management products, allowing them to more effi-
ciently use established exit channels. According to Altius Associates (2012), KKR has
successfully tapped new sources of liquidity in the last several years. For example, the
firm, along with Goldman Sachs, sold a 25 percent stake in German forklift manufac-
turer Kion to Chinas Shandong Industries in late 2012 and announced an additional
sale of 10.8 percent of Kion in January 2014. Selling portfolio companies to public com-
panies such as Shandong Industries is a relatively new phenomenon for PE firms and
the trend is expected to continue as companies search for alternative sources of liquidity.
Additionally, sovereign wealth funds are potential capital providers to PE firms
trying to exit boom year deals. Sovereign wealth funds are government-owned funds that
invest in real and financial assets. These funds invest globally and are funded by foreign
exchange reserves and export revenues. According to Preqin (2013a), sovereign funds
had $5.35 trillion in AUM in early 2014. Sovereign fund AUM are expected to increase
to $8 trillion by 2020 (Spector 2014). An increased demand exists for investment op-
portunities from sovereign funds. In many cases, however, sovereign wealth funds are
choosing direct investments, rather than investments via PE funds, as a means of avoid-
ing PE management fees (Dunkley 2014). Yet, direct investment strategies have risks.
According to some industry leaders, sovereign fund managers may have the resources
for co-investment deals with VCs and other PE participants, but may not have the ex-
pertise to best execute direct investments.

ASIA
Strong PE investment has occurred in Asia for many years and the trend is expected
to continue. In an early examination of PE investments in Asia, Bruton, Dattani, Fung,
Chow, and Ahlstrom (1999) examine the differences between PE in China versus the
western world. The authors conduct 20 interviews with PE fund managers and experts
in Chinese PE markets. Their evidence shows that Chinas socialist economy with some
capitalist characteristics affects the goals and objectives of PE investing. Chinese PE
firms face challenges that are unknown in mature economies such as the United States
and Western Europe.
In an effort to more clearly define these challenges, Bruton and Ahlstrom (2003)
expand upon the research by interviewing 24 Chinese VCs in 36 independent inter-
views to ascertain the differences between VC investing in China versus the West. The
authors identify two key concerns. First, the regulatory environment in China is weak.
Second, accounting data are generally less reliable in China than in the West. These two
508 t r e n d s i n p r i vat e e q u i t y

structural weaknesses make investing in China difficult. The authors also note that the
Chinese government may influence the goals of portfolio companies. This means that
VCs cannot necessarily assume a profit motive as is common in the West. Bruton and
Ahlstrom conclude that the Chinese PE market is complex and success is often elusive
for investors. As a result of these fundamental differences with the western PE model,
the authors further contend that the VCs relationships with entrepreneurs are even
more critical in China than in the United States and Europe. These relationships be-
tween capital providers and managers of portfolio companies will be critical as PE con-
tinues to expand in Asia, especially to lesser known markets.
Despite the difficulties with PE investing identified in early studies, Chinese financial
markets continue to liberalize. China has emerged as the largest PE market in Asia, sur-
passing Japan, which had the greatest number of PE firms in Asia in the late 1990s. Ac-
cording to Altius Associates (2012), new legislation in China facilitated PE investment
by local investors in 2005. Since local Chinese investors have difficulty providing capital
to offshore funds (i.e., U.S. and Western European funds), PE has become increasingly
popular with local investors and local fund raisers. Cash-rich sovereign funds are prev-
alent in the market as well. While generalists initially managed these sovereign funds,
local government authorities have begun to see PE as a means of attracting capital to
specific geographic locations, targeting economic growth.
Besides local investment in Chinese PE funds, some large U.S. firms have received
permission to raise renminbi (RMB) funds. The renminbi is the official currency of the
Peoples Republic of China and literally means peoples currency. While Chinese of-
ficials typically use the term to describe the Chinese currency, it is too formal for every-
day use. The term yuan is used colloquially in reference to the Chinese currency and
is the actual unit of currency in China (as the dollar is the actual unit of currency in the
United States). RMB funds are raised in local currency. An RMB fund is purchased with
renminbi (i.e., yuan) and gains and losses are based on the monetary value of Chinese
currency at the time of purchase or sale of the asset (Benge 2014).
Chinese PE firms such as New Horizon, CDH, and Orchid have successfully raised
RMB funds (Deng 2013). Also, in recent years some U.S. firms have received licenses to
raise RMB funds. Beginning in 2007, global PE funds started raising RMB funds. For ex-
ample, Blackstone successfully raised an RMB fund in 2009 and 2010. The Carlyle Group
and TPG Capital have also raised RMB funds. When raising these funds, the U.S. firms
hoped that the local currency fund would be treated as a domestic fund in China, with the
same rights as local funds (i.e., access to investment targets and local investors comparable
to those of Chinese funds). However, some uncertainty exists about whether the Chinese
government considers these funds domestic or foreign, so investments and fund raising
has slowed. A slowing Chinese economy and declining returns has exacerbated the down-
ward trend. According to Deng, in 2012, U.S. firms invested just $71 million worth of yuan
in China compared with the $723 million invested by Chinese PE companies.
Batjargal and Liu (2004) examine 158 VC investment decisions in the Peoples Re-
public of China to determine the role that social capital plays in the investment deci-
sions of VCs. The authors find an entrepreneurs social capital influences the investment
decisions of VCs. They also find that strong ties between VCs and entrepreneurs affect
valuations, covenants, and investment delivery. This study reinforces the notion alluded
to previously that relationships are critical when doing business in China.
Th e F u t u re of P riv at e E qu it y 509

L AT I N A M E R I C A
Funds allocated to Latin America reached a six-year high in 2013, according to the Latin
American Private Equity and Venture Capital Association (2014). Firms committed
$8.9 billion in 233 deals to Latin America in 2013. Funds allocated to Latin America
increased more than 13 percent relative to 2012, with $7.9 billion invested. Brazil con-
tinued to garner the greatest attention in Latin America, attracting 72 percent of the
money invested and 62 percent of the deals in 2012. However, in 2013, major buyouts
were funded in Brazil, Chile, and Colombia. Trends show that in Latin America smaller
firms close the majority of deals. Also, funds are looking not only at Brazil, which is the
most popular area for investing in Latin America, but also more broadly for investments
in countries such as Chile and Colombia.

EMERGING MARKETS
Historically, investors interested in emerging markets focused on Brazil, Russia, India,
and China, known as the BRICs. According to Delevingne (2014), the PE industry is
beginning to shift its attention away from the BRICs toward Southeast Asia and sub-
Saharan Africa. China and India continue to be strong markets for PE capital flows but
investors are looking for new growth opportunities. PE funds raised $88 billion in 2011
and 2012 for emerging market investments, which were deployed in both BRIC and
non-BRIC markets in 2013. According to Maryam Haque, EMPEAs Head of Data and
Analysis (Emerging Markets Private Equity Association 2014b, p. 3),

While capital continued to flow to the BRICs, fund managers also looked
to traditionally less active markets and segments as they seek the next wave
of growth. The less favorable macroeconomic conditions in some emerging
markets can actually make it easier for fund managers to more clearly view
the finances and operations of a company to better identify value. By mapping
where long term investors are deploying capital, we begin to see the next wave
unfold.

In a report on emerging market investments, Pries and Berla (2012) note that the
benefits of emerging market investing are substantial and accrue to both PE funds and
investors and to the countries themselves. For example, capital can help companies
grow, leading to job creation and higher tax revenues. Citizens may also be empowered
by greater financial strength to demand accountability from governments. Education
and better healthcare are also benefits that develop from economic growth. These ben-
efits will continue to appeal to socially conscious investors in the next decade.
PE is likely to continue to expand across the globe. David Rubenstein, co-founder
of the Carlyle Group, believes that one major change the market will see in the next
decade will be the emergence of large, global PE firms developing outside the United
States (Garten 2013). Currently, all large global firms are based in the United States.
Rubenstein also suggests that U.S.-based PE firms will continue to expand their global
reach. As PE investors increasingly face difficulty finding good deals at home, firms
will likely continue to look to untapped markets such as Southeast Asia and East Africa
510 t r e n d s i n p r i vat e e q u i t y

for growth opportunities. According to Rubenstein, the dollar may not be the domi-
nant currency in the future, which will benefit PE firms outside the United States. Fi-
nally, he proposes that a more multilateral, distributed governance structure will emerge
worldwide, which will benefit emerging markets. The GDP of emerging markets is fast
approaching that of the developed world, giving emerging markets more power in both
the global political and economic landscapes. All of these factors are likely to affect PE
investments in the coming decades.

Regulatory Changes
Historically, one appealing aspect of PE to funds and private wealth investors was the
lack of regulation in the industry. The PE industrys preferred regulatory approach was
self-governance. However, the period of nearly perfect autonomy ended after the finan-
cial crisis and the increased regulation that resulted. What makes PE and hedge funds
private is, according to Spangler (2013, p. 1), a series of affirmative decisions taken
by the proposed funds sponsors, and explicitly agreed to by prospective investors, to
operate within designated exemptions to securities laws and financial regulations. After
the financial crisis, the Dodd-Frank Wall Street Reform and Consumer Protection Act
(the Dodd-Frank Act) in 2010, and other regulatory changes have permanently altered
the landscape for PE and hedge funds. The next section provides a brief summary of
some of the key regulatory changes that will affect the future of the PE industry.

THE DODD-FRANK ACT


The Dodd-Frank Act (or simply Dodd-Frank), which is large and complex, will have
both direct and indirect effects on the PE industry. Many expect it to be a boon for
lawyers as firms try to comply with the requirements of the Act. This section provides
a discussion of some provisions of the Dodd-Frank Act that will affect the future of PE
in the United States.

The Volcker Rule


A key provision of the Dodd-Frank Act in section 619 that will affect PE is the Vol-
cker Rule, which went into effect December 10, 2013. It intends to prevent banks and
other financial institutions that benefit from Federal Deposit Insurance Corporation
(FDIC) deposit insurance from engaging in risky behaviors. Taxpayers in essence guar-
antee borrowed funds (i.e., deposits). Additionally, banks can borrow at the Federal Re-
serves discount window. As such, many policymakers, including Paul Volcker, former
Chairman of the Federal Reserve, argue that banks should be prohibited from engaging
in speculative or proprietary trading because taxpayers are ultimately responsible for
losses incurred by risky trading. As a result, the Volcker Rule prohibits banking entities
from engaging in proprietary trading and investing in or sponsoring hedge funds and PE
funds. This rule is subject to certain exceptions.
According to the International Association of Risk and Compliance Professionals
(IARCP) (2014), the hedge fund and PE fund provisions of the Volcker Rule gener-
ally prohibit a banking firm from investing in, or having certain relationships with, any
Th e F u t u re of P riv at e E qu it y 511

fund that is structured under exclusions commonly used by hedge funds and PE funds
under the Investment Company Act of 1940, commonly referred to as the Investment
Company Act. Specifically, the Volcker Rule amends the Bank Holding Company Act
of 1956 to prohibit banks from engaging in proprietary trading and taking an owner-
ship interest in a hedge fund or a PE fund. This provision of Dodd-Frank is critical for
the future of PE because it means that bank debt, used to finance LBOs, will be much
scarcer in the future. This provision means PE firms must find other potentially more
expensive sources of capital to finance buyouts.

Adviser Registration
PE firms have long relied on the Principal Advisers Act exemption, commonly referred
to as the fewer than 15 clients exemption, for registering with federal and state author-
ities, including the Securities and Exchange Commission (SEC). Before Dodd-Frank,
PE firms having fewer than 15 clients were generally exempt from SEC registration. In-
vestment advisers could count any fund that they advised as one client. In other words,
firms did not have to count each individual investor in a fund as a client. This meant that
few PE firms had to register as investment advisers and thus did not have to abide by the
disclosure obligations required under the Investment Advisers Act. Now, as a result of
Dodd-Frank, most advisers must register as this exemption is no longer available. Firms
with $150 million or more in AUM must now register with regulators as investment
advisers. For large PE firms, much of the information required is already reported to
LPs so the change may not make a major difference. Still, it will improve record keeping
and increase accounting and legal costs for most firms that were previously unregistered
(Kreutzer 2012). Registered firms will be subject to the SEC regulatory rules, oversight,
and examination to which mutual funds have long been subject.
The reason registration is such a critical issue for the future of PE is that much in-
formation that was proprietary in the past will now be made public as part of the re-
porting requirements. Also, most firms will be open to SEC examinations and many are
unprepared for this greater level of scrutiny. The SEC is launching an initiative to con-
duct focused, risk-based examinations of investment advisers (Securities and Exchange
Commission 2011a). The Office of Compliance Inspections and Examinations (OCIE)
within the SEC will administer the exams. These rules will fill a key gap in the regulatory
landscape, said SEC Chairman Mary L. Schapiro (Securities and Exchange Commission
2011a, p. 1). She goes on to state, In particular, our proposal will give the Commission,
and the public, insight into hedge fund and other private fund managers who previously
conducted their work under the radar and outside the vision of regulators.
What these new rules mean for the future of PE is increased investment in infrastruc-
ture and personnel to handle the new demands imposed by Dodd-Frank and the SEC.
Firms must develop robust compliance policies and also hire and train personnel to
deal with the new regulatory oversight. For those firms without the in-house structure
in place, outside firms will likely step in and help firms ramp up compliance procedures
and infrastructure. This change will mean added expense for small and large firms alike.

Form PF
As part of the Dodd-Frank Act, the SEC proposed new reporting rules that went
into effect June 15, 2012. The rule requires registered investment advisers who have
512 t r e n d s i n p r i vat e e q u i t y

at least $150 million in AUM to file a Form PF (Securities and Exchange Commis-
sion 2011b). Large PE investment advisers, with at least $1 billion in AUM, are sub-
ject to additional reporting requirements in addition to the Form PF. These firms must
report monthly valuations of assets including hard-to-value illiquid assets, information
on credit providers, risk profiles, performance measures, investor concentrations and
more, according to the Private Equity Growth Capital Council (2011). This change will
increase PE operating expenses as firms are forced to improve compliance procedures.

Reporting and Record Keeping


According to Dodd-Frank (Security and Exchange Commission 2011a), private fund
managers have to provide operational information such as the types, sizes, and owner-
ship structures of funds as well as other general fund data. Funds managers must also
identify the gatekeepers. Gatekeepers perform critical roles for advisers and the private
funds they manage and include auditors, prime brokers, custodians, administrators, and
marketers. The goal of these new rules is to detect potential misconduct by fund manag-
ers and other personnel earlier. Before the Dodd-Frank Act, relevant information that
could lead to detecting misconduct in PE and hedge funds was not disclosed. As a result
of these changes, PE firms in the United States are hiring compliance staff and creating
the infrastructure needed to handle the new requirements imposed by Dodd-Frank.
Krainer (2012), who provides an extensive review of the Dodd-Frank Act, identifies
two approaches to achieve future financial stability. The first approach is to create thor-
ough and thoughtful laws and regulations that prevent financial firms from engaging in
risky activities that ultimately impact the economy as a whole. Lawmakers followed this
logic when constructing and enacting the Dodd-Frank Act. The second approach sug-
gested, but not adopted by policy and lawmakers, is to tax financial institutions that are
systemically important. Krainer contends that this tax system would result in payment
by those responsible for increasing risk and the resulting fallout.

NEW MARKETING RULES


Historically, PE and hedge funds could not advertise to attract funds and could only raise
capital from accredited investors. An accredited investor is one who earns $200,000 per
year (or $300,000 jointly with a spouse) or has net assets of $1,000,000 excluding the
principal residence. The notion behind accredited investors is that they possess sufficient
wealth and sophistication to have the financial expertise to fend for themselves. The
goal is to prohibit retail investors from investing in private funds. In other words, those
not classified as accredited investors should not put their retirement funds in PE and
hedge funds because they are not sophisticated enough to understand the risks. The dis-
tinction of an accredited investor is also used in the United Kingdom and across Europe.
Some criticism exists of this distinction because many accredited investors failed to
recognize warning signs regarding previous financial disasters such as Bernie Madoff s
Ponzi scheme. Some of Madoff s investors were hedge funds that should have had the
financial expertise to recognize the suspicious activities of the firm. In the United States,
per capita wealth has increased since the term accredited investor was originally de-
fined. Thus, the SEC is beginning to question who really has the financial expertise to
make regulation superfluous. As such, the Dodd-Frank Act excludes a persons primary
Th e F u t u re of P riv at e E qu it y 513

residence from the net worth calculation, and the SEC will review the definition of an
accredited investor as it applies to natural persons at least once every four years. A nat-
ural person is a legal term used to describe a living, breathing human being as opposed
to a corporation.

The JOBS Act


PE and hedge funds were exempt from SEC oversight mainly because they only raised
capital from accredited investors. In the United States, a ban on general advertising to
raise capital for private funds has existed for a long time. If a firm advertised for in-
vestors through newspapers, magazines, television stations, or radio stations, the fund
would lose its oversight exemption. In a somewhat surprising and controversial move,
President Obama signed into law the Jumpstart Our Business Startup Act, known as the
JOBS Act, on April 5, 2012, with bipartisan support. The JOBS Act is intended to help
small businesses raise financing by relaxing some securities laws.
As a result of the JOBS Act, PE and hedge funds can now market their funds to the
public, but only to accredited investors (Quinlivan 2012). This change will affect angel
investors, also known as angels, as well as PE firms and hedge funds because syndicates
of accredited investors can be assembled via the Internet. According to Ante and Rusli
(2013), angels, as well as other investors, can solicit funds on their websites, through
social media, or through investor syndicate services such as AngelList, which allows a
lead investor such as an angel or seed fund to set the terms of a deal and then syndicate
the deal online. Investors indicate how much they want to contribute to the venture and
the syndicate leader decides which investors to accept. The syndicate leader has a 10 to
20 percent claim on future profits, 5 percent of which goes to AngelList. Platforms such
as AngelList provide investment opportunities previously unavailable to many investors
along with new sources of funding for start-up firms.

Crowdfunding
Crowdfunding is the financing of a project by individuals instead of professional par-
ties such as VCs, angel investors, or banks. Crowdfunding received a major boost when
Congress passed the JOBS Act in March 2012, encouraging individuals to provide
capital to new ventures directly. Banks have historically played the intermediary role
between businesses and households. Individuals save money in bank accounts, while
banks act as intermediaries, lending the funds to businesses and households. In the in-
stance of crowdfunding, the individual invests directly in the entrepreneurial venture
rather than indirectly via an intermediary such as a bank. Funds are typically raised
through the Internet.
Perhaps the most well-known crowdfunding ventures so far have supported the
arts. Kickstarter, the largest crowdfunding platform, has raised funds for high-pro-
file artistic projects such as movies. For example, the platform raised $5.7 million to
produce the Veronica Mars movie from a cancelled television show. Funding for the
movie was completed April 12, 2013. According to Kickstarter Statistics (2014), the
company has raised more than $1 billion, pledged by more than six million people,
funding 59,000 projects. An ironic fact about Kickstarter is that it financed its business
by raising funds from VCs and angel investors, rather than through crowdsourcing.
514 t r e n d s i n p r i vat e e q u i t y

Mollick (2014) reports that equity crowdfunding is largely prohibited worldwide,


including in the United States. Equity crowdfunding is funding by investors who receive
ownership stakes in the new venture in exchange for capital. Equity crowdfunding ac-
counts for less than 5 percent of all crowdfunding investments. Accordingly, the goals of
funders fall largely into three categories:

Philanthropy Model. Some funders see a product or project in which they believe and
provide capital to the venture. In this model, funds are often donated with no expec-
tation of a return on invested capital.
Lending Model. Funds are loaned to the projects founders and those lenders expect a
return on invested capital.
Reward-based Model. Funders receive an award for providing capital. For example,
the award could allow investors creative input on a project, or a small speaking part
in a film if they provide enough capital. Products such as signed CDs could be of-
fered to capital providers if a record is produced.

Currently, the notion of crowdfunding does not apply to private funds or hedge
funds because one cannot crowdfund a fund. However, the ability to market funds
changes the game and may be especially relevant in periods of uncertainty. Marketing to
investors is unlikely to change PE or hedge funds in the short-run because the SEC has
publicly stated that it will scrutinize fund claims about performance (Quinlivan 2012).
Also, if firms use Financial Industry Regulatory Authority (FINRA) placement agents
to assist in fundraising, placement agents must abide by FINRA rules. Finally, the In-
vestment Advisers Act contains anti-fraud measures. Since many PE firms must register
with the SEC as investment advisers under the Dodd-Frank Act, they must abide by the
terms of the Investment Advisers Act. The JOBS Act did not change those terms. Some-
thing that may be helpful to PE firms under the JOBS Act is the IPO On-Ramp pro-
vision. The IPO On-Ramp provision is intended to help emerging companies go public
by relaxing some of the regulatory requirements. Although the goal of the IPO On-
Ramp is job creation, it might also help PE firms use equity markets as an exit strategy.
The SEC is currently reviewing a proposed change to the Securities Act of 1933 and
the Securities Exchange Act of 1934 that would potentially provide for equity-based
crowdfunding. Comments had to be made on the proposed change by February 3,
2014, and as of this writing, have not been acted on (Securities and Exchange Commis-
sion 2013). However, according to Inside the Firm (2014), a forum devoted to careers
in PE and VC, this regulatory change would be good for VC and other PE firms. Al-
though PE funds would likely be unable to raise money via crowdfunding directly, this
change would provide the start-up funds for many potentially profitable ventures. These
funds would mitigate risks for future capital providers and allow firms to choose from
a variety of new ventures with proven track records, without directly providing seed
money to these ventures.

THE BASEL III ACCORD


The Basel III Accord (Basel III) follows the Basel II Accord in establishing standard
capital requirements for the banking industry. The primary purpose of Basel III is to
Th e F u t u re of P riv at e E qu it y 515

increase bank liquidity and decrease bank leverage (Bank of International Settlements
2010). Although Basel III does not have requirements specific to PE, indirect effects
exist. Some banks are divesting their PE investments because Basel III requires banks
to hold more capital against these types of assets (Realdeals 2011). Basel III could also
increase demand for PE if the availability and attractiveness of bank loans declines as a
result of the new rules.

A LT E R N AT I V E I N V E S T M E N T F U N D M A N A G E R S D I R E C T I V E
The Alternative Investment Fund Managers Directive (AIFMD) regulates hedge
funds and other alternative investment funds in the European Union (EU). The pur-
pose of the 2013 regulation is to increase transparency and risk management functions
(European Securities and Markets Authority 2013). The rule is intended to govern
hedge funds and PE more similar to mutual funds and pension funds. According to
Dunkley (2014), industry leaders indicate that the AIFMD has already had a big
impact on the industry and that the regulations will force the PE industry to become
more transparent. This affects both global funds and domestic EU funds.

F O R E I G N A C C O U N T TA X C O M P L I A N C E A C T
The Foreign Account Tax Compliance Act (FATCA), known as the anti-tax dodge
law, was created to mitigate offshore tax evasion. Foreign banks, insurers, and invest-
ment funds are required to send the Internal Revenue Service (IRS) information about
offshore accounts of Americans and permanent residents that are worth more than
$50,000. FATCA is part of the 2010 Hiring Incentives to Restore Employment (HIRE)
Act and was signed into law in March 2010.

P R O P O S E D TA X C H A N G E O N C A R R I E D I N T E R E S T
Following the 2012 presidential election in the United States, a debate ensued about
taxing PE managers (i.e., GPs) profit participation, called carried interest. GPs re-
ceive a management fee, which is taxed as a salary, and share in the funds profits that
exceed a pre-determined hurdle rate. This profit share is taxed as a capital gain (The Tax
Policy Center 2012). Former U.S. presidential candidate Mitt Romney came under
attack because much of his income was taxed at the 20 percent capital gains rate rather
than the ordinary income rate, which caps at 39.6 percent (Appelbaum 2014). A key
issue relating to the taxation of fund managers share of profits involves whether PE
partners are taking an active management role in the portfolio company. In simple
terms, if partners are actively managing portfolio companies, then the carried interest
should be treated as income, like a salary. If not, then the income should be treated as
investment earnings. While this issue continues to be debated, no change has been
made to the tax law as of this writing. David Rubenstein, co-founder of the Carlyle
Group, and considered by many an expert on national politics affecting the PE indus-
try, said in 2014 that he does not believe the current Congress will pass any tax reform
(Spector 2014).
516 t r e n d s i n p r i vat e e q u i t y

R E G U L ATO R Y G U I D A N C E O N L B O S
In March 2013, the Office of the Comptroller of the Currency (OCC) and the Federal
Reserve issued guidance warning banks against financing buyouts in which the leverage
was more than six times earnings before interest, tax, depreciation, and amortization
(EBITDA). Regulators are attempting to crack down on risky banking practices. In part
as a result, several banks, including Citigroup, JP Morgan Chase, and Bank of Amer-
ica, have recently decided against financing some corporate takeovers (Tan 2014). This
change will make deals more expensive for PE firms, as they often rely on bank debt to
provide inexpensive financing for takeovers. Martin Pfinsgraff, the OCCs senior deputy
comptroller for large bank supervision (Tan 2014, p. 1) said:

The impact on private equity, a significant driver of what we see as risky prac-
tices, is an intended consequence of our actions. As regulators, we certainly
hope to change bad practices and remove the extraordinary froth thats expe-
rienced at the peak of a credit cycle. If we can mitigate that, it reduces the size
of the valley to follow.

As Tan (2014) indicates by referencing Pfinsgraff, regulators intend to cut off bank
financing to PE for high leverage deals. This means that going forward, PE firms will
have to rely on other sources of financing, perhaps from non-bank lenders or securities
dealers, which are not subject to the strict regulatory authority of banks.

Investor Trends
Several trends have emerged since the financial crisis of 20072008. The PE investor
base continues to move toward institutions such as pension funds and foundations.
Also, investors continue to become more sophisticated and demand more from funds.
The following section provides a discussion of these and other trends.

I N S T I T U T I O N A L I N V E S T M E N T I N P R I VAT E E Q U I T Y
A key trend in PE that is expected to continue is the increased investment by institu-
tions in the asset class (Metrick and Yasuda 2011). Total AUM exceeded $3 trillion
in 2013 (Preqin 2013b). Much of the growth from 1994 to 2013 (i.e., from $500
billion in AUM to about $3 trillion in AUM) has been from institutional investors.
For example, according to privateequity.com (2013) pension fund target allocations
to PE increased from 7.5 to 8.3 percent between January 2012 and January 2013. For
large pension funds with a minimum of $5 billion in AUM, the increase was even
greater, up to 9.7 percent from 8.3 percent. Reasons cited for the increase generally
focus on making up losses suffered during the financial crisis of 20072008. Also, as
Metrick and Yasuda note, the large foundation allocation to PE reached 12 percent
during the pre-crisis. At the time of this writing in mid-2014, bond yields remain low
and many pensions are underfunded, forcing pension funds to look for alternatives
to increase returns.
Th e F u t u re of P riv at e E qu it y 517

Probitas Partners (2012) discuss results from an online survey conducted in Oc-
tober 2012 on PE trends for 2013. The 126 respondents to the survey include senior
investment executives from around the world, including institutional investors such as
corporate and public pension funds, foundations, endowments, fund-of-funds, banks,
consultants, and advisers. Of the respondents, 78 percent were from Europe or North
America. Investors identified the following PE trends:

A steady and potentially increasing demand for PE going forward.


A limited appetite for new managers but instead a focus on consolidating current
fund manager relations.
A focus on smaller and mid-market buyout and growth capital funds in Europe and
the United States and on country-focused funds in Asia.
A greater concern with risks, such as political risk in emerging markets, with a re-
duced inclination to believe in high growth.
The continued dominance of Brazil and China in emerging market investment with
Turkey and Indonesia as rising stars and investors showing declining interest in
India.
A rising popularity in credit-oriented strategies in the United States.
A continued low interest in VC investing.
The development of co-investment programs by large investors to enhance returns.
A co-investment is a minority investment made directly to an operating company.

I N V E S TO R D E M A N D S
One factor that distinguishes PE firms from angel investors or private investment com-
panies is that PE firms do not use their own capital. Today, investors are often pension
funds, endowments, and other institutional investors. Potential agency conflicts exist be-
tween GPs and LPs. An agency conflict arises when one group that has been given author-
ity to look after another groups interests acts in its own self-interest instead. Contractual
provisions in the limited partnership agreement (LPA) may be used to mitigate agency
costs. Additionally, portfolio companies are private, increasing informational asymme-
tries, which can increase agency costs. Finally, PE firms must return money to investors,
unlike corporations that make strategic investments. All of these factors distinguish PE
from mutual funds, hedge funds, and corporations (Metrick and Yasuda 2011).
Following the financial crisis of 20072008, PE investors have become more de-
manding. According to Dunkley (2014), investors are demanding more co-investment
opportunities with LBO firms. Investors are also looking for lower fees and requesting
earlier access to deal pipelines. Many PE firms are raising capital again. As Shah (2013)
notes, LPs are tightening deal terms involving both performance and management fees.
According to the Deloitte Center for Financial Services (2012, p. 2), increasingly so-
phisticated LPs are also prompting PE shops to improve their compliance capabilities
by asking for improved transparency and customized reporting. The report indicates
that LPs are engaged in more substantial due diligence efforts, prolonging the due dili-
gence process and comprehensively digging into a firms operational and organizational
infrastructure to assess its resilience (Deloitte Center for Financial Services 2012, p. 2).
518 t r e n d s i n p r i vat e e q u i t y

In an industry that has not been required to consider compliance much in the past, a
culture of compliance will take time but will be necessary for firms to attract funds as
investors become increasingly focused on this issue.
Traditionally, PE firms have relied on GDP growth, multiple expansion, and lev-
erage to drive alpha (i.e., positive excess risk-adjusted returns), according to the De-
loitte Center for Financial Services (2012). Going forward, these sources of alpha
may not be enough to satisfy LP performance demands. PE firms must look at oper-
ational inefficiencies, both at the portfolio firm level and the PE firm level, to increase
returns. According to David Rubenstein of the Carlyle Group, PE is not a financial en-
gineering game, despite what many people believe. He states that the bulk of private
equity returns come from operational improvements made in portfolio companies
(Spector 2014). Several opportunities are emerging for PE firms to increase operational
efficiencies including:

Data management. PE firms, like other firms, are expected to outsource data manage-
ment storage and retrieval to third parties.
Customer relationship management (CRM). CRM is a system for managing a compa-
nys interactions with current and future customers, using technology to organize,
automate, and synchronize sales, marketing, customer service, and technical sup-
port. PE firms will increasingly outsource CRM.
Waterfall automation technologies. One of the most complex tasks of a PE firm is dis-
tribution waterfalls, which refer to the process of dividing the proceeds from realized
investments between investors and fund managers. GPs also receive their perfor-
mance and carried interest through distribution waterfalls. Typically, firms calculate
the distributions manually but doing so is difficult and labor intensive. PR Newswire
(2013) reports that BNY Mellon, a firm that provides services to PE firms, an-
nounced the industrys first third party administrator waterfall solution.
Document management systems. Document management software can be used to
manage deal flows in PE firms.

Historically, PE firms have generally relied on manual processes for data management
and reporting. Regulatory changes such as the Dodd-Frank Act, FATCA, and AIFMD,
along with the increasing costs associated with investors demands for reporting and
increased transparency, mean that firms will be assessing their in-house skills needed
to meet these needs and requirements. However, firms can realize improvements and
efficiencies as a result of these technological solutions. As such, PE firms are expected
to increasingly rely on outsourcing and technology going forward. These technologies
have the potential to create efficiencies and are a tangible lever to deliver alpha (De-
loitte Center for Financial Services 2012, p. 4) in the industry in the next decade.

Summary and Conclusions


Over the last 50 years, PE has continued to evolve. From the early days when firms were
known as buyout firms to the 1980s when firms became leveraged buyout firms,
to today, a period when firms must determine the next step in their evolution, PE has
Th e F u t u re of P riv at e E qu it y 519

been resilient. Looking forward, the future holds major challenges for the industry.
This period is especially difficult because firms must adjust to the most dramatic regu-
latory changes that have occurred in the industrys history. As a result of Dodd-Frank,
firms will, for the first time, have to reveal aspects of their businesses that they have long
kept secret. Most firms must now register with, and be subject to, examination by the
SEC. Additionally, PE firms and hedge funds are affected as banks continue to come
under regulatory scrutiny for risky behaviors. The Volcker Rule prohibits banks from
proprietary trading and equity ownership of PE firms and hedge funds. PE firms will
not have access to the inexpensive debt that has fueled much of the takeover business in
the United States in the last several years. Additionally, investors and regulators alike are
demanding more transparency from firms.
Firms are also facing a period during which they must consider operational ef-
ficiencies to fuel returns; not simply leverage, growth in GDP, and price/earnings
(P/E) multiple expansions. This means considering outsourcing and technology as a
way to improve efficiencies in an industry that has long relied on manual processes for
data management. Firms may also have to provide more co-investing opportunities
for institutional investors, as they balk at high management fees. Taxpayers are also
looking at the industry in the wake of the 2012 presidential election in the United
States and are engaging in widespread discussion of the taxation of carried interest.
While nothing has happened so far, this topic will continue to be discussed by many
in Congress.
Finally, globalization continues to influence the industry, both in terms of oppor-
tunities for investment in emerging markets and competition from new PE firms in
emerging markets. Firms are now looking outside the obvious investment areas such
as India and China to new opportunities in places such as sub-Saharan Africa and East
Asia. The largest firms will likely see competition in the coming decade as more firms
locate outside the United States. Sovereign fund investing is also expected to rise in
the coming decade, again with the demand for higher returns and co-investing oppor-
tunities. Following the financial crisis of 20072008, financial firms need to be more
flexible. PE firms will need to be adaptable in the face of increasing globalization, reg-
ulatory intervention, and a new breed of investors who are both knowledgeable and
demanding.

Discussion Questions
1. Explain the evolution of PE in terms of geography and discuss globalization trends
for the next decade.
2. List the drivers of PE returns in the 2000s and early 2010s and explain what is likely
to have a bigger influence on returns in the latter half of the 2010s and beyond.
3. Discuss the primary trend in institutional investing in PE and the forces driving that
trend.
4. Define a sovereign wealth fund and explain the role that sovereign wealth funds play
in PE.
5. Discuss how the Dodd-Frank Act affects PE.
520 t r e n d s i n p r i vat e e q u i t y

References
Altius Associates. 2012. 10 Challenges Facing the Private Equity Industry in 2013. Available at
http://www.altius-associates.com/downloads/AA_Q1-2013_newsletter.pdf.
Ante, Spencer E., and Evelyn M. Rusli. 2013. Breaking Down the Walls for New Angel Investors:
Easing of Government Restrictions Allows Funding Syndicates Such as AngelList to Flourish.
Available at http://online.wsj.com/news/articles/SB1000142405270230444140457912381
0278384116.
Appelbaum, Eileen. 2014. Private Equity Tax Breaks, How Long Will They Last? Available at
http://finance.fortune.cnn.com/2014/04/10/private-equity-tax-breaks-how-long-will-
they-last/.
Bank of International Settlements. 2010. Group of Governors and Heads of Supervision Announce
Higher Global Minimum Capital Standards. Available at http://www.bis.org/press/
p100912.pdf.
Batjargal, Bat, and Mannie Liu. 2004. Entrepreneurs Access to Private Equity in China: The Role
of Social Capital. Organization Science 15:2, 159172.
Benge, Vicki A. 2014. What Is an RMB Fund? Available at http://www.ehow.com/info_8644375_
rmb-fund.html#ixzz2ybnjmjZd.
Bottazzi, Laura, Marco Da Rin, and Thomas Hellmann. 2008. Who Are the Active Investors?
Evidence from Venture Capital. Journal of Financial Economics 89:3, 488512.
Bruton, Garry, and David Ahlstrom. 2003. An Institutional View of Chinas Venture Capital
Industry: Explaining the Differences between China and the West. Journal of Business Ventur-
ing 18:2, 233259.
Bruton, Garry, Maneksh Dattani, Michael Fung, Clement Chow, and David Ahlstrom. 1999. Pri-
vate Equity in China: Differences and Similarities with the Western Model. Journal of Private
Equity 2:2, 713.
Caselli, Stefano, Stefano Gatti, and Francesco Perrini. 2008. Are Venture Capitalists a Catalyst for
Innovation? European Financial Management 15:1, 92111.
Davis, Steven J., John C. Haltiwanger, Kyle Handley, Ron S. Jarmin, Josh Lerner, and Javier Miranda.
2013. Private Equity, Jobs and Productivity. Working Paper. National Bureau of Economics,
19458.
Davis, Steven J., John C. Haltiwanger, Ron S. Jarmin, Josh Lerner, and Javier Miranda. 2011. Private
Equity and Employment. NBER Working Paper no. 17399.
Delevingne, Lawrence. 2014. PE Firms Swap BRICs for Southeast Asia, Africa. Available at http://
www.cnbc.com/id/101576057.
Deloitte Center for Financial Services. 2012. 2013 Private Equity Fund Outlook: In Search of Firm
Footing. Available at http://www2.deloitte.com/content/dam/Deloitte/global/Documents/
Financial-Services/dttl-fsi-US-FSI-2013PEOutlook-121912.pdf .
Deng, Chao. 2013. For Private Equity, Raising Yuan Loses Its Shine. Available at http://blogs.wsj.
com/moneybeat/2013/04/22/for-private-equity-raising-yuan-loses-its-shine/.
Dezember, Ryan. 2013. Private Equity Enjoys a Record Year. Available at http://online.wsj.com/
news/articles/SB10001424052702304361604579290793412549248.
Dunkley, Dan. 2014. The Challenges Facing Private Equity. Available at http:/blogs.wsj.com/
privateequity/2014/02/20/the-challenges-facing-private-equity.
Emerging Markets Private Equity Association. 2014a. Key Elements of Legal and Tax Regimes
Optimal for the Development of Private Equity. Available at http://empea.org/resources/
empea-guidelines/.
Emerging Market Private Equity Association. 2014b. Ten Notable EM PE Investment Trends in
2013. Available at http://www.empea.org/_files/listing_pages/FINAL_-_Press_Release_-_
Full-Year_2013_Statistics.pdf.
Engel, Dirk, and Max Keilbach. 2007. Firm Level Implications of Early Stage Venture Capital In-
vestment: An Empirical Investigation. Journal of Empirical Finance 14:2, 150167.
Th e F u t u re of P riv at e E qu it y 521

European Private Equity and Venture Capital Association. 2014. About Us: Who We Are. Availa-
ble at http://empea.org/_files/listing_pages/EMPEAGuidelines_v1_web.pdf.
European Securities and Markets Authority. 2013. Final Report: Guidelines on Key Concepts of
the AIFMD. Available at http://www.esma.europa.eu/system/files/2013-2611_guidelines_
on_key_concepts_of_the_aifmd_-_en.pdf.
Fenn, George W., and Nellie Liang. 1998. New Resources and New Ideas: Private Equity for Small
Businesses. Journal of Banking and Finance 22: 68, 10771084.
Garten, Jeffrey. 2013. Whats the Future of Private Equity? Interview with David Rubenstein.
Available at http://insights.som.yale.edu/insights/what%E2%80%99s-future-private-equity.
Inside the Firm. 2014. The Impact of Crowdfunding on Venture Capital. Available at http://www.
jobsearchdigest.com/insidethefirm/3792/the-impact-of-crowdfunding-on-venture-capital/.
International Association of Risk and Compliance Professionals. 2014. The Volcker Rule. Availa-
ble at http://www.volcker-rule.us/.
Kaplan Steven N., Berk A. Sensoy, and Per Strmberg. 2008. Should Investors Bet on the Jockey or
the Horse? Evidence from the Evolution of Firms from Early Business Plans to Public Compa-
nies. Journal of Finance 64:1, 75115.
Kickstarter Statistics. 2014. Statistics. Available at https://www.kickstarter.com/help/
stats?ref=help_nav.
Krainer, Robert E. 2012. Regulating Wall Street: The Dodd-Frank Act and the New Architecture of
Global Finance: A Review. Journal of Financial Stability 8:2, 121133.
Kreutzer, Laura. 2012. Putting Dodd-Frank to the Test: Is Private Equity Ready? Available at
http://blogs.wsj.com/privateequity/2012/08/28/putting-dodd-frank-to-the-test-is-private-
equity-ready/.
Latin American Private Equity and Venture Capital Association. 2014. Highlights from 2014 In-
dustry Data and Analysis Report. Available at http://lavca.org/2014/02/20/lavca-releases-
highlights-from-2014-industry-data-analysis/.
Mass, Allison. 2013. Trends in Investment Banking: Private Equity. Available at http://www.
goldmansachs.com/our-thinking/trends-in-our-business/investment-banking/mass/.
Metrick, Andrew, and Ayako Yasuda. 2011. Venture Capital and the Finance of Innovation. Hoboken,
NJ: John Wiley and Sons.
Meyer, Thomas. 2006. Private Equity, Spice for European Economies. Journal of Financial Transfor-
mation 18, 6169.
Mollica, Marcos, and Luigi Zingales. 2007. The Impact of Venture Capital on Innovation and the
Creation of New Businesses. Working Paper, University of Chicago.
Mollick, Ethan. 2014. The Dynamics of Crowdfunding: An Exploratory Study. Journal of Business
Venturing 29:1, 116.
Ogg, John C. 2013. The Hottest IPOs of 2013. Available at http://finance.yahoo.com/news/
hottest-ipos-2013-101858504.html.
Popov, Alexander, and Peter Roosenboom. 2009. Does Private Equity Investment Spur Innova-
tion? Evidence from Europe. European Central Bank Working Paper Number 1063.
Preqin. 2013a. 2014 Preqin Sovereign Wealth Fund Review. Available at https://www.preqin.
com/item/2014-preqin-sovereign-wealth-fund-review/1/7594.
Preqin. 2013b. Private Equity Assets under Management as at December 2012September 2013.
Available at https://www.preqin.com/blog/101/7508/private-equity-aum.
Pries, Gerhard, and Vivina Berla. 2012. Private Equity in Emerging Markets: Exits from Aid, Steps
Towards Independence. Available at http://www.impactassets.org/files/ImpactAssets_
IssueBriefs_6.pdf.
privateequity.com. 2013. Private Equity Investor Allocations 2013. Available at http://private-
equity.com/tag/private-equity-investor-allocation/.
Private Equity Growth Capital Council. 2011 Issues: The Dodd-Frank Act. Available at http://
www.pegcc.org/issues/the-dodd-frank-act-summary/sec-form-pf/.
Private Equity Growth Equity Council. 2014. About the PEGCC. Available at http://www.pegcc.
org/about/.
522 t r e n d s i n p r i vat e e q u i t y

PR Newswire. 2013. BNY Mellon Announces Automated Waterfall Process for Private Equity
Fund Administration. Available at http://finance.yahoo.com/news/bny-mellon-announces-
automated-waterfall-113500531.html;_ylt=A0LEVvFU7lJT9FkAnwMPxQt.;_ylu=X3oDM
TBybnV2cXQwBHNlYwNzcgRwb3MDMgRjb2xvA2JmMQR2dGlkAw--.
Probitas Partners. 2012. Private Equity Investor Trends for 2013 Survey. Available at http://www.
probitaspartners.com/pdfs/probitas_private_equity_survey_trends2013.pdf.
Quinlivan, Steve. 2012. JOBS Act Effect on Private Equity and Hedge Funds. Available at http://
dodd-frank.com/jobs-act-effect-on-private-equity-and-hedge-funds/.
Realdeals. 2011. Basel III Claims another Private Equity Scalp. Available at http://realdeals.eu.
com/article/17394.
Schuetze, Arno. 2014. Private Equity Poised to Pounce on Southern Europe. Available at http://uk.
reuters.com/article/2014/02/25/uk-privateequity-conference-idUKBREA1O1HW20140225.
Securities and Exchange Commission. 2011a. SEC Adopts Dodd-Frank Amendments to Invest-
ment Advisers Act. Available at http://www.sec.gov/news/press/2011/2011-2133.htm.
Securities and Exchange Commission. 2011b. Form PF, Paper Version. Available at http://www.
sec.gov/about/forms/formpf.pdf.
Securities and Exchange Commission. 2013. Securities and Exchange Commission Proposed Rules
on Crowdfunding. Available at http://www.sec.gov/rules/proposed/2013/33-9470.pdf.
Shah, Sumeet. 2013. The Past, Present, and Hopeful Future of Private Equity. Available at http://
www.firmex.com/blog/the-past-present-and-hopeful-future-of-private-equity/.
Spangler, Timothy. 2013. Private Equity, Hedge Funds Are Ready for Their Close-Up. Available at
http://www.forbes.com/sites/timothyspangler/2013/08/31/private-equity-hedge-funds-
are-ready-for-their-close-up/.
Spector, Mike. 2014. U.S. Private Equity Tax Change Doubtful This Year, Says Carlyle Co-Founder.
Available at http://online.wsj.com/news/articles/SB1000142405270230380130457940648
4186743754.
Strmberg, Per. 2008. The New Demography of Private Equity. SIFR, in The Global Economic
Impact of Private Equity Report 2008. Working Paper, World Economic Forum.
Strmberg, Per. 2009. The Economic and Social Impact of Private Equity in Europe: Summary of Re-
search Findings. Available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1429322.
Tan, Gillian. 2014. Banks Sit Out Riskier Deals: Regulatory Pressure Push Some Lenders to Let
Lucrative Deals Go. Available at http://online.wsj.com/news/articles/SB100014240527023
04302704579334820201530010.
The Tax Policy Center. 2012. Business Taxation: What Is Carried Interest and How Should It Be
Taxed. Tax Policy Briefing Book. Available at http://www.taxpolicycenter.org/briefing-book/
key-elements/business/carried-interest.cfm.
Discussion Questions and Answers

Chapter 2 Economics of Private Equity


1. Explain the nature and functions of PE firms.
PE firms are financial intermediaries that raise money from high net worth indi-
viduals, pension funds, endowments, and other financial institutions to invest and
acquire equity in private companies, often taking the companies private if they are
initially public. PE firms are financial buyers and not strategic buyers. They often
take on high levels of debt to buy out companies. This process is commonly called a
leveraged buyout (LBO).
2. Explain the difference between GPs and LPs and how they are compensated.
The managers of a PE firm are known as the GPs whereas the investors who
commit capital in a PE fund are called LPs. GPs are compensated through a man-
agement fee, which is typically 2 percent of the assets under management (AUM),
and a performance fee, which is usually 20 percent after crossing a certain hurdle
rate. The LPs receive the rest of the returns net of the GPs fees.
3. Identify the criticisms and benefits of PE.
Some criticize PE for eliminating jobs, increasing the likelihood of bankruptcy
for investee companies, exploiting the tax code because the carried interest is differ-
entially taxed, stripping the investee companies of assets, and taking on high levels
of debt to award dividends to PE investors. Despite some misconceptions about PE
that are prevalent in the popular media, academic studies document several benefits.
One of the benefits is reducing agency problems in investee companies by concen-
trating ownership and bringing in new capital, expertise, and better corporate gov-
ernance practices. Studies provide strong evidence that PE firms do not destroy jobs
but create value for the investee companies.
4. Discuss how PE firms improve corporate governance in their investee companies.
PE firms improve the corporate governance of their acquired companies through
monitoring and aligning incentives. PE firms actively track and review the perfor-
mance of management teams of their investee companies. To mitigate agency prob-
lems, PE firms incentivize management teams of the investee companies either by

523
524 d i s c u s s i o n q u e s t i o n s a n d a n s w e r s

making them partial owners or by rewarding high performance. Such measures


help to align the interests of the managers of the investee firms with those of the GPs
and LPs.
5. Discuss how PE firms create value in their portfolio companies.
PE firms provide their expertise and advice on the operational, financial, and stra-
tegic decisions of their portfolio companies. The portfolio companies also benefit
from the PE firms prior experience and relationships in that particular industry. PE
firms sometimes bring in a new management team to handle a firms operations.
They use tools such as monitoring and incentive alignment to ensure value creation
in their portfolio companies. PE firms also create value through a successful exit
such as a reverse LBO.

Chapter 3 Private Equity in the United States


and Europe: Market and Regulatory Developments
1. Discuss how recent actions taken by central banks in the late 2000s and early 2010s
in the United States and Europe influenced PE activity.
To combat the financial crisis of 20072008 and sovereign debt crises, the Fed-
eral Reserve Bank and the European Central Bank adopted extraordinary loose
monetary policies aimed at reviving stock and credit markets. The actions of cen-
tral banks were successful in boosting stock prices and reviving credit markets. PE
sponsors took advantage of this opportunity to exit their investments through IPOs,
refinance the debt of portfolio companies at better rates, and add debt to companies
to fund their own payouts. Accommodating credit markets also allowed PE bidders
to finance their acquisitions at better rates.
2. Discuss general trends in the U.S. PE industry.
The U.S. PE industry has three general trends. One trend is the gradual transfor-
mation of the major PE firms into more broad-based asset management firms. Their
expansion into multiple lines of business such as real estate, investments in hedge
funds, credit extension, and financial advisory services is likely to continue. Expand-
ing their product offerings allows PE firms to deliver superior returns to their in-
vestors, diversify their source of income, and benefit from economies of scale. The
second trend involves more PE firms choosing to become or contemplate becoming
listed on a public stock exchange allowing them to raise permanent capital and their
founders to cash-out their stakes in the firms. The third trend is the continual in-
crease in secondary buyouts as a means of exiting PE investments. In a secondary
buyout, the PE investor sells holdings to another PE investor. Some criticize sec-
ondary buyouts for aggravating the agency costs between investors and managers.
3. Compare the regulatory approach of the United States and the European Union
toward the PE industry.
Regulating the PE industry after adopting the Dodd-Frank Act in 2010 mostly
concerned the potential contribution of the PE industry to systemic risk. Although
the industry is unlikely to be a source of systemic risk, the Dodd-Frank Act con-
tains rather limited disclosure requirements for assessing systemic risk that are not
d i s c u s s i o n q u e s t i o n s a n d a n s w e r s 525

expected to burden PE firms with insurmountable compliance costs. Furthermore,


a PE firm or fund is unlikely to be designated as a systemically important financial
institution. Besides, the Volcker Rule is likely to have a positive effect on the PE in-
dustry. Stand-alone PE funds will no longer face unfair competition from PE funds
belonging to banking institutions operating under explicit and implicit government
guarantees. In contrast, the AIFM Directive that went into effect in 2011 contains
provisions aimed at tackling the systemic risk of the PE fund industry and protect-
ing investors in PE funds. The AIFM Directive requires fund managers to comply
with burdensome and complex requirements, despite the unlikely possibility that
the PE industry is a source of systemic risk. Also, investors in PE funds are sophis-
ticated market players who can protect themselves and enter mutually valuable ar-
rangements with PE firms. Overall, the AIFM Directive is expected to substantially
increase compliance costs for PE firms operating in Europe. The restrictions im-
posed by the Directive and targeting asset stripping should have a profound impact
on deal structuring and exits.
4. Discuss the future of PE in the United States and Europe.
The financial crisis of 20072008 and the sovereign debt crises that followed
caused a profound transformation of the PE industry. The revival of PE activity is
threatened by adverse market regulatory developments in the United States and
Europe. The effects of these developments are likely to be adverse in Europe because
the European banking sector is still recovering from the sovereign debt crisis with a
severely curtailed ability to finance LBOs. Furthermore, the growth of the European
economy is likely to remain sluggish for the coming years. A particularly interven-
tionist and burdensome regulatory approach toward the PE industry complements
these adverse market developments. By contrast, U.S. regulators recognizing the
crucial role of PE in spurring growth and innovation decided to adopt a measured
approach toward the risks posed by the PE industry. U.S. PE firms are capitalizing on
the growth of the U.S. real economy, which is showing signs of a strong revival. How-
ever, as central banks around the world exit the extraordinary measures adopted
after the crises, PE investors will have to adapt to tighter capital markets and a more
modest increase in stock prices.

Chapter 4 Venture Capital in Europe: Myths


and Reality
1. Identify and discuss the main features of the VC industry in Europe.
The European VC industry is the second-largest venture hub for fundraising in
the world. It is young and the median starting date of European VC firms is 1998.
Government agencies are the major players and provide 40 percent of funds raised
in 2012. The community of European VC investors is stable but this stability has
both positive and negative aspects. On the positive side, European VC investors have
gained experience and built strong relationships among themselves. On the negative
side, European VC investors seem unreceptive to new players and are not prone to
changing their habits. A class of serial entrepreneurs has emerged in Europe.
526 d i s c u s s i o n q u e s t i o n s a n d a n s w e r s

2. Discuss whether the American and European VC practices have converged


European VC investors have seemingly adopted some practices of their U.S.
peers. This change may partially be due to the fact that about one-third of European
VCs have some work experience in the United States. However, some practices of
European VC investors still differ from those of their U.S. colleagues. For example,
European VCs resort less often to complex instruments of financing than their U.S.
counterparts. Also, the percentage of syndicated investments remains less impor-
tant in Europe than for the United States. Furthermore, European VCs replace the
entrepreneurs at the head of the company less often than U.S. VCs. Finally, U.S. VCs
rapidly change the destination of investments to quickly capture the best investment
opportunities. European VCs are less mobile and seem tied to industries deemed
strategic for European governments.
3. Indicate two approaches for analyzing the performance of PE funds.
PE investing broadly refers to investing in securities through a negotiated process.
PE investments are in unlisted companies, but some PE funds combine stocks of
these firms with other assets such as monetary investments, shares, and bonds. Two
approaches are available for analyzing the performance of PE funds: (1) examine
their performance for each investment and (2) examine the funds as a whole.
4. Explain why returns of PE funds have a J-curve effect.
The J-curve effect is a common phenomenon in PE funds characterizing the re-
turns patterns over the life of the fund. It shows the tendency of these funds to de-
liver both negative returns and cash flows in the early years and investment gains
and positive cash flows later in the investment funds life as the portfolio companies
mature and are gradually exited.

Chapter 5 Leveraged Buyouts


1. Discuss the role of leverage in an LBO.
In an LBO, leverage provides an additional funding source for buyout funds to
purchase larger companies and enables more portfolio companies to boost the re-
turns for the funds equity investors. Funding an acquisition of a portfolio company
with debt has several valuable effects for equity returns. Interest expenses on debt
are tax-deductible, which increases the value of the acquired portfolio firm. Further,
the equity stake in the company increases due to paying down debt using the port-
folio companys own cash. This effect, often compared to financing a house using a
mortgage, boosts equity value for the investors if the company can repay most of the
debt during the investment phase.
2. Identify typical restructuring mechanisms of LBO funds in their portfolio compa-
nies and discuss why some restructuring mechanisms may be detrimental to the
portfolio companies.
LBOs have four distinct groups of restructuring activities: (1) financial engineer-
ing, (2) operational engineering, (3) governance intervention, and (4) manage-
ment monitoring.

Financial engineering describes recapitalizing the portfolio company using high


leverage. Additionally, buyout funds optimize the financial accounting of the
d i s c u s s i o n q u e s t i o n s a n d a n s w e r s 527

portfolio companies through earnings management in order to lower taxable


income or artificially inflate the accounting income before an exit.
Operational engineering describes the process in which the portfolio compa-
nys operational business is restructured to improve efficiency. The goal is to ei-
ther increase revenues or lower costs in order to create higher free cash flows.
Frequent tools to achieve this goal are engaging in layoffs, stopping the produc-
tion of unprofitable products or closing divisions, and strengthening the profit-
able businesses.
Governance intervention and management monitoring address the reduction
of agency costs in portfolio companies. By either actively changing a companys
governance structure or monitoring its management, agency costs are lowered
to increase firm value. A buyout fund usually aligns the interests of equity inves-
tors and the companys management by giving ownership stakes to the manage-
ment. Also, managers introduce strict monitoring mechanisms, such as monthly
reporting systems, to keep close track of the companys operations.

Some restructuring activities are detrimental to portfolio companies because


they are not aimed at increasing economic value in the company, but to free up cash
flows, which can be paid out to the equity investors. For example, dividend recapi-
talizations put a financial burden on the portfolio company in the form of excessive
leverage.
3. Define dry powder and discuss how it affects a funds investment decisions.
Dry powder is the fraction of committed capital that a buyout fund has not yet
used for buyout investments. Having too much dry powder might adversely affect a
fund managers reputation. If a GP cannot invest the entire committed capital, po-
tential future investors might regard this as a negative signal for the funds invest-
ment prowess or portfolio company selection capabilities. Dry powder can also be a
sign of investment pressure. As the buyout fund ages, investors expect a fund man-
ager to invest the committed capital. This expectation can lead LBOs to use less debt
and more equity in structuring their financing because a fund managers goal shifts
toward spending the funds capital rather than selecting the best target companies.
High levels of dry powder can also lead to an increased amount of secondary buy-
outs because of the availability of such transactions. After all, some other buyout
fund might be looking to exit.
4. Discuss the importance of maintaining a relationship with banks for a buyout fund.
Leverage is an integral part of LBOs because it helps to increase the return on
equity for buyout investors. Debt also constitutes a device that helps to increase the
total amount of investment capital, which allows larger target companies to be ac-
quired. Maintaining a good and ongoing relationship with banks reduces agency
costs because a bank learns more about a specific GP in business practice, reliability,
and risk profile. Additionally, banks are willing to further reduce the cost of debt
if the potential exists for cross-selling banking products. As buyout funds acquire
many companies over time, banks might value the opportunity to provide other
services to portfolio companies even after the exit of the buyout fund. Banks are
also the main conduit to syndicated loan markets, which are an important source of
debt capital especially in large-scale LBOs. Every syndicated loan has a lead arranger
that usually is a bank.
528 d i s c u s s i o n q u e s t i o n s a n d a n s w e r s

Chapter 6 Mezzanine Capital and Commercial


Real Estate
1. Identify the hallmark of mezzanine investment and discuss some other important
characteristics.
The hallmark of any mezzanine investment is its inferiority to at least one class of
senior debt and its superiority to common equity. Other important characteris-
tics include a lack of any direct security interest in the investment real estate itself,
subordination to all of the real estate owners debt whether secured, and a position
in the capital stack supporting the real estate investment that is more consistent
with an equity risk profile than debt.
2. List the factors leading to the upsurge in mezzanine investment in real estate during
the 20-year period preceding 2008 and identify which still apply today.
The primary factors driving the surge in mezzanine real estate investment be-
tween 1988 and 2008 were the constrictions in numbers and underwriting liberality
of conventional real estate lenders resulting from the S&L crisis, the advent of secu-
ritization in mortgage-debt CRE financing, and the attraction to mezzanine debt
more conservative, debt-minded investors. Mortgage debt securitization has now
emerged from its profound slump in 20082012 and is approaching earlier levels,
but mortgage lenders and rating agencies remain cautious in terms of loan-to-value
and other underwriting criteria. The potential demand for mezzanine financing thus
remains great. What has probably changed most is the size and nature of the pop-
ulation of potential mezzanine investors. Uncertainty surrounds whether more en-
trepreneurial types of investors, such as hedge funds, will be sufficient to replace
departed investor classes, such as foreign banks, and whether the new investors will
be content with the CMBS-modeled forms of mezzanine investment.
3. Compared to mortgage debt, identify the primary risks inherent in mezzanine in-
vestments in real estate and discuss how they differ between preferred equity or
mezzanine loans.
Aside from B-piece loans and the now largely discarded second mortgages,
mezzanine investments lack the direct interest in the investment real estate that
secures mortgage loans and are generally subordinate to all obligations of the real
estate owner. Thus, whether in the form of debt or equity, mezzanine investments
in their risk profile resemble equity investments, subject to the operating risks of
the property owners business and not just the risk of a severe decline in real estate
values. These risks do not differ substantially between mezzanine debt other than B-
piece debt and preferred equity. After all, the security for the CMBS-type mezzanine
loan is just equity in the property owner.
4. Standardization of documentation appeared to be one feature of mezzanine loans
attracting investors such as foreign banks. Discuss whether such packaging of invest-
ment products should weigh heavily in the choice of mezzanine investment forms.
Relative uniformity in documenting mezzanine loans in CMBS-structured
financings may facilitate underwriting of a mezzanine debt investment by a
post-origination investor, but such an investor has little opportunity to fix defects
in the documents. The real advantages of standardization are to the CMBS loan
d i s c u s s i o n q u e s t i o n s a n d a n s w e r s 529

originators, which can avoid the complication of having mezzanine investors pres-
ent at creating their loans. From the investors perspective, standardization means
the documentation does not benefit from a thorough vetting by an eventual risk-
taker and does not necessarily fit the investors particular needs. For small invest-
ments, supposed cost-savings could outweigh the added risks of flawed documents;
for larger ones, such savings could come with a large future cost.
5. Discuss how the rating agencies shaped CRE mezzanine investment.
The agencies rating guidelines, by all but forbidding a mezzanine investor to have
a direct security interest in the investment real estate or the ability to put the prop-
erty owner into bankruptcy, have shaped the essentially equity-like nature of mez-
zanine debt. The rating agencies also have dictated the terms of the intercreditor
agreement between the mortgage and mezzanine lender, further hamstringing the
latters remedies when its loan is in default.
6. Given that rating agency guidelines favor preferred equity over mezzanine loans in
rating CMBS debt, explain why CMBS lenders so strongly favored mezzanine debt
as the mezzanine level in their structured financings.
Even with the severe limits that the standard intercreditor agreement places on the
ability of a mezzanine lender to turn troublesome, the rating agencies favor preferred
equity when the borrower requires mezzanine financing. The strong preference that
CMBS originators have had for mezzanine debt, even at the cost of some demerits from
the agencies rating their loan pools, likely stemmed from the greater ease with which
mezzanine loans in the past at least could be sold in secondary investment markets.

Chapter 7 Distressed Debt Investments


1. Discuss how the passage of the Bankruptcy Act of 1978 set the stage for the growth
of distressed debt investing.
The Bankruptcy Act of 1978 provided a new legal regime to efficiently process
business reorganizations and to better define creditor rights. It also had the explicit
goal of rehabilitating debtors in a way that preserved the going concern value of
companies. When Drexel subsequently developed a market for high yield bonds,
this led to the creation of a new class of securities with substantially higher default
risk. The combination of riskier corporate capital structures combined with an effi-
cient bankruptcy process in which investors could confidently predict restructuring
outcomes fueled the growth of distressed debt investing.
2. Compare and contrast the main strategies used by distressed debt investors.
Distressed debt investors employ four main strategies: (1) distressed for control,
(2) loan-to-own, (3) special situations, and (4) turnaround investing. In the first
two strategies, the investors goal is to control the targets post-reorganization equity
by converting the targets debt securities into new, post-reorganization equity se-
curities. The difference between the two is that in distressed for control investors
acquire the targets debt securities through secondary market purchases of exist-
ing issues whereas loan-to-own investors originate a new loan directly to the target.
Special situation investors do not want to control the target; instead, they identify
misvalued debt securities in the targets capital structure and make a profit when a
530 d i s c u s s i o n q u e s t i o n s a n d a n s w e r s

catalyst causes those securities to become properly valued. Turnaround investors


make direct equity investments into the target and make the majority of their invest-
ment return by improving its operations.
3. Explain whether distressed investors can add alpha if their returns tend to depend
on favorable market environments,
In analyzing the returns of distressed investment funds, the starting date or vin-
tage of the fund is clearly an important determinant of that funds performance. That
said, normalizing for vintage, certain fund managers persistently outperform vin-
tage cohort mean performance suggesting that these managers do add alpha.
4. Explain why distressed debt investors are sometimes characterized as vulture
investors.
Hilary Rosenberg infamously characterized distressed investors as vulture inves-
tors in a book by that title. The analogy is that distressed investors dissect the capital
structures and pick over the carcass of distressed companies as a way to generate
investment returns.

Chapter 8 Valuing Private Equity


1. Explain the difference between simple-equity cash flow and complex-whole cash flow.
Simple-equity cash flow is the most basic but not the most economically useful
measure of a companys cash flow. It combines net income plus depreciation ex-
pense and represents a measure of cash flow available to a companys equity holders.
Complex-whole cash flow is often called free cash flow (FCF). FCF begins with
simple-equity cash flow and converts it to operating cash flow by adding back the
tax shield on interest expense. Next, the change in deferred taxes is added back, and
then capital expenditures and the change in net working capital are deducted. The
resulting FCF represents a more accurate, complete, and economically sound meas-
ure of the cash flows that a company generates.
2. Discuss how to use the DCF method to estimate the value of a PE firm.
Analysts project FCFs over a forecast period until the cash flows can be assumed
to stabilize at a steady state and can be projected beyond that point by applying a
constant growth rate. The forecast often begins with such items as the prior years
sales, margins, capital expenditures, and investment in net working capital. Consid-
ering these historical drivers of value and future expectations, analysts forecast the
drivers into the future over the forecast period. Analysts also estimate the appro-
priate weighted average cost of capital (WACC) based on the firms riskiness and
capital market conditions. They estimate a terminal value at the end of the forecast
period by applying a constant growth rate assumption to the cash flow at the end of
the forecast period. Finally, analysts estimate the firms enterprise value by taking the
present value of the FCFs, including the terminal value, discounted at the WACC.
3. Explain why multiples-based valuation and the APV method may be appropriate in
evaluating a PE firm.
Multiples-based valuation provides an estimate of a firms value using valu-
ation multiples such as EV-to-EBITDA for similar companies trading in the cur-
rent market, or for recent acquisitions of similar companies. Analysts apply these
d i s c u s s i o n q u e s t i o n s a n d a n s w e r s 531

multiples to the relevant base such as EBITDA of the target firm to estimate its value
in the current market. This method helps to ensure other calculations and assump-
tions are reasonable and aids companies when the inputs for DCF are difficult to
obtain such as in high-tech start-ups. APV is useful for highly leveraged transactions
in which the acquirer finances the acquisition by increasing the target firms financial
leverage and paying down this debt over time. The WACC used in a standard DCF
valuation assumes that the company keeps a constant capital structure, which is not
the case in a highly leveraged transaction. APV bypasses this assumption by valuing
the company as the present value of the operating cash flows discounted at an unlev-
eraged cost of equity, plus the present value of the interest tax shield from the debt
used to finance the transaction.
4. Discuss the means by which PE firms can increase the value of companies they
acquire.
PE firms succeed by understanding a firms current value and its potential value
after making operational and financing improvements. For example, the PE firm may
be able to change the target firms management thereby altering the targets strategy
and direction. The PE firm may also hold similar companies in its portfolio of invest-
ments that can be used to leverage new ideas for the target. These changes may result
in higher revenue growth, higher margins, more efficient rates of investment in capital
expenditures and net working capital, and higher overall growth rates for the target
firm, leading to an increase in value. The PE firm may also financially reengineer the
target firm by moving to a more highly levered capital structure, which lowers the
target firms WACC and raises its value. If the PE firm purchases a target company for
a reasonable premium over its current market price and then makes these changes to
increase the company value to more than the price paid, it can earn a profit on the
difference between the potential value of the company and the price paid.

Chapter 9 Cost of Capital of Private Equity


1. Discuss whether the cost of capital of PE differs from the standard paradigm estab-
lished by the modern finance theory.
Although computing the cost of capital of private and public equity is similar,
it should consider specific nuances related to PE. These include the differences in
perceived risk in an environment described by a certain opacity (e.g., data are pri-
vate and often difficult to obtain and opacity of financial statements is increased),
illiquidity, and the stale valuation problem. Such nuances typically increase the cost
of equity, affect the risk level of debt, and may lead to an increase in the cost of debt.
The main risk factors are the market premium, size, book-to-market, and liquidity.
Some adjustments may be required to address the stale valuation problem.
2. Identify which asset pricing models should be used to evaluate the cost of capital of PE.
The most common models used to evaluate the cost of capital for PE are the
standard asset pricing models developed in the financial literature: the CAPM and
the Fama-French three-factor model. Analysts rarely include the momentum factor
introduced by Carhart to value the cost of PE. However, recent literature suggests
including a liquidity factor.
532 d i s c u s s i o n q u e s t i o n s a n d a n s w e r s

3. Explain the main consequences of illiquidity or stale valuation on PE.


The illiquidity or stale valuation issue observed in PE returns is related to returns
being serially correlated. Serial correlation leads to underestimating risk and over-
estimating excess returns. Estimates are statistically biased and require correction.
4. Discuss how adjusting for liquidity premiums improves the valuation process.
Adding a fourth risk-factor related to a liquidity premium to the Fama-French
model captures an important characteristic of PE returns. An augmented Carhart
model including the liquidity premium results in the same conclusion. Because the
coefficient associated with the liquidity factor loading is generally statistically sig-
nificant, including the liquidity factor in the model can improve the computation of
the cost of capital of PE. As an example, for PE funds (as reported by Table 9.5) with
a mean of 0.79 percent for the liquidity premium (Table 9.1) and a factor loading of
0.65 (in the augmented Carhart model), the total liquidity premium to be added is
equal to 0.51 percent (0.65 x 0.79 percent).

Chapter 10 Liquidity Issues in Private Equity: Lessons


Learned in the PostFinancial Crisis
1. Identify and discuss the two types of liquidity risk investors face in PE.
The two types of liquidity risk for PE investments are market (asset) liquidity
risk and funding (cash flow) liquidity risk. For PE investment, market liquidity risk
refers to exit risk, which is the risk of being forced to sell an investment at a discount
or to be locked up in an investment longer than desired without being able to exit.
This risk is mainly due to the nature of PE funds that are mostly organized in the
form of limited partnerships with finite lives of typically 10 years. Partnership agree-
ments specify restrictions on transferability and LPs have no right to ask that sales
be made. PE is intended to be a long-term investment for buy-and-hold investors
and its returns originate from capital gains in exit transactions. This exit risk is an
important part of liquidity risk. Another liquidity risk called funding liquidity risk
is due to the institutional features of the PE industry. PE commitments are not fully
funded at the beginning of the initial investment. Upon launch, only a certain por-
tion of the committed capital is payable and the balance is drawn over time as the
GP performs capital calls when the PE firm decides to invest in selected companies.
This unpredictable timing of capital calls over the life of a PE investment creates
funding liquidity risk for the LPs.
2. Explain the term denominator effect for PE and the GFC and discuss its impact on
the PE secondary market.
The denominator effect is a way of describing the impact of the GFC on portfolios,
especially those of institutional investors such as pension funds or university endow-
ments. In general, these long-term investors plan their target asset allocation as a per-
centage of their total AUM. Due to the decrease in the values of other asset classes
such as public equity during the GFC, investors became superficially over-allocated
in alternative investments such as hedge funds or PE. Thus, while the denominator
(total portfolio) shrank, the relative size of the numerator (PE) grew. To rebalance
their portfolios and reestablish desirable ratios, investors tried to sell their PE fund
interests, which was difficult in such markets. As a result, the best option for LPs was
d i s c u s s i o n q u e s t i o n s a n d a n s w e r s 533

to sell their PE funds in the PE secondary market. This sales pressure led to a substan-
tial decrease in prices in the secondary market and to the deterioration of the effective
liquidity of the market. The deal volume also decreased and bid-ask spread increased.
3. Describe the impact of the GFC on PE GPs in terms of liquidity.
The GFC affected PE fund managers through a funding liquidity channel. During
the crisis, financial institutions saw a deterioration of their balance sheets and greater
exposure to severe potential losses. As a result, they tried to expand their cash buf-
fers and decreased their loans to borrowers, among which were PE firms. Therefore,
PE fund managers had difficulty refinancing their investments and had to liquidate
them or accept higher borrowing costs.
4. Explain the avenues through which PE LPs faced liquidity problems during the GFC.
Due to the decrease in the values of other asset classes such as public equity, in-
vestors became superficially over-allocated (the so-called denominator effect) in
alternative investments such as hedge funds or PE. Consequently, a large increase
occurred in the number of LPs trying to sell their PE fund interests. Because PE in-
vestments were illiquid, long-term investments; effectively exiting such investments
was difficult. Another problematic issue involving liquidity was the exclusive right
of the GPs to make capital calls and ask investors for more money. During the GFC,
distributions decreased but GPs continued to exercise their right to place capital
calls but at somewhat lower amounts. These capital calls placed further liquidity
pressures on LPs. To fulfill these calls, PE investors had to sell other assets in their
portfolios at a time when market values were low. The result was a deteriorating li-
quidity profile of investors portfolios.
5. Explain how the Dodd-Frank Act and AIFMD relate to PE liquidity risk.
The major legal provisions affecting PE are the Dodd-Frank Wall Street Reform
and Consumer Protection Act (Dodd-Frank Act) in the United States and the Alter-
native Investments Fund Managers Directive (AIFMD) in the European Union. The
objective of the Dodd-Frank Act is to control systemic risk through stricter rules and
standards for systemically important entities. As such, if the authorities consider a
PE fund to be systemically important, its manager will be subject to heightened stan-
dards including liquidity and leverage requirements. The AIFMD regulates manag-
ers of alternative investment funds. Liquidity risk is an important area of focus within
the Directive, reflecting the increase in liquidity risk following redemption requests
during the crisis. According to AIFMD, managers are required to adopt appropri-
ate liquidity management procedures for each alternative investment fund, except
for unleveraged closed-end funds. This Directive also requires managers to conduct
stress testing under normal and exceptional liquidity conditions and to show to the
authorities that proper liquidity management systems are in place.

Chapter 11 Private Equity Portfolio Management:


Challenges, Approaches, and Implementation
1. List and describe differences between PE investments and tradable assets such as
common stock and bonds.
Most common stock and bonds are traded frequently and cheaply on public
exchanges in a standardized format. PE investments are contractual obligations
534 d i s c u s s i o n q u e s t i o n s a n d a n s w e r s

between parties whose terms usually stay private and are rarely fully standardized,
and whose outcome is not consistently and fully disclosed. The investments are
evaluated only periodically or when a relevant event such as a third-party transac-
tion or business event happens. This valuation is an estimate of the present value of
the expected future cash flows, and one cannot be confident to be able to buy or sell
for that amount. Common stock and bonds have a last market price within a certain
range in which the investor should be able to buy or sell. As a result, PE investments
cannot be easily sold, cannot be sold under compulsion, and are typically held long
term. PE is a structurally illiquid asset class that institutional investors expect to
carry a compensatory risk premium. To benefit, they must have a liability profile
that allows them to lock in capital for a minimum of 10 years.
2. Explain why PE does not fit into the standard Markowitz portfolio optimization.
The Markowitz portfolio optimization is based on the existence of a market that
constantly, efficiently, and transparently prices its assets. The market prices can be
recorded, periodic returns computed, and its time series can be analyzed. The av-
erage return, the volatility of returns and correlations among the returns are then
computed. These historical characteristics can be inputted to forecast the future
average, volatility, and correlations among returns. Markowitz shows how to com-
pute the optimal proportion of different assets with different return, risk, and cor-
relations. PE investments do not have a market with constant, efficient, and trans-
parent pricing. Thus, studying time series of market prices is very difficult. Several
approaches have been used, but no one method is completely satisfactory. Funds
and PE assets that are buy-to-sell could be seen as cash flow assets. Such assets
usually cannot be traded profitably, create cash flows, and need to be sustained
through a timely provision of liquidity because the opportunity costs associated
with undrawn commitments require explicit or implicit over-commitment strate-
gies. Making tradable assets and cash-flow assets comparable involves either map-
ping market prices on a cash-flow model or using a cash-flow model to determine a
fair value, assuming that this would fetch the same price in the market. This process
poses complications for portfolio management approaches based on modern port-
folio theory (MPT), which relies on market prices and the ability to transact at any
point in time.
3. Describe two alternative frameworks to MPT.
Observations of investor behavior including portfolio composition show that
many investors, especially individuals, do not follow the Markowitz optimal port-
folio based on MPT. Economists have searched for reasons to understand the di-
vergence from Markowitz in efficient markets, and have proposed several explana-
tory frameworks of which behavioral finance and the adaptive market hypothesis
(AMH) are the most prominent. Behavioral finance rightly points out that humans
are not perfectly rational economic agents using complete and perfect information.
The human brain and its surrounding culture have developed many thinking short-
cuts, so-called heuristics, to survive in an uncertain world with imperfect informa-
tion. For example, humans are more concerned with losing than not gaining, which
could explain why many people first set aside emergency savings and then invest, in-
stead of having one portfolio. The AMH is based on evolutionary ideas. An investor,
d i s c u s s i o n q u e s t i o n s a n d a n s w e r s 535

such as a PE investor, needs to adapt his strategy to find niches in an inefficient but
evolving market. These two explanatory frameworks might be especially useful in
PE because PE is an uncertain market with imperfect information.

Chapter 12 The Role of Private Equity Firms in Initial


Public Offerings: The Case of Venture Capital Firms
1. Explain the common exit routes taken by VC firms and identify the most popular
exit route.
VC firms engage in five major exit routes.

An initial public offering (IPO) is the process by which a private company goes
public by selling shares in the financial market for the first time.
A trade sale occurs when VCs sell their investee companies in whole or in part to
another company.
A management buy-out (MBO) occurs when the existing management team
with the help of external funding buys a firm.
A secondary sale is a form of refinancing.
A liquidation is a write-off.

Traditionally, IPOs were the most popular exit route due to their large returns.
However, between 2006 and 2013, trade sales became the most common exit route
for PE and VC firms.
2. Discuss the role of VCs in IPO underpricing.
The role of VCs in IPO underpricing is still being debated. Some researchers doc-
ument that VCs prefer less underpricing because it alleviates information asymme-
try and leaves less money on the table at the time of IPOs. Others find that VCs
increase underpricing. Recent international survey evidence by Ernst & Young fails
to offer a definitive answer on this topic.
3. Describe the monitoring role of VCs in their investee companies.
The necessity of monitoring investee companies arises because entrepreneurs as
firm insiders often have more information than the VCs. Such information asymme-
try results in agency problems such as adverse selection and moral hazard. To miti-
gate such problems, the VCs, who act as principals in such a situation, continuously
monitor the behavior and performance of the entrepreneurs, who are the agents.
Studies provide both indirect and direct evidence of the monitoring role of VCs in
investee companies.
4. Discuss the certification and grandstanding hypotheses in relation to IPO under
pricing.
The certification hypothesis asserts that VCs certify the true value of a firm that
intends to go public. Market participants generally have more confidence in reputa-
ble VCs and tend to accept the offer price set by a venture capitalist. This action leads
to a lower IPO underpricing. Early empirical studies find support for this argument
536 d i s c u s s i o n q u e s t i o n s a n d a n s w e r s

by showing that the underpricing level, as denoted by the first-day returns, of VC-
backed IPOs is significantly lower than those of non-VC-backed IPOs. The grand-
standing hypothesis contends that to raise more funds in the following periods, VCs
are willing to accept higher underpricing to make investors more interested in the
IPOs. Such a strategy increases the probability of a successful IPO and leaves inves-
tors with a good impression. Some believe that IPO firms benefit from this approach
because they can set a higher price for their shares during secondary equity offerings
(SEOs). More recent empirical studies find support for the grandstanding hypoth-
esis, especially for VC firms with shorter track records or lower reputation levels.
5. Identify several unanswered research questions about the role of PE and VC firms in
IPOs.
Some unresolved issues still surround the role of VCs in IPOs. These unanswered
questions include: Why do VCs choose an IPO as an exit route versus other exit
routes? Do the firms that the VCs choose to take public have certain characteristics
that systematically differ from those for which VCs choose a trade sale or a second-
ary sale? What is the effect of VC exits once the companies become public? If VCs
play an important role during IPOs, their eventual exits should also affect newly
listed firm performance. Can newly listed firms maintain good operating perfor-
mance and raise funds through SEOs effectively in the post-VC era?
Another important issue needing further examination is the dynamics between
the venture capitalist and the entrepreneur in terms of the principalagent roles in
managing the firm. The literature has extensively explored the model in which the
venture capitalist is considered to be the principal and the entrepreneur the agent.
Models contending that in such a set-up the venture capitalist acts as an agent for the
entrepreneur have been comparatively less examined and needs further research.

Chapter 13 Exit Strategies in Private Equity


1. Describe the traditional exit routes for a PE fund.
Three traditional exit routes for a PE fund are a trade sale, an IPO, and a second-
ary buyout. A trade sale occurs when a strategic acquirer buys one of the PE firms
companies. The strategic buyer is usually a non-PE firm, and the acquisition is in
the buyers strategic interest whether the motive is for market growth, patents, inno-
vative products, or synergies. The acquirer can be the portfolio company itself, re-
purchasing its own shares from the PE firm or fund. An IPO is a method in which a
companys shares are listed on the stock market for the first time. This strategy offers
a PE firm a way to exit by selling its shares of a company within its portfolio. A sec-
ondary buyout occurs when the PE fund sells a portfolio company to another finan-
cial sponsor in a buyout transaction. The transaction may or may not be leveraged.
2. Discuss the advantages and disadvantages of a trade sale as an exit strategy.
A trade sale has both advantages and disadvantages. The buyer expects a greater
competitive advantage and market share in its respective industry, intending to hold
the acquisition over the long term. The buyer often agrees to pay the value of strate-
gic options embedded in the target price because it expects a higher future operating
cash flow from the target company. Therefore, an advantage of a trade sale is that it
d i s c u s s i o n q u e s t i o n s a n d a n s w e r s 537

usually commands the highest sale price, allowing the PE vendor immediate liquid-
ity. Additionally, a trade sale is an easy way to exit because it implies a business-
to-business negotiation, and a small transaction cost if low information asymmetry
exists between the vendor and the acquirer. Despite its benefits, a trade sale has two
potential disadvantages. First, a companys management may resist the trade due to
the risk of its own replacement. Second, the portfolio company risks exposing con-
fidential business information during the negotiation process.
3. List the three main reasons that a firm initiates an IPO.
Several reasons explain why a firm initiates an IPO. One reason is to gain atten-
tion from existing and potential third parties (e.g., customers to increase sales, sup-
pliers to get better buying conditions, and skilled people to employ them). Other
reasons are to raise diversified funds to finance the firms development, gain a first
movers advantage, and allow existing shareholders to sell their holdings in order to
receive a return. The founders can diversify their portfolio and the other investors
such as the PE and capital investment funds can exit their investment.
4. Identify the main geographic areas where PE backed-IPO deals took place between
January 2005 and June 2014.
Between January 2005 and June 2014, the Americas and the EMEA region led
the PE-backed IPO market: Western Europe and North America account for 44.9
and 34.3 percent, respectively, of the value of all IPOs. These two regions are tra-
ditionally the main areas where PE-backed IPOs take place but Central Asia and
China in particular should represent a bigger proportion in the future.
5. List the main reasons VCs want to exit an investment
VCs have three major motives for exiting an investment. First, they consider the
exit as a way to reward their LPs after a few years. Second, VCs cannot afford to
finance portfolio companies in their late stage because of their limited financial re-
sources. Third, VCs do not have the competencies to manage these maturing firms.
4. Discuss the pros and cons of bottom-up and top-down portfolio constructions.
Deciding on a bottom-up or top-down approach is not really an active choice in-
vestors have because they have to cater to their expertise and constraints. However,
they need to find the most appropriate approach for them. The bottom-up approach
focuses on screening all investment opportunities with quality being the overriding
criterion, irrespective of portfolio considerations such as sector or geographical di-
versification, which are assumed to have a lesser impact. The positive aspects of this
approach include a focus on best quality investments and managers and the ability
to out-perform if superior skills exist. On the negative side, the bottom-up approach
can lead to an unbalanced portfolio and more risk.
The top-down approach is based on a construction of the overall portfolio
by determining allocation ranges and then searching for opportunities that fit
these allocations. Its advantages include automatic diversification, never being
completely wrong, and the assumption that selection skill is less critical. The
disadvantages of the top-down approach include the possibility of being unable
to find high-quality investments in each sub-class and most return coming from
company-specific risk (i.e., the success depends mostly on the companys unique
qualities) and not sub-class. Apart from the questions associated with determin-
ing allocation weights, the major shortcoming of a top-down approach is that a
538 d i s c u s s i o n q u e s t i o n s a n d a n s w e r s

strict adherence to these ranges is impossible in reality. When investing in PE


funds, finding and accessing a sufficient number of superior managers to fill in
each pre-determined sub-class allocation can be difficult. Indeed, often only one
or two superior fund managers operate in a particular sector, and they raise capi-
tal only every three or four years. 5. List and explain measures that can be used to
monitor a portfolios evolution.
The portfolio manager should establish measures to monitor the portfolios past
performance, current status, and possible future evolution. These measures include
the following:

Cash-flow measures: Total drawdowns, total paid-out, total amount invested,


and net asset value as an estimate of future cash flows.
Performance measures: The internal rate of return (IRR), multiples, and residual
value divided by paid-in fund investment.
Diversification measures: Total invested funds in sectors, stages, geographic ar-
eas, and individual teams.
Target measures: Cash needed from drawdowns within the next six months,
cash flow at risk, and the percentage of drawdown paid out.
Future measures: Implementing a cash-flow model leading to a distribution of
future measures (e.g., the distribution of the final IRR or multiple).

Chapter 14 Private Equity Returns: Cross-Country


Evidence
1. Explain why the PE return results reported in this chapter should be viewed with
some skepticism.
Academic research often uses two primary data sources: VentureXpert and Ven-
tureOne. Both data sources rely on self-reported data. Kaplan, Sensoy, and Strm-
berg (2002) detail many issues associated with these data sources. Others (Co-
chrane 2005; Phalippou and Gottschalg 2009) have identified oversampling firms
from a specific region, as well as missing stages of financing, which biases returns
calculated for other financing stages. Potentially the biggest issue is survivorship
bias. Because PE and VC funds invest over an extended period, a snapshot at any
given time includes exits and ongoing investments. Exits are presumably represen-
tative of successful investments, while continuing investments may eventually be
positive but could possibly be negative at the moment. Also, target companies are
private concerns. Thus, getting an accurate valuation measure is impossible, and any
returns based on a valuation may be positively skewed.
2. Explain whether investors should include PE in a diversified portfolio.
Based on an examination of PE equity indexes from their inception, arguing that
investors receive risk-adjusted portfolio performance from including PE in their
portfolio would be difficult. Given the short time horizon, basing a conclusion on
these data alone would be inappropriate. However, research suggests that, net of
fees, both VC and PE introduce negative alphas, especially after adjusting for the in-
herent riskiness associated with PE. Looking at the difference between average and
d i s c u s s i o n q u e s t i o n s a n d a n s w e r s 539

median returns, some winners appear, but most PE does not deliver on the promise,
except for the GPs.
3. Identify the factors that affect cross-sectional PE BO activity and identify any global
trends that have occurred between 2006 and 2013.
The primary factors affecting cross-sectional PE buyout activity are: (1) eco-
nomic activity, (2) well-functioning capital markets, (3) taxes, (4) corporate gov-
ernance and investor protection, (5) the social environment, and (6) the existence
of entrepreneurial opportunities. Each factor plays a part in shaping the size and
strength of a countrys PE market. From 2006 to 2010, almost 50 percent of all deals
occurred in North America in most years. In each year, Europe came in second, Asia
was in the third position, and the rest of the world followed. The deal data confirm
the value of the attractiveness indexs power to discern markets and countries where
investors will actually invest. Examining the proportion, rather than the level, of ac-
tivity across geographical regions over the period reveals that the proportions revert
in 2013 to about where they started in 2006 as shown in Figures 14.3 and 14.4. Of
total global BO activity, 77.0 percent occurred in North America in 2006 but by
2013 that value had declined to only 62.6 percent. In 2006, Europe created 26.7 per-
cent of the total value of global PE BO value, and by 2013 that value had increased
slightly to 27.0 percent. Asia ends 2013 higher than in 2006, as its total value contri-
bution to global PE BO activity increased from 2.7 to 7.0 percent.
4. Identify the two largest PE markets in the world and factors contributing to the size
of these markets.
The two largest PE markets are North America and Europe. While the United
States is the epicenter for PE investments, Europe including the United Kingdom
has grown into the second largest PE market. Kaplan and Strmberg (2008) note
that Western Europe represented a larger share of the PE transactions from 2000
to 2004 than the United States (48.9 vs. 43.7 percent). Within Europe, the United
Kingdom is the largest market because many firms are headquartered there and
the United Kingdom has deep capital markets. Sommer (2013) identifies 40,682
completed deals in Europe from 1990 to 2009 with the United Kingdom, France,
Germany, and the Netherlands dominating this list. This represents between 60 and
70 percent of PE deals in any given year. Notably, according to ImadEddine and
Schwienbacher (2011), 25 percent of the capital in European PE funds originated in
North America, specifically with those investors who also maintain local European
facilities such as a branch office for an insurance company.

Chapter 15 Benchmark Biases in Private


Equity Performance
1. Discuss the pros and cons of measuring PE performance using IRR, PME, and
RCR.
IRR measures the annual return and can be used to compare performance across
capital and investment horizons. One drawback is the implicit assumption that cash
flows can be reinvested at the IRR. IRR also does not reflect the length of the invest-
ment, thereby overvaluing quick flips. PME is calculated for the whole investment
540 d i s c u s s i o n q u e s t i o n s a n d a n s w e r s

horizon of the PE fund by comparing PE performance with the performance of an


index investment with equivalent cash flows (i.e., investments and disbursements).
Yet, PME does not consider the costs of holding liquidity for capital calls. The RCR
also takes liquidity costs into consideration assuming that the non-called part of the
committed capital is invested at the risk-free rate. A weakness of RCR is that the li-
quidity, assumed necessary to hold, might be exaggerated as investors can get access
to liquidity to meet calls by using credit lines or by selling assets.
2. Discuss the factors to consider when choosing a benchmark, the drawbacks of using
a public equity index as a benchmark, and how to adjust the index to compensate for
these shortcomings.
A benchmark should have a risk and return pattern close to that of the measured
asset. Because a public equity index is much more liquid than a PE fund, the alpha
of the fund should compensate for the liquidity premium. The index might have a
different beta than the measured asset. The performance of the benchmarks fund
should be adjusted for the beta difference.
3. Explain the difference between measuring performance based on return gross of
fees and net of fees and why each measure is used.
The difference between the two measures reflects the fees paid to the manag-
ers (e.g., the 2 percent management fee and the 20 percent performance fee of the
return exceeding a hurdle rate of 8 percent). The gross performance is relevant for
measuring the managers ability to create value. The net performance measures the
return to the investor and is the relevant measure for asset allocation decisions as it
can be directly compared to the return of other investments.
4. Explain the stale pricing problem and how to address it.
When the measured asset is valued less frequently than the benchmark, the price
of the measured asset becomes stale and the correlation between the asset and the
benchmark is underestimated. Thus, the result may overestimate the alpha value.
This problem can be addressed by including lagged correlating effects and correct-
ing beta. When lagged effects are included, higher correlation leads to a higher beta
value and a lower alpha value. The lower alpha value reflects that less return is ex-
plained by specific PE performance and a larger part is due to the development of
the benchmark such as the stock market.
5. Identify the most important types of selection bias and explain their effect if the
sample is not corrected for them.
The most important biases are due to differences in survivorship, vintage year,
fund size, update frequency, and skill of managers or investors. Non-correction for
survivorship bias leads to an overestimation of performance because dead funds
are excluded. PE funds should be compared with funds from the same or a nearby
vintage year to avoid business cycle mix. Selecting only large funds typically over-
estimates performance because large funds tend to outperform small funds. Fund
managers may stop reporting due to bad performance. Keeping valuation from non-
reporting funds unchanged could lead to overestimating performance. Finally, some
managers are better than others at finding high performing companies, and some
investors are better at spotting successful funds. Using a sample from just one man-
ager or investor likely provides an unrepresentative picture of the performance of PE
funds as an asset class.
d i s c u s s i o n q u e s t i o n s a n d a n s w e r s 541

Chapter 16 Return Persistence: Finding Top


Quartile Managers
1. Discuss the factors to consider when evaluating a fund based on its cash-on-cash
returns (multiples) or the IRR.
Multiples are not annualized as is common practice with most other return meas-
ures. They also ignore the time value of money because cash-on-cash multiples are
based solely on absolute numbers. The investment process for BO funds however
differs substantially from that associated with other asset classes where multiples
might be adequate return measures. The capital committed to the fund is invested
over several years and the funds subsequent distributions are irregular in both size
and frequency. Unlike open-end investment vehicles such as mutual funds, PE
funds are typically closed-end funds where the GP fully controls a fixed pool of cap-
ital with the associated investment process. Thus, the timing of investments and di-
vestments is a major part of the value generation of GPs, and needs to be considered
with regard to returns. Because cash-on-cash multiples do not reflect that cash-flow
timing, multiples are an inappropriate stand-alone return attribution methodology
for evaluating GP performance and can be misleading. Therefore, a value-weighted
return measure is needed that accounts for the timing of investments and divest-
ments by the fund manager. IRRs use annualized returns and incorporate the time
value of money. Despite also having shortcomings and restrictions, the IRR has
emerged as the industry standard.
2. Explain what the PME approach toward benchmarking PE returns seeks to
accomplish.
The PME is the ratio of the terminal wealth obtained when investing in a PE fund
and reinvesting intermediate cash flows in a given public market benchmark com-
pared to the terminal wealth obtained when investing the same amount of money in
the benchmark. It offers a simple solution to the benchmarking problem related to
time-weighted returns. The concept determines the opportunity cost of capital for
buyouts, creating a mimicking portfolio of PMEs such as the S&P 500 index for the
United States. These investments are designed to replicate the risk profile of buyouts
in terms of timing and systematic risk. PMEs can provide a complete performance
ranking of all available funds but do not consider financial leverage and systematic
risk.
3. Explain the relationship between a PE firms fundraising and the valuation levels it
reports for its PE investments. How does that behavior influence the search for top-
quartile managers?
To inform investors on the value of their investments, GPs self-report a quarterly
NAV that reflects the value of continuing investments (i.e., the investment portfolio
companies). Because portfolio companies do not trade publicly, measuring NAVs
objectively is difficult and strategic valuation decisions of the fund managers can
affect valuations. When GPs prepare for a new round of fundraising, they try to pres-
ent their preceding funds in the best light. Thus, GPs have an incentive to report
attractive valuations of their portfolio companies. Given that much leeway exists
when reporting NAVs, GPs are tempted to report inflated valuations. This behavior
542 d i s c u s s i o n q u e s t i o n s a n d a n s w e r s

artificially inflates return expectations for funds, thereby manipulating NAVs and di-
recting investors toward the wrong funds when making their investment decisions.
4. Identify the practical limitations of implementing an investment approach based on
top quartiles.
Selecting GPs based on past top-quartile performance has several limitations.
First, due to ambiguous performance criteria, some leeway exists in defining top-
quartile performance. This ambiguity explains why much more than one-quarter
of funds claim to be classified as top quartile. Further, a fund moves among quar-
tiles over its life. PE groups claiming top quartile with a fund often raise subsequent
funds before potential investors can accurately measure the previous funds perfor-
mance. Thus, for investors to act on any positive relationship between predecessor
and follow-on funds, they need to know a funds final performance after only a few
years of operation when the fund is still in its investment phase. Interim perfor-
mance is a weak indicator of eventual performance, which is itself a weak indicator
of future performance.
5. Explain why the worst performing BO funds over the long term are often the largest
funds by investment size.
Several reasons explain why the worst performing BO funds tend to be the largest
funds. Partners at successful, large firms might become less eager to take the risks to
get an outstanding return. Becoming more risk averse can lead to more modest re-
turns. Further, because a partners capacity to look after portfolio companies is lim-
ited, increasing the fund size might lead to less management attention and therefore
inferior performance. A refinement of areas of expertise and personnel additions/
changes due to increased fund size coupled with changing market dynamics can also
alter the strategy pursued by the GP as well as the environment in which the GP will
invest the funds.

Chapter 17 Private Equity Due Diligence


1. Define due diligence and discuss its importance in PE.
Due diligence is a measure of responsibility, diligence, and prudence that is gen-
erally expected from and exercised by any rational investor under different circum-
stances. It involves investigating the feasibility of a potential investment. Due dili-
gence is important to PE firms because it helps identify material facts about a target
company before entering a contract and prevents unnecessary harm to either party
involved in the transaction. Due diligence also helps to identify the risk and return
profile of the fund offer. Through due diligence, PE firms try to build value. Due
diligence helps GPs gain a better understanding of the companies which they are
considering for investment. A well-structured due diligence process is the initial step
for any contract and contains a complete analysis of the company profile, allowing
the investor to sort through the most promising funds in the market. Finally, it helps
to ensure that PE firms get what they expect from a transaction.
2. Explain the importance of commercial due diligence in PE.
Commercial due diligence is the process of extracting information about buyers
and suppliers of the respective target company. It also involves understanding the
d i s c u s s i o n q u e s t i o n s a n d a n s w e r s 543

factors that influence the performance of the respective company. The main factors
that can affect the business are the performance of the market, driving forces of the
industry in which the company operates, and regulatory reforms. Investors select PE
to gain profits. By ignoring relevant factors, they may experience a loss. For exam-
ple, if a particular industry in which the company operates is governed by the bar-
gaining power of the suppliers, the cost to produce can rise depending upon market
conditions, industry dynamics, and product demand. In such a case, the profits may
decline when suppliers use their bargaining power. Also, when the cost of switching
companies is less for the customers, they can use their power to drive down product
prices. In this situation, the companys revenues may decrease. Moreover, compa-
nies operating in government-regulated industries can experience drastic losses if
the new regulatory reforms negatively affect their business. The key factors to assess
when considering commercial due diligence are (1) industry attractiveness, (2) po-
sitioning of the target company, (3) opportunities for value creation, and (4) exit
strategies.
3. Explain legal aspects of conducting due diligence in PE.
The following are various legal aspects of due diligence in PE. First, the legal strat-
egy of due diligence requires various investment decisions with respect to tax ex-
emptions and the approval mechanisms of the various committees. It also involves
a thorough review of the terms and conditions of the agreement regarding various
partners involved in the fund and its portfolio. Second, due diligence involves iden-
tifying the status of investment managers with respect to other investment man-
agers. Ascertaining the fund managers qualification also falls under this category.
Third, risk due diligence requires various aspects involved in risk mitigation at vari-
ous levels and investments at different stages, especially upon exit.
The following are other legal aspects to be considered for due diligence in PE:

Any past criminal or administrative proceedings or investigations against the


firm, its affiliated entities, and/or its current and former team members.
Any past investigations by an industry regulatory body of the firm, its affiliated
entities, and/or its current or former team members.
Any pending or ongoing litigation/investigation against the firm, its affiliated
entities, and/or its current or former team members.
Any accusation and/or conviction of fraud or misrepresentation against any of
the firms current or former team members.
The activities of the firms support functions (e.g., finance and fund administra-
tion, human resources, and compliance/legal).
The processes and procedures for capital movements (e.g., capital calls, trans-
fers of cash, and investment acquisitions and distributions).
The third parties (e.g., law firms, custodians, fund administrators, prime bro-
kers, consultants, and banks) providing services to the firm or fund.
How the firm manages counterparty risk related to these third party arrangements.

4. Discuss strategic due diligence.


Strategic due diligence helps explain the variability of a respective companys
future cash flows. These future cash flows are used to determine the value of the
544 d i s c u s s i o n q u e s t i o n s a n d a n s w e r s

company in which the PE investors want to invest. Strategic due diligence helps
explain the position of a respective company in the market. In this sense, position
refers to the size of the company (e.g., whether it is a large cap, mid cap, or small
cap), closest possible competitors that are working on selling products and services
to the same target market, and driving forces that affect the company by considering
its size. After evaluating the companys position, the PE investor can evaluate the
future cash flows of the company. These future cash flows help determine both the
present value of the company and its equity.
5. Discuss how both corporate and national culture can influence a deal.
The recent growth of the international PE industry has gained much attention due
to the high level of capital flowing into the industry. When executing deals, manag-
ers need to identify any major people issues that may act as opportunities or threats
to the deal. PE managers should identify the similarities and differences in culture to
harness them and to explore differences to mitigate the risks associated with them.
PE firms should be sensitive to the fact that deals have financial implications and
involve the confrontation of two different organizational cultures. For cross-border
deals, different national cultures and their implications need to be understood.
Perceptions toward work environment vary from one individual, organization,
and country to another. These perceptions influence the way people respond to
different situations and influence their decision-making processes. When PE firms
undertake investments without studying the differences and similarities in respec-
tive cultures, financial benefits cannot be assured. The advisors and regulators in PE
firms need to understand cultural issues before making investments, and advisors
should continue to monitor any cultural issues even after making the investments.
Organizations generally exhibit two layers of culture: organizational culture and
national culture. Organizational culture refers to the informal set of values, norms,
and beliefs that control the way people and groups in an organization interact with
each other and with people outside the organization. Most members of the organ-
ization have a common set of values and attitudes that shape the work behavior of
employees, their perception of a situation, and their responses to it. Hence, the deci-
sions that people make in organizations are largely dependent on their values and
attitudes. National culture is critical to the success or failure of PE deals. Accord-
ing to Hofstede, Hofstede, and Minkov (2010) national cultures worldwide can be
understood through six dimensions (1) power distance, (2) individualism vs. col-
lectivism, (3) masculinity vs. femininity, (4) uncertainty avoidance, (5) long-term
orientation, and (6) indulgence vs. self-restraint.
6. Identify the challenges faced in the due diligence process.
The two key challenges faced in the due diligence process are as follows. First,
teams for conducting due diligence vary. For example, the team conducting financial
due diligence may not be the same team conducting the operational or the strategic
due diligence. If the teams do not have the right amount of communication among
themselves, the deal managers may have difficulty taking necessary action on prob-
lems that are highlighted by a particular team. Second, firms that have limited access
to the target companies pose a challenge during the due diligence. Therefore, in a
situation where PE investors have poor access to information, they need to redefine
the due diligence process to get a clear view of the targets business.
d i s c u s s i o n q u e s t i o n s a n d a n s w e r s 545

Chapter 18 Institutional Investors and Private Equity


1. Describe an institutional investor and list major types.
An institutional investor is an individual or non-bank organization trading finan-
cial securities in large volume and dollar amounts. This framework allows such in-
vestors to qualify for preferential treatment, mainly in the form of lower transaction
costs. Major institutional investors include banks, finance companies, insurance
companies, labor union funds, pension funds, endowments, and foundations. Insti-
tutional investors are subjected to fewer regulations because they are assumed to be
more capable of defending against poor investment decisions. Institutional inves-
tors generally invest on behalf of their clients.
2. Describe the growth and development of the PE market.
PE dates from 1946 when the American Research and Development Company
publicly traded closed-end funds to individuals. The explosive growth of the PE in-
dustry resulted from the decline in capital gain taxes in 1978 and the U.S. Depart-
ment of Labors reinterpretation of labor rules in 1979. In 1991, a total of about
200 deals had a combined value of $7.5 billion. By 2013, about 2,410 deals were
worth $455 billion. PE deals peaked in 2007 at $533 billion. In 2013, institutional
investors contributed capital at the highest level since 2008. In 2013, $162.19 bil-
lion was invested in U.S. PE funds, which are essentially portfolios of PE investment
opportunities including buyout, industry-focused, diversified PE, restructuring, dis-
tressed debt, and real asset funds. BO funds accounted for $74.17 billon, with 45
percent coming from five firms: Apollo Global Management, Bain Capital, Carlyle
Group, Kohlberg Kravis Roberts & Co., and Silver Lake.
3. Explain the structure of PE.
GPs in PE provide capital and management expertise as well as earn most of the
fixed revenue from management fees. In contrast, LPs provide capital via funds and
have their investment contracts often locked in for 10 to 12 years. The fund, either
by itself or as a fund-of-funds, is a pool of capital with direct operations. Investors
(institutional or private) acquire interests in the fund, which they in turn invest in a
portfolio of companies. GPs have legal power to act on behalf of the investors. The
management company or investment advisor plays an advisory role to the fund.
4. Outline the terms of the covenants that establish the contract between institutional
investors and PE.
The structure of the partnership agreement between institutional investors (LPs)
and the fund manager (GP) contains covenants that outline the rules of engagement.
A covenant contains a description of the investment mandate, including the target
companys stage of development. Covenants also outline restrictions on investment
activities such as the limits on investment horizon, amount of equity being invested,
and prudential limits on geographical exposure. Further, they provide limits on fi-
nancing such as maximum borrowings, cross-fund investing between two managers,
and reinvestment of capital. Additionally, covenants indicate the rights of the in-
vestors such as removing managers, reporting and accounting, and making investor
representations on advisory committees. Finally, covenants outline the economics
of the fund, such as the management fee (typically 1 to 3 percent of capital commit-
ted to the fund) and performance fee (typically 20 to 30 percent of the profits above
546 d i s c u s s i o n q u e s t i o n s a n d a n s w e r s

the return on capital and after the fund has crossed a minimum threshold rate of
return estimated to be 8 to 10 percent annually).
5. Distinguish between the behavior of individual and institutional investors in PE.
Understanding the difference between individual and institutional investors in
PE involves examining their contrasting behaviors in general investments, partly
due to the paucity of research on PE. Institutional investors hold more equity se-
curities in their portfolios than do individual investors. The larger number of stocks
owned by institutional investors provides them with more choices when selling.
Additionally, attention is not as scarce a resource for institutional investors as it is
for individuals. Institutions may limit their search to stocks in a particular sector
or meeting specific criteria. Institutional investors, particularly mutual funds, have
a propensity to herd, in which they buy and sell the same stocks at the same time.
They are often momentum investors, buying past winners although not systemati-
cally selling past losers. In PE, institutional investors are likely to also have a larger
information set and thus a propensity for herding.
6. Explain how institutional investors determine their choices in PE.
Typically, institutional investors commit a certain amount of capital to PE usually by
participating in a PE fund. The fund manager calls these investments over a period of
time when investment opportunities arise. Institutional investors generally invest in PE
for diversification purposes. Benefits from diversification will accrue if PE has a low cor-
relation with public equity. Unpredictability of cash flows and illiquidity of the markets
are factors in institutional investors decisions to invest. Institutional investors need an
efficient recommitment strategy to stave off both under-investing (thus losing economic
value) and over-investing (thus creating cash shortages and problems with liquidity).
Institutional affiliation also affects investment choice. Funds affiliated with banks
prefer investments in life sciences, bio-tech, and health-related industries. In con-
trast, funds sponsored by life insurance companies and pensions show little interest
in start-ups. Finally, the presence of agency costs affects the choice of investing. In-
vestors in privately held corporations are mainly concerned about the expropriation
of their interests at the hands of the controlling shareholders. Fear of minority ex-
propriation is important for institutional investors because the law bans them from
being controlling owners and participating in decision-making that suggests control.

Chapter 19 Private Equity and Value Creation


1. Identify the main value creation factors for LBO transactions.
Several items contribute to value creation in LBO transactions.
Reducing the agency costs between the management and the shareholders
is of the utmost importance. Jensens (1986) free cash flow theory considers the
discipline effect of leverage to achieve this reduction. Others consider additional
factors, such as the motivational schemes implemented because these are engi-
neered to achieve full alignment of interests. Reducing agency costs between credi-
tors and shareholders is an important factor because it allows creditors to be more
comfortable in providing large amounts of external capital to a company. Several
financial techniques exist to reach this goal including strip financing and covenants
d i s c u s s i o n q u e s t i o n s a n d a n s w e r s 547

on financial indebtedness. LBO transactions have been useful when identifying


undermanaged and undervalued targets. For example, this sample includes public
to private deals where the stock market undervalued the public company. LBO
transactions benefit from the tax shield on debt financing (i.e., their tax expenses
are reduced by the amount of interest paid on the acquisition debt). LBO transac-
tions are successful in improving the cash-flow generation of their target companies
through better working capital management.
2. Discuss the factors contributing to the failure of the traditional LBO models as ex-
plained by valuation multiples.
The failure of the traditional LBO models are exhibited between 2005 and 2007
when leverage played the most important role in explaining the valuation multiples.
For this period, LBO transactions reached high valuation levels that were justified
mostly by inexpensive and abundant leverage. As a result, LBO funds focused less
on other measures of value creation, such as corporate governance or cash-flow im-
provement and aimed to reach their target IRR only by loading target companies
with as much debt as possible. Being able to raise a large sum of funds through lev-
erage helped LBO funds paying valuation multiples in excess of historical levels.
This framework has resulted in a situation in which LBO funds abandoned their
traditional goal of identifying undervalued and under managed targets. At the same
time, overconfidence in the discipline effects of debt on agency costs led LBO funds
to put less emphasis on other corporate governance measures that traditionally also
helped them reduce agency costs.
3. Explain the relationship among leverage, the IRR, and industry movements accord-
ing to the study presented in this chapter.
For the whole sample, leverage explains about one-third of the IRR performance
of LBO companies whereas the evolution of market multiples explains about 20
percent. This result provides evidence for the discipline effects of debt. Company-
specific performance explains the remaining 50 percent. Above industry growth and
governance mechanisms have the same cumulated effect as leverage. Leverage seems
to serve as a powerful disciplinary tool to reduce agency costs, more so than corpo-
rate governance measures. However, for the vintage year transaction between 2005
and 2007, company-specific performance explains only 36 percent of the IRR calcu-
lations while leverage explains almost 40 percent. This finding is another example of
the important role that leverage played for driving LBO transactions in that period.
4. Discuss key performance indicators for which PE ownership had the most signifi-
cant impact and compare these indicators to those for strategic transactions.
A clear difference exists on which key performance indicators for LBO transac-
tions and strategic transactions have an impact. For strategic deals, the most impor-
tant post transaction driver is seen in top line growth (in absolute terms or relative
to the industry) and the EBITDA margin evolution. In contrast, the impact of LBO
ownership seems to be the most important factor on EBITDA growth and the im-
provement of the working capital and capital expenditures efficiency ratios. This
result is consistent with the fact that LBO funds seek to maximize their exit value.
One of the best ways to maximize the IRR is to increase the EBITDA in the exit
year as much as possible versus the entry year EBITDA. Improving cash manage-
ment (working capital and capital expenditures) also helps to improve the IRR as
548 d i s c u s s i o n q u e s t i o n s a n d a n s w e r s

LBO funds are driven to reimburse as much debt as possible before the companys
disposal, thus increasing the value of the equity stake (all other things else being
equal). Excess cash can be diverted to debt reimbursement by improving efficiency
thus leading a higher IRR.

Chapter 20 Compensation Structure


1. Consider a fund with a 20 percent carried interest and a hurdle rate of 8 percent. The
committed capital is $100 million and the entire amount is invested on the first day
of its operation. The investment is sold for $110 million in one year and the fund is
liquidated on that day. Describe how profits are split between LPs and GPs under a
100 percent catch-up provision and a 50 percent catch-up provision.
The funds IRR at the end of the first year is 10 percent. Since the hurdle rate is 8
percent, the LPs will first receive $8 million [($100 million)(1.08) $100 million].
Under the 100 percent catch-up provision, the GPs will receive profits until they earn
20 percent of the cumulative profits. The LPs share of $8 million corresponds to
80 percent of the cumulative profits; hence the assumed cumulative profits are $10
million ($8 million/0.80). Therefore, the GPs earn the next $2 million [0.20($10
million)]. Under the 50 percent catch-up provision, the GPs will share 50 percent of
the incremental profits until they receive 20 percent of the cumulative profits. Let x
be the incremental profits. Then 0.50 x = 0.20($8 million + x) and x = $5.33 million.
Therefore, when the incremental profits are below $5.33 million or the cumulative
profits are below $13.33 million = $8 million + $5.33 million, the GPs and LPs split
the profits 50/50. Since the cumulative profits are $10 million of which $8 million is
already distributed to the LPs, the GPs and LPs each receive $1 million [= 0.50($2
million)]. On balance, with the 100 percent catch-up provision, the LPs receive $8
million and the GPs earn $2 million. However, with the 50 percent catch-up provi-
sion, the LPs earn $9 million (= $8 million + $1 million) and the GPs earn $1 million.
2. Consider two funds that each invest $100 million today and produce $120 million
over the next three years. Both have an 8 percent hurdle rate, a 100 percent catch-up
provision, and a clawback provision. The two funds generate cash flows at different
times as shown below. Compute how the LPs and GPs will share the profits.

0 1 2 3

Fund A 100 + 108 +2 + 10


Fund B 100 + 100 + 10 + 10

This problem shows how compounding affects the preferred return (i.e., hurdle rate).
Fund A. In year 1, the fund earns 8 percent return and the $8 million profit goes
to the LPs (with the contributed capital). In year 2, the GPs catch up and earn the
d i s c u s s i o n q u e s t i o n s a n d a n s w e r s 549

$2 million as indicated in question 1. In year 3, the GPs and LPs split the profits ac-
cording to the preset carry rule (20/80) (i.e., $2 million and $8 million, respectively).
Fund B. In year 1, the LPs receive the contributed capital ($100 million). In year
2, with a hurdle rate of 8 percent, the GPs should return $16.64 million [= ($100
million) (1.08)2 $100 million] of the profits to the LPs. Therefore, the $10 million
goes to the LPs. Since all the contributed capital is returned to the LPs, the pre-
ferred return will not be earned on the contributed capital from year 2 to 3. In year
3, however, the unreturned preferred return of $6.64 million goes to the LPs and the
rest ($3.36 million) goes to the GPs. Assuming that the unreturned preferred return
should earn the preferred return from year 2 to 3, then the LPs receive $7.17 million
[= ($6.64 million)(1.08)] in year 3 and the GPs take the rest.
3. Denote c as the carried interest, h as the hurdle rate, and u as the catch-up. Deter-
mine the IRR of a fund at which the catch-up provision expires (i.e., GPs and LPs
share profits according to the preset carry rule).
Let r percent be the threshold IRR. The LPs first receive h percent. Then the GPs
take u percent of the next (r h) percent until they receive c percent of the cumu-
lative return (i.e., r percent). Therefore, u(r h) = cr. Solving for r, r = uh/(u c).
4. Discuss how the relative bargaining power of LPs and GPs affect various fee terms in
the LPAs.
LPs try to negotiate lower management fees and carried interest. Also, when pos-
sible, they try negotiating a budget-based management fee in which the GPs receive
an approval for a planned operating budget. The LPs do not want the GPs to receive
deal-related fees so that the GPs interest lies in making profits via carried interest.
The LPs also like to set an appropriate level of preferred return and clawback clause
with a provision ensuring the implementation of the clawback clause. Additionally,
LPs prefer having a back-end-loaded carry waterfall provision in which profits from
the fund are split at the end of the funds life according to a set carry rule.

Chapter 21 Global Regulatory and Ethical Framework


1. Discuss whether PE funds are investment companies subject to regulation under
the Investment Company Act of 1940 and whether they must register offerings of
their investment interests under the securities laws.
PE funds are exempt from registration under the securities laws and are not clas-
sified as investment companies because their investors have certain wealth char-
acteristics. Under the securities laws, to gain exemptions, investors are required
to have a net worth of at least $1 million, excluding their personal residence, but
including retirement savings. Qualified purchasers for exemption from investment
company registration must have $5 million of investments, exclusive of the PE fund
investments. Investors with considerable wealth often can bear the risk of loss from
an investment without jeopardizing their standard of living or compromising their
future needs. The wealth threshold, however, may be too low to identify investors
who can bear the risk of loss. Assets of $1 million include investments in retirement
plans and may over-identify wealthy individuals. The $5 million investments thresh-
old for qualified purchasers in PE funds is a more satisfactory level of wealth but PE
550 d i s c u s s i o n q u e s t i o n s a n d a n s w e r s

funds tend to be illiquid for long periods so a threshold focusing on liquid wealth to
meet the investors ongoing need might be better. Wealth does not indicate sophis-
tication and understanding of investment risks. Many people attain wealth through
inheritance, good fortune, or other means and may be vulnerable to promoters of
inappropriately risky investments or investment fraud that registration under secu-
rities laws might inhibit. The accredited investor definition recognizes the need for
both wealth and sophistication and contemplates that accredited investors will have
wealth and either sophistication or an investment representative with sophistica-
tion. For qualified purchasers, $5 million of other investments, if sufficiently liquid,
should be ample for current needs, but the wealth level will not protect investors
from fraudulent or excessively speculative investments.
2. Explain the laws and private strategies limiting the success of hostile corporate take-
overs in the United States.
Poison pills and state antitakeover laws give incumbent management considera-
ble ability to prevent hostile takeovers by reducing the economic attractiveness of
target entities and restricting shifts of control and the acquirers ability to restructure
the target, sell its assets, or distribute to shareholders. Hostile takeovers rarely occur
because of antitakeover laws and poison pills.
3. Discuss whether the laws and strategies used in the EU are similar to those in the
United States.
The laws and strategies used in the EU are similar to those in the United States
but are more limited in nature. The Takeover Directive focuses on the best price for
shareholders and assuring that all shareholders have the opportunity to get the same
price for their shares. The Takeover Directive suspends defensive strategies while
an offer is outstanding. The AIFM Directive, however, defers post-takeover capital
impairments for 24 months.
4. A former German minister described PE funds as a swarm of locusts. Interpret the
meaning of his statement and discuss whether he was correct in drawing that analogy.
In seeking profit from acquisitions, PE funds restructure and occasionally liq-
uidate target enterprises or sell off some of their assets. Hence, PE funds destroy
enterprises and leave devastation in their wake as they capture windfall profits as
locusts destroy crops. Considerable disagreement exists on the value of the PE fund
function. The opposing view is that PE funds enhance enterprise value by increas-
ing efficiency through the disposal of inefficient operations and excessive liquidity.

Chapter 22 Private Investment in Public Equity


1. Describe a PIPE offering.
Companies typically issue PIPE securities under Regulation D, which allows
public companies to issue stocks privately to accredited investors without the need
for public registration as long as the seller follows a set of listed requirements. After
closing a PIPE transaction, the issuer prepares and files a resale registration state-
ment with the SEC.
2. Explain how a PIPE offering differs from a SEO and why PIPEs have become an al-
ternative to SEOs.
d i s c u s s i o n q u e s t i o n s a n d a n s w e r s 551

In contrast to a traditional private placement, a PIPE closing does not depend


upon the SEC review process, making PIPE issuance a time-efficient mechanism.
However, investors cannot resell purchased securities until the SEC declares that
the registration statement is effective. Another advantage is that the financial con-
tracting template used in a PIPE allows for several state-contingent terms, which
can be finely tailored to match the particular needs of a given investment. The PIPE
market became popular because it solved an important matching problem: small
public firms with high information asymmetry that are traditionally barred from the
SEO market can now get funding in a timely fashion from investors such as hedge
funds and PE funds seeking a superior risk premium.
3. Identify the major investors in the PIPE market and how their objectives affect PIPE
deal structure.
In terms of capital invested, the major investors in the PIPE market are hedge
funds, pension and government funds, corporations, mutual funds and institutional
advisors, buyout firms, VC firms, brokers and dealers, banks, insurance companies,
and charitable, educational, and family trusts. Strategic investors are often involved in
actively monitoring management after a PIPE issuance and often strive to maintain
their investment in the PIPE issuer for a relatively long time horizon. By contrast,
financial investors prefer short-run cash profits from PIPE transactions and are rela-
tively passive investors even when they hold block stakes. PIPE issuers with strategic
investors such as corporations and VC funds often offer investors more control rights
and less superior cash-flow rights than those with financial investors. Firms financed
by strategic PIPE investors significantly outperform those with financial investors.
4. Identify the three categories of contingent contractual terms in a typical PIPE offer-
ing, give an example for each category.
The three categories of contingent contractual terms in a typical PIPE offer are:
(1) investor protections, which contain terms that attach various protections to a
PIPE investors stock; (2) trading restrictions, which contain terms that determine
how an investor can trade the underlying stock after an offering; and (3) issuer
rights, which contain terms that give an issuer the right to force an investor to take
certain actions. The terms in these latter categories favor issuers at the expense of
investors. Examples for investor protections include dividend, interest and warrants,
investor registration rights, anti-dilution protection, right of first refusal, investor
call options, and redemption rights. Examples for trading restrictions are prohibi-
tions on shorting or hedging and a lock-up period. Examples for issuer rights are
company-forced conversions, company put options, and optional redemptions.
5. Discuss the role that placement agents play in a typical PIPE deal.
Placement agents play several roles in a typical PIPE deal. First, they conduct
due diligence on the issuer. Second, placement agents identify a group of investors
interested in the deal. Third, they bring the deal over the wall, which refers to an
investor expressing serious interest in the PIPE offering. Once the deal is over the
wall, the agent reveals the identity of the issuer and gives the investor more detailed
financial and operational information. Finally, placement agents advise the issuer on
negotiating the PIPE offering.
6. Define a CMPO and discuss its major advantage relative to traditional PIPEs and
SEOs.
552 d i s c u s s i o n q u e s t i o n s a n d a n s w e r s

Some refer to CMPOs as wall-crossed, pre-marketed, or overnight offerings.


A major feature of a CMPO is that it includes a public offering phase following a pri-
vate marketing phase to a selected group of accredited institutional investors. Once
the underwriter has obtained an indication of interest from the investors contacted
during the confidential phase, the issuer announces the offering to the public at the
closing of the market and files the selling documents. The underwriters then market
the offering more broadly to institutional and retail investors. This public phase con-
tinues into the evening and the pricing is finalized and announced before the market
opens the following day. CMPOs are typically priced at a discount compared to the
current market price. Its public phase is similar to a traditional SEO.
CMPOs have certain advantages over a traditional SEO and a standard PIPE. For
instance, CMPOs are executed very quickly, often just days in comparison to a fully
marketed SEO that often takes weeks or months. The confidential feature of the first
phase also allows the issuer to test investors appetite for its security and to negotiate
the pricing and terms without risking a potential decline in the stock price due to
speculative trading before the announcement. Additionally, since the shares sold in
CMPOs are freely transferrable, the transaction has no illiquidity premium, which
investors usually request in a standard PIPE.

Chapter 23 Listed Private Equity


1. Explain why listed PE might be more commonly considered by institutional inves-
tors based in the United Kingdom than their counterparts in continental Europe.
The United Kingdom has the most liquid stock market and home bias is widely
documented. Thus, greater institutional investor interest is expected in listed PE
in the United Kingdom than in continental Europe. Furthermore, broker research
analysts who are primary London-based cover the listed PE sector and provide in-
sightful research to their clients. Also, current evidence is consistent with the view
that English legal origin countries afford greater protection to investors for publicly
listed companies.
2. Discuss why listed PE is less likely to be considered by an institutional investor
whose decision-making is delegated to an equities team.
Human resources can have a pronounced impact on an institutional investors
interest in investing in listed PE. The identification and manager selection process
in PE is relatively costly and successful manager selection can increase the proba-
bility of generating excess returns. Early adopters of PE have developed specialist
investment teams that have cultivated the industry networks and the skills to access
private placements of the best fund managers. Recent studies report significant dif-
ferences in institutional investors ability to select successful investment managers.
Some have investment decisions made by more generalized equities teams that are
more specifically trained at stock picking and arguably have less time, experience,
and skill to carry out due diligence and negotiate contracts for limited partnerships.
The equities team is often more comfortable with investing in listed PE.
3. Discuss why listed PE enables institutional investors to achieve their target PE al-
locations quicker, and as such, institutions that invest in listed PE are less likely to
d i s c u s s i o n q u e s t i o n s a n d a n s w e r s 553

adjust their listed PE allocations over time in response to slower adjustments to lim-
ited partnership PE allocations.
Most institutions investing in PE have budgeted annual allocations to the asset
class. These targets may not necessarily be met as the institution is dependent on
the expertise of its PE team to source, identify, negotiate terms, and commit the
budgeted amount within the stipulated time frame. Unfortunately, PE investment
through private placements has a distinct disadvantage of taking many months if
not years for an investor to achieve the desired exposure level (i.e., capital invested).
Listed PE provides two advantages to an investor, especially in its relation to pri-
vate placements. First, the institution can achieve relatively rapid exposure to PE
through listed PE funds. Second, maintaining a listed PE exposure alongside private
placements provides a dynamic adjustment mechanism for an investors overall PE
exposure. For example, the investor could start at 100 percent listed PE and then
reduce listed PE exposure as limited partnership exposure increases. Used in this
manner, listed PE reveals an investors liquidity-time preferencesthe willingness
to trade-off exposure tomorrow from private placement with the exposure and flex-
ibility today through listed vehicles.
4. Discuss how listed PE funds have performed over time since their first listing date.
Several studies examine the performance of listed PE since its first listing date. For
example, Lahr and Kaserer (2009), who analyze 79 ordinary funds and 21 listed PE
funds between 1992 and 2008, find that listed PE funds start at an initial premium of
2.5 percent and adapt to the long-term average of 21 percent after two years. They
note that premia predict future returns and are explained by liquidity but not by in-
vestor sentiment or the funds investment degree. PE fund premia seem to depend
on credit markets and systematic risk. Jegadeesh, Krussl, and Pollet (2009) examine
the performance of PE fund-of-funds into unlisted PE funds, and compare the per-
formance to listed PE funds. Based on data from 26 PE fund-of-funds and 129 listed
PE funds from 1994 to 2008, they estimate the markets expectation of unlisted PE
funds (via fund-of-funds) abnormal returns (and net of their fees) to be 1 to 2 per-
cent above the market accounting for risk, while the markets expectation for listed
PE abnormal returns is zero to marginally negative. The authors find that the betas
of listed PE and unlisted PE (via fund-of-funds) are close to one. PE fund returns are
positively correlated with gross domestic product (GDP) growth and negatively cor-
related with credit spread. Godineni and Megginson (2010) contend that listed PE
became relatively more popular in 2007 due to market changes and the need for fund
managers to seek alternative ways of raising capital. They also argue that listed PE af-
fords insiders an alternative way to cash out. Blackstones 2007 IPO performance is
consistent with these rationales compared to contemporaneous returns on the S&P
500 index. The performance of listed PE improved after the financial crisis of 2007
2008, despite early poor performance of some newly listed PE funds. Raising capital
continues to grow. In 2014, more than 200 listed PE funds are available worldwide.
More recent performance statistics shows that listed PE does better than the S&P
500 index and the MSCI Europe index, although they are more volatile in nature.
5. Explain the values of the listed PE fund betas.
Based on data from 26 PE fund-of-funds and 129 listed PE funds over 1994
2008, Jegadeesh et al. (2009) find the betas of listed PE and unlisted PE (via
554 d i s c u s s i o n q u e s t i o n s a n d a n s w e r s

fund-of-funds) to be close to one. More recent performance statistics shows that


listed PE does better than the S&P 500 index and the MSCI Europe Index, although
more volatile (beta greater than 1, and indeed the beta for listed PE funds such as
Blackstone is greater than 2). A problem with estimating listed PE betas is that in-
vestments not yet sold can be sticky (i.e., do not change much over time). Thus,
betas may be understated.
6. Discuss why listed PE is a more attractive asset class for smaller institutional
investors.
Empirical studies consistently show PE investment is a specialized asset class
involving high information and search costs as compared with listed equities. As
a result, an investors willingness to incur search costs (e.g., information collec-
tion and access to high quality managers) in order to generate risk-adjusted excess
returns over public markets will be associated with size of the investor, type and
ease of access such as that facilitated by locational advantages. With respect of size,
smaller institutions do not have the time and possibly the experience or skills to
incur high search costs in identifying high-quality managers, or to negotiate limited
partnership contracts. Smaller institutions also may face limited access to more rep-
utable PE funds with existing institutional investors investing in multiple fund vin-
tages within the same PE firm, which is a major disadvantage in view of evidence of
significant performance persistence among PE funds. The idea of significant perfor-
mance persistence among PE funds is extremely important for limited partnership
and listed PE funds alike and shows that manager selection is critical.

Chapter 24 Private Equity Growth in International


and Emerging Markets
1. Describe the global trend of the PE investment volume and the development in tra-
ditional versus emerging countries.
Global PE activity reached an all-time high with about $270 billion at the begin-
ning of the new millennium. The burst of the new economy stock market bubble also
led to a resulting depression in the PE market. Annual activity fell to less than $100
billion globally between 2002 and 2004. Then, activity revived and grew to almost
$200 billion before the financial crisis of 20072008 affected the PE market again.
Since then, global investment volume has been volatile and has not yet reached the
2007 levels as of 2014. Nevertheless, the trends in traditional and emerging PE mar-
kets differ. The traditional markets (e.g. the United States, Canada, and the United
Kingdom) still encompass most PE investments. However, the historic peak activ-
ity levels have not yet been attained, while emerging markets gain momentum and
market share. The BRICs (Brazil, Russia, India, and China) are attracting growing
amounts of capital and are gradually increasing investment volume over time. China
ranks among the top countries for PE activity, closely followed by India. Several
other emerging countries might follow the BRICs development path, notably Indo-
nesia, Mexico, the Philippines, and Turkey. Nevertheless, several conditions need to
be improved in those countries to attract large-scale PE allocations.
d i s c u s s i o n q u e s t i o n s a n d a n s w e r s 555

2. Discuss why PE investments are concentrated in particular economies.


The most important criterion for geographic PE allocation decisions of institu-
tional investors is the expectation of promising deal flow. If investors believe that
a country can produce sufficient deal flow, they allocate fractions of their portfo-
lio accordingly. Deal flow arises with economic activity and growth, but requires
additional favorable settings. This favorable environment exists in many developed
countries but first needs to be developed in emerging and frontier markets.
3. Discuss the six key criteria that lead to deal flow and attract investors.
Groh, Liechtenstein, and Karsten (2010) suggest using six criteria for a coun-
trys attractiveness for institutional PE investors: (1) economic activity, (2) depth
of the capital market, (3) taxation, (4) investor protection and corporate govern-
ance, (5) human and social environment, and (6) entrepreneurial culture and deal
opportunities. The single most important key driver is debatable. However, com-
paring these criteria with the actual PE investment activity across countries reveals
that these characteristics are well-developed and rather balanced in countries with
a large number and amount of investments. Hence, countries need to be successful
with respect to all the key drivers to attract international PE capital. Viewed from
a different perspective, a high correlation exists among the key driving forces and
countries need to have reached a certain level of development to allow the PE trans-
action model. Black and Gilson (1998) refer to this correlation as a chicken and
egg problem because determining causality or the factors that are most required for
successful PE equity investing is difficult, if not impossible.
4. Explain why the BRICs attract increasing amounts of PE capital.
The unique combination of economic catch-up potential and population in the
BRIC countries yields substantial demand for financing. China has been among the
top countries of PE activity in recent years, followed closely by India. In parallel to
their economic progress, the BRICs, but notably China and India, have meanwhile
developed strong capital market activity, which supports the PE deal-making infra-
structure. Nevertheless, the key driving factors of PE investments are unbalanced in
these countries. Investor protection, the human and social environment, absence of
bribery and corruption, and the entrepreneurial culture, which provides appropri-
ate deal opportunities, are poorly developed compared to the traditional PE mar-
kets. If the BRICs manage to improve these conditions, their PE markets should
become even more vibrant.
5. Discuss whether a threshold of economic development exists in which PE activity
emerges.
Maturity in terms of economic development is an important factor in the inter-
national allocation of institutional investors. A country has to fulfill minimum re-
quirements to achieve investors attention. First, size matters because institutional
investors might not notice smaller countries. Their transaction costs simply deter
investments in markets that are not large enough. Although allocations to regional
funds might help to overcome this issue, these funds impose additional fees to in-
vestors. Second, an expectation of sufficient deal flow must exist. This expectation is
determined by the socioeconomic environment in a country and can be expressed
by the six key drivers. If these key drivers are sufficiently developed, PE activity
emerges.
556 d i s c u s s i o n q u e s t i o n s a n d a n s w e r s

Chapter 25 Diversification Benefits of Private Equity


Funds-of-Funds
1. Discuss the two typical forms of PE funds-of-funds.
Investing in funds-of-funds can be accomplished through proprietary funds-
of-funds and pooled funds-of-funds. The asset class of self-made funds-of-funds is
called proprietary funds-of-funds allowing investors to meet strategic, diversifi-
cation, and tax targets. Proprietary funds-of-funds can be set up in two ways. The
first method is by investing in many funds directly. The advantages to this strategy
are that an investor can build up long-term relationships with GPs, gain industry
knowledge, and have full control over the investment strategy. The second method
involves employing an external fund-of-funds management team, which customizes
the fund-of-funds. Pooled funds-of-funds collect capital from investors to invest in
funds. Investors of pooled funds-of-funds benefit from internal cost-savings because
they do not have to negotiate and monitor several single investments. Pooled funds-
of-funds offer the advantages of professional management teams and due diligence,
as well as access to information that would otherwise be expensive or possibly un-
obtainable by an individual investor. The main drawback of pooled funds-of-funds
is that they involve an additional layer of fees. Besides the fees the fund-of-funds
charges, all fees are passed from the underlying portfolio funds to its investors.
2. Discuss an alternative structure to the general 220 framework for managerial
compensation.
Some GPs establish a management fee based on committed capital, but with a re-
duced percentage after the initial investment period. Others charge a stable fee per-
centage but shift the fee basis from committed capital during the investment period
toward net invested capital during the divestment period. In some LPAs, a combi-
nation of both a reduced fee percentage and a shift in fee basis is agreed upon. These
alternative structures allow the customization of efforts to better align managerial
incentives for GPs to minimize potential agency problems.
3. Describe the typical fees of a PE fund and funds-of-funds.
Fees in PE consist of fixed and variable components. The fixed part is the man-
agement fee, which is used to compensate management and cover operating costs.
The LPA sets forth the parameters to calculate this fee, fee percentage, and fee basis.
These parameters often vary over time. The typically agreed contractual fee percent-
ages range between 1 and 3 percent. Carried interest is the variable component of the
GPs remuneration and is its share in the capital gains. GPs are compensated for the
risk they take with their ownership stake of usually 1 percent of total funds. GPs only
receive carried interest, after LPs have received at least their initial investments plus
a preferred return (hurdle rate), which is typically established at 8 percent a year.
4. Explain how to calculate carried interest for a fund using the following data and as-
suming a $100 million fund. Assume a 2 percent management fee, 20 percent car-
ried interest rate, and 100 percent catch-up period.
Year 1 2 3 4 5 6 7 8 9 10 Total
CF 50 30 10 10 60 85 7 20 40 38 150
d i s c u s s i o n q u e s t i o n s a n d a n s w e r s 557

The carried interest of 20 percent is often linked to a preferred return or hurdle


rate. GPs only receive carry when LPs have received at least their initial investments
including management fees plus a preferred return, which is typically 8 percent a
year. Carry is calculated as the agreed upon carried interest times all previous cash
flows. For year eight, the carry is calculated in the following manner:
Carry = 0.20($52 $32 $12 $12 + $58 + $83 + $5 + $18) = $11.20
The hurdle rate is measured by the internal rate of return (IRR) on all previous
cash flows. The IRR of year seven for example is calculated by
7
CFt
NPV = = $0
1 (1 + i)t
NPV = $52/1.08 $32/1.082 $12/1.083 $12/1.084 + $58/1.085 + $83/1.086
+ $5/1.087 $0
with NPV as the net present value of the investment. It is calculated as the sum
of the discounted cash flows (CF). The discount rate i is the IRR that has to be esti-
mated in a first step and t stands for the year the cash flows occur. If the NPV is above
(below) zero the next estimation of IRR needs to be higher (lower). This process is
repeated until the equation for NPV = 0 holds true.
Most LPAs have a catch-up period during which GPs receive a pre-specified per-
centage of net returns until net returns are distributed in the agreed-up carried in-
terest ratio between GPs and LPs. After the catch-up, all further proceeds over the
hurdle rate are distributed in this ratio.

Year 1 2 3 4 5 6 7 8 9 10 Total
CF 50 30 10 10 60 85 7 20 40 38 150
Fees 2 2 2 2 2 2 2 2 2 2 20
CF 52 32 12 12 58 83 5 18 38 36 130
IRR 1 1 1 1 0.19 0.07 0.08 0.11 0.15 0.17 0.17
Carried 0 0 0 0 0 0 0 11.20 7.60 7.20 26
Interest

Chapter 26 Publicly Traded Private Equity


1. Discuss the advantages and disadvantages of PTPE compared with conventional PE
from an investors viewpoint.
From an investors perspective, PTPE has various advantages and disadvantages
compared with conventional PE. A company goes public for various reasons includ-
ing obtaining financing, enhancing liquidity and public relations, improving moti-
vation and incentives for management and employees, and exploiting mispricing by
trading its own stock. Besides pension funds and wealthy individuals who invest in
conventional PE, PTPE firms may gain access to a wider range of investors. Being
publicly traded suggests a certification benefit but may be more costly due to the
legal requirements of being a listed company. A public listing offers a possibility for
the PE firms owners to divest their shares. Investors can use PTPE vehicles to invest
558 d i s c u s s i o n q u e s t i o n s a n d a n s w e r s

excess cash from the non-listed PE portfolio at short notice. Finally, PTPE funds
provide a benchmark for non-listed funds.
2. Describe the organizational forms of PTPE and explain the similarities and differ-
ences to non-listed PE.
Listed PE funds mirror non-listed ones, while listed management companies are
similar to the GP in non-listed funds. Investment companies combine both organ-
izational forms and are sometimes difficult to distinguish from ordinary holding
companies if they do not explicitly say they follow a PE business model. Finally,
listed funds-of-funds are vehicles that invest in non-listed PE funds. PE firms can be
managed internally or externally. Listed PE funds and investment companies often
also invest in other PE, which sometimes puts the private in listed private equity
at risk.
3. Explain the reasons for discounts to NAVs in listed PE funds.
Differences between NAVs and market prices of listed PE can best be explained
by investor sentiment and liquidity. Investor sentiment and noise trader risk are
likely explanations because of the difficulty in setting up an arbitrage portfolio of
private companies to profit from relative price divergence. Managerial ability that is
revealed over time may also explain discounts in closed-end equity funds, but this
explanation finds little support in PE funds. Discounts also depend on credit mar-
kets, which correspond to the tendency to use leverage in portfolio firms.
4. Discuss how different organizational structures can explain variation in market risk
across PTPE firms.
Differences in market or systematic risk across PTPE firms can arise for several
reasons. Operating activities could induce high or low levels of business risk, lever-
age could amplify this business risk, or tax regimes could differ across companies.
Market risk is high in internally managed vehicles but low in externally managed
ones suggesting that different sources of cash flow such as management fees and
carried interest cause variation in systematic risk. The organizational structure in PE
funds creates an asymmetry between the management company (GP) and share-
holders of the fund. This is because the management company receives a dispropor-
tionately large share of gains and may also participate in losses to a greater extent
than the fund it manages. If fund returns are correlated with market returns, market
risk will be different for the fund and its GP.

Chapter 27 The Future of Private Equity:


A Global Perspective
1. Explain the evolution of PE in terms of geography and discuss globalization trends
for the next decade.
PE has its roots in the United States, beginning in the mid-1950s. The firms started
as buyouts and evolved into LBO firms in the 1980s. PE firms began expanding into
Europe, mostly Western Europe, in the 1990s. The 2000s saw expansion of U.S. PE
firms into Asia and, in the last few years, funds have begun entering Latin American
markets. In the next decade, firms will continue to look to new markets for profitable
opportunities. The BRICs will still be important but Southern European countries
d i s c u s s i o n q u e s t i o n s a n d a n s w e r s 559

such as Greece, Spain, and Italy are appealing as government reforms take hold.
Also, Sub-Saharan Africa, East Asia, and developing countries in South America are
likely to garner investor attention.
2. List the drivers of PE returns in the 2000s and early 2010s and explain what is likely
to have a bigger influence on returns in the latter half of the 2010s and beyond.
Historically, the driving forces of PE returns have been GDP growth, P/E mul-
tiple expansion, and cheap debt. In the next decade, firms will have to generate
returns from operational efficiencies at both the portfolio company level and the
PE firm level. This is true for several reasons. Bank debt will likely be more expen-
sive or even unavailable due to regulatory changes following the financial crisis of
20072008. GDP growth has slowed in some countries and may continue to do
so. Also, while P/E multiple expansion may occur in certain periods and sectors, it
cannot solely be relied upon for long-term growth and profitability.
3. Discuss the primary trend in institutional investing in PE and the forces driving that
trend.
Institutional investing in PE and VC has increased since the financial crisis of
20072008 and is expected to continue to increase. The primary trend driving the
increase in institutional investing in PE is the demand for higher returns. In the
United States, for example, many states and firms have underfunded pension plans.
High PE returns are appealing to lawmakers trying to fix the underfunding prob-
lem. Foundations also are seeking higher returns as states continue to cut university
budgets. Further, sovereign funds are increasing investments in PE.
4. Define a sovereign wealth fund and explain the role that sovereign wealth funds play
in PE.
Sovereign wealth funds are government-owned funds that invest in real and finan-
cial assets. These funds invest globally and are generally funded by foreign exchange
reserves and export revenues. AUM of sovereign funds are expected to increase
to $8 trillion by 2020 from around $5.35 trillion in 2014. These sovereign wealth
funds, along with other large investors, are becoming more demanding, choosing
direct investments, rather than investments via PE funds, as a means of avoiding PE
management fees.
5. Discuss how the Dodd-Frank Act affects PE.
The Dodd-Frank Act, passed in 2010 in response to the financial crisis of 20072008
is a massive piece of financial reform legislation that affects banks, consumers, PE funds,
VC, angel investors, and all others involved with financial markets and institutions. Sev-
eral provisions of the Dodd-Frank Act affect PE. A key provision of the Dodd-Frank
Act that affects PE is the Volcker Rule, which prohibits banks from taking an equity
ownership position in a PE fund or hedge fund. Another provision that is relevant is
the Advisor Registration provision, which requires all PE funds with $150 million or
more in AUM to register with the Securities and Exchange Commission (SEC). This
change means that many more PE firms have to register and reveal previously private
information about their operations to the SEC and public. Firms will be required to fill
out Form PF detailing information about the type, size, ownership structure, and oper-
ations of various funds. Reporting and recordkeeping requirements are increased under
Dodd-Frank as well, increasing costs for most PE firms.
Index

(Page numbers in italics refer to tables (t) and figures (f) within the text.)

A-B loans, 9394 comprehensive valuation example, 133, 134t,


abnormal returns 140, 143t144t
cost of capital for private equity, 149, 151, 156 using cash flows to estimate value, 130131
accountants, due diligence, 293 advertising. See marketing/advertising, new rules
accredited investors, 312, 380381, 397398 adviser. See investment advisers
distinguished from retail investors, 512 AIFM Directive, 381
global regulatory and ethical framework, alternative investment fund (AIF), 38, 175
380381 liquidity issues during global financial crisis,
institutional investors, 312 175
net worth calculation change, Dodd-Frank Act, alternative investment fund manager (AIFM)
512513 acquisition regulation, 390
new marketing rules, 512513 European Union regulatory developments,
acquisition debt multiples, 111, 113f post-financial crisis, 38, 41
acquisition regulation, 382391 Alternative Investment Fund Managers Directive
AIFM directive, takeovers, and private equity (AIFMD)
funds under national law, 390 liquidity issues during global financial crisis,
alternative investment fund manager (AIFM), 174175, 177
390 Alternative Investment Fund Managers Directive
Alternative Investment Fund Managers (AIFMD) (European Union), 33, 38, 4142
Directive (AIFMD), 390, 391 acquisition regulation, 390, 391
closely held business acquisition, 382383 global regulatory and ethical framework, 378,
corporate combinations, eliminating minority 380, 381382
shareholders, 386387 liquidity issues during global financial crisis,
corporate takeover defenses, 387391 174175, 177
discounted cash flow (DCF) valuation, 387 new marketing rules, 515, 518
disposition of acquired companies, 390 UCITS Directive (Article 5 of Directive
EU Takeover Directive, 42, 385, 387, 389391 2009/65/EC), 41
hostile takeovers, 383384 alternative investment market (AIM), 486, 490
minority shareholders, elimination, 386387 American depository receipt (ADR), 202, 218
national law, AIFM directive, takeovers, and American Research and Development (ARD),
private equity funds under, 390 127, 311
poison pills, 387389, 390391 American-style waterfall provisions, 365367
publicly traded corporation acquisition, hostile Amro International, S.A. v. Sedona Corp., 411
takeovers, 383384 angel investors, 290, 513, 517
state antitakeover laws, 388391 due diligence, 293
tender offers and regulatory balancing, fallen angels, 102, 117
384385, 391 AngelList, 513
Williams Act, 383, 384385, 389, 390391 Angola, private equity attractiveness, 245, 450f
adaptive market hypothesis (AMH), 181, 191 asset class, private equity return measures, 275
adjusted present value (APV) asset liability management (ALM), 187

561
562 i n d e x

asset pricing models, cost of capital for private buyout, 258


equity, 149154 discounted cash flow (DCF) valuation, 266
capital asset pricing model (CAPM), 150, distribution to paid-in-capital (DPI), 261
151154, 155, 156, 158 dot-com bubble, 268
data, 151, 152t finding appropriate data and benchmark,
descriptive data statistics of computed indexes, 258260
151, 152t buyout and venture capital, 258
errors-in-variables and linear asset pricing capital asset pricing model (CAPM), 259
models, 154155 private equity, defined, 258
estimation of cost, decomposition of risk, relevant data, 259
155161 right benchmark, 259260
illustration of private equity risk using the mature funds, 266268, 271
capital asset pricing model, 151154 measurement methods, 260262
private equity risk using CAPM, 151154 internal rate of return, 260261
buyout (BO) index vs. S&P 500 index, 153 public market equivalent (PME), 260, 261262
growth index (GRO) vs. S&P 500 index, total value to paid-in capital (TVPI), 261
152153 net present value (NPV), 261
mezzanine index (MEZ) vs. S&P 500 PrEQIn database, 260
index, 153 risk and liquidity adjustments, 262267
private equity index vs. S&P 500 index, 154 liquidity adjustments, 264265
asset stripping, prohibition, 42 measuring gross or net of fees, 267
assets under management (AUM) risk adjustments, 263264
future of private equity, global perspective, 503, stale pricing problem, 265266
505, 507, 511512, 516 valuation challenges, 266267
institutional investors, 317 sample selection bias, 267270
at-the-market (ATM) offering, 414415 outliers and weighting, 270, 272
attractiveness index. See Country Attractiveness size bias, 269
Index skill bias, 269272
Australia survivorship bias, 267268
institutional investors, financial assets update bias, 269
held by, 323t vintage bias, 268269
private equity attractiveness, 244t venture capital, 258
return on capital (ROC) performance, 20 benefits of private equity firms, 2228, 28
avoidable costs, 1920 bankruptcy rate, default and recovery, 2627
corporate governance structure improvements,
2526
Baltics, private equity attractiveness, 460 dividends, misconceptions regarding, 28
Bangladesh, private equity attractiveness, 444 fund performance, mixed evidence of,
Bank Holding Company Act of 1956, 325, 326, 2324
381, 511 innovation and operational efficiency, 28
banking industry job creation and unemployment, 2728
Basel III Accord, 514515 value creation, 2324
new marketing rules, leveraged buyout value creation for limited partners, 2425
(LBO)s, 516 beta of listed private equity (LPE) funds,
bankruptcy 420421, 424
Lehman Brothers bankruptcy, 109, 112 book-to-market factor
Bankruptcy Act of 1978 (Chapter 11), 100, cost of capital for private equity, 152t, 155, 156,
101102, 118 157t, 158, 159t, 160t, 161t, 163t
bankruptcy rate, default and recovery, 2627 Brazil
banks, due diligence, 293 globalization trends in private equity
Basel III Accord, 514515 industry, 509
behavioral aspects private equity attractiveness, 245, 441, 443,
underpricing issued securities, 207 444f, 449f, 451f, 452456, 460
behavioral finance, 190191 private equity returns, cross-country evidence,
core-satellite method, 181, 190191 241, 242, 245
benchmark biases in private equity performance, 3G (Brazilian private equity firm), 36, 80
1011, 257272 BRICs (Brazil, Russia, India, and China)
i n d e x 563

globalization trends in private equity cost of capital for private equity, 150, 151154,
industry, 509 155, 156, 158
private equity attractiveness, 245, 441, 443, liquidity issues during global financial crisis,
444f, 449f, 451f, 452456, 460 173174
private equity returns, cross-country evidence, valuing private equity, 133, 135
241, 242, 245 capital calls
Buffet, Warren, 3, 15, 36, 103 liquidity, 169
Burundi, private equity attractiveness, 244, liquidity issues during global financial crisis,
245f, 450f 171, 172173, 176177
business development company (BDC), 490 capital cash flow (CCF) method, 131
business model capital expenditures (CAPEX)
ability of countries to support. See Country due diligence, 303
Attractiveness Index leveraged buyouts (LBOs), 353t, 355t
generally. See economics of private equity valuing private equity, 125, 136t138t
business model structure capital for private equity, cost of. See cost of capital
buyout boom 20032007, European market for private equity
developments, 37 Carnegie Steel Company, 16
buyout index (BO) carried interest, 17
estimation of cost, decomposition of risk, 156, compensation structure of private equity funds,
158, 159t 363367, 369
S&P 500 index vs., 153 funds-of-funds, fee structure of private equity
buyout private equity firms, 116117 funds, 468, 470, 478
buyout process, parties involved, due diligence, leveraged buyout (LBO) funds, 69, 333
292293 leveraged buyouts (LBOs), 34
buyouts proposed tax change, 515, 519
benchmark biases in private equity cash-on-cash multiples, 275276
performance, 258 catch-up potential, 443, 460
defined, 3, 465 Cede & Co. v. Technicolor, Inc., 387
public-to-private transactions, 465 center of main interest (COMI), 115
types of certainty equivalent level of final wealth concept,
leveraged buyouts. See leveraged buyouts 471, 477478, 481
(LBOs) certification hypothesis, 203, 205206, 229
MBOs. See management buyouts (MBOs) Chad, private equity attractiveness, 244, 245f, 450f
SBOs. See secondary buyouts (SBOs) chicken and egg problem, causality
(sponsor-to-sponsor buyout) determinations, 447
buy-to-hold (buy-to-keep), 182, 183, 184 Chile
buy-to-sell, 182, 183, 184, 190 globalization trends in private equity industry,
509
private equity attractiveness, 448, 449f, 451f,
California Public Employees Retirement Systems 452, 460
(CalPERS), 173 China
Canada private equity attractiveness, 245, 441, 443,
institutional investors, financial assets 444f, 449f, 451f, 452456, 460
held by, 323t private equity industry, 507508, 509
private equity attractiveness, 244, 248 private equity returns, cross-country evidence,
return on capital (ROC) performance, 20 241, 242, 245
capital assembly, regulation of, 379382 renminbi (RMB) funds, 508
equity capital, 380381 socialist capital, role of, 508
fund managers and compensation, socialist economy, capitalist characteristics,
381382 507508
issuance of debt, 382 structural weaknesses, 507508
private equity fund structure for, 378379 venture capital, 507508
Sun Capital Partners III, LP v. New England China Securities Regulatory Commission
Teamsters & Trucking Indus., 379 (CSRC), 223
capital asset pricing model (CAPM) closed-end fund puzzle, 492, 493494
benchmark biases in private equity closely held business acquisition, 382383
performance, 259 acquisition regulation, 382383
564 i n d e x

closely held business acquisition (continued) corporate takeover defenses


DeBaun v. First Western Bank and Trust Co., 383 acquisition regulation, 387389
Perlman v. Feldman, 383 poison pills, 387391
cluster analysis, 195 state antitakeover laws, 389391
Code Red, Marine Corps, 239 cost of capital for private equity, 9, 147164
coercive exchange offer, 116 abnormal returns, 149, 151, 156
collateralized debt obligation (CDO), 88 asset pricing models, 149154
collateralized loan obligation (CLO), 115 capital asset pricing model (CAPM), 150,
Colombia 151154, 155, 156, 158
globalization trends in private equity industry, 509 data, 151, 152t
private equity attractiveness, 451f, 452, 460 descriptive data statistics of computed
commercial mortgage-backed securities (CMBS), indexes, 151, 152t
8688, 9091, 9395, 9798 errors-in-variables and linear asset pricing
commercial real estate (CRE) financing. See models, 154155
mezzanine commercial real estate (CRE) estimation of cost, decomposition of risk,
financing 155161
commitment, institutional investors, 321 illustration of private equity risk using the
committed capital, 316317 capital asset pricing model, 151154
commonality-in-liquidity effect, 166 private equity risk using CAPM, 151154
Companies Act of 1981, 240 buyout (BO) index vs. S&P 500 index, 153
compensation structure of private equity funds, growth index (GRO) vs. S&P 500 index,
12, 360374 152153
American-style waterfall provisions, 365367 mezzanine index (MEZ) vs. S&P 500
capital assembly, regulation of, 381382 index, 153
carried interest, 363367, 369 private equity index vs. S&P 500
deal-level fees, 368 index, 154
dot-com bubble, 374 book-to-market factor, 152t, 155, 156, 157t,
European-style waterfall provisions, 366, 367 158, 159t, 160t, 161t, 163t
fees and expenses, 367 buyout index (BO)
fund-of-funds. See funds-of-funds, double layer estimation of cost, decomposition of risk,
of fees 156, 158, 159t
hurdle rate (preferred return), 69 S&P 500 index vs., 153
illustration of profit-sharing between limited capital asset pricing model (CAPM), 150,
partners and general partners, 365f 151154, 155, 156, 158
leveraged buyout (LBO) funds, 69 Dagenais and Dagenaiss higher moment
private equity firms, criticisms, 18 estimator (DDHME), 149, 154155,
2-and-20 standard compensation structure, 360, 162
467, 468 estimation of cost, decomposition of risk,
waterfall provisions, 363, 365367 155161
confidentially marketed public offerings buyout index (BO), 156, 158, 159t
(CMPOs), 415416 growth index (GRO), 156, 157t
corporate combinations, eliminating minority mezzanine index (MEZ), 158, 160t
shareholders, 386387 private equity index, 158, 161t
Cede & Co. v. Technicolor, Inc., 387 Fama-French three-factor model, 149, 150, 155,
Santa Fe Industries v. Green, 386 156, 157t, 158, 159t, 160t, 161t, 163t
Singer v. Magnavox, 386 growth index (GRO)
Weinberger v. UOP, 386 estimation of cost, decomposition of risk,
corporate governance 156, 157t
benefits of private equity firms, 2526 S&P 500 index vs., 152153
endogeneity illiquidity and the cost of equity in private
private equity firms, 26 equity, 162, 163t
VC-backed IPOs, monitoring investments, liquidity premium, 150, 155, 156, 158, 162
210 mezzanine index (MEZ)
leveraged buyouts (LBOs), governance estimation of cost, decomposition of risk,
intervention and management monitoring 158, 160t
mechanisms, 76 S&P 500 index vs., 153
organizational culture, 301302 modified internal rate of return (MIRR), 148
quality of legal system, 446 private equity index
i n d e x 565

estimation of cost, decomposition of risk, investor protection and corporate


158, 161t governance, quality of legal system,
S&P 500 index vs., 154 446
public market equivalents (PMEs), 148 tax regimes, impact of, 446
stale valuation, 149, 150, 162 private equity country maturity and historic
Country Attractiveness Index, 243246, 448459. private equity returns, 444, 451452,
See also specific country 457459
attractiveness ranking, 448449 ranking
BRICs, 452456 changes in ranking, medium-term trends,
country attractiveness index values 450452
(20102014), 245t medium-term trends in attractiveness
Indonesia, Mexico, the Philippines, and ranking, 449452
Turkey, following the path of BRICs, private equity attractiveness ranking, 448449
453455 private placement memoranda (PPM),
key drivers of BRICs private equity 457458
attractiveness, 452, 453f top-down approach to ranking, 445
country maturity scores and historic private top three and bottom three countries for
equity returns, 458, 459f private equity attractiveness, graph, 245f
current situation, 448459 upward bias, 458
deal flow, 444, 445, 451, 452, 455 crowdfunding
factors, 243244 anti-fraud measures, 514
historic private equity returns, 444, 451452, defined, generally, 513
457459 IPO On-Ramp provision ( JOBS Act), 201, 514
Indonesia, Mexico, the Philippines, and Turkey lending model, 514
BRICs, following the path of, 453455 philanthropy model, 514
key drivers of private equity attractiveness, reward-based model, 514
454, 455f CTS Corp. v. Dynamics Corp. of Am., 389
minimum maturity level for private equity cultural due diligence, 301302
activity, 454455, 456f national culture, dimensions, 302
least attractive countries for private equity, organizational culture, 301302
244245, 450f cumulative abnormal returns (CARs), 24, 403
medium-term trends in attractiveness ranking, customer relationship management (CRM), 518
449452
attractiveness rank changes, 450452
country private equity attractiveness score, Dagenais and Dagenaiss higher moment
448, 450f estimator (DDHME), 149, 154155, 162
minimum maturity level for private equity data management, 518
activity, 442, 455457 deal-level fees, 368
attractiveness scores and private equity death-spiral transactions and SEC enforcement,
activity of 118 countries, 456, 457f 411413, 417
highly granulated determinants of DeBaun v. First Western Bank and Trust Co., 383
attractiveness for Indonesia, Mexico, debt capital markets and leveraged buyout (LBO)
the Philippines, and Turkey, 454455, funds, 7173, 81
456f debt tax shield, 17, 19
most attractive countries for private equity, Dell Inc., 36, 71
244, 441 denominator effect (portfolio composition),
private equity business model, ability of 172, 188
countries to support, 444448, 459460 discounted cash flow (DCF) valuation, 124, 557
chicken and egg problem, causality acquisition regulation, 387
determinations, 447 alternative, multiples-based valuation method,
key drivers of private equity maturity, 131132
determinants, 445448, 459460 benchmark biases in private equity
depth of the capital market, bank centered performance, 266
and stock market centered, 445 cash flows to estimate value, 127129
economic activity, 445 valuing a potential private equity investment
entrepreneurial culture and deal (valuation example), 133, 134t
opportunities, innovation, 447 private equity ownership assumptions,
human and social environment, 446447 138t139t, 140
566 i n d e x

disposition of acquired companies, 391392 cluster analysis, 195


distressed debt investments, 8, 100118 diversification level, 192193
defined, 3 foreign exchange, 194
dry powder, 113, 117 market timing, 193194
fallen angels, 102, 117 natural hedges, 194
generally, 104106 reducing interdependence, 194195
Harrahs Entertainment, 17t, 116 risk profile of venture capital, 192, 193f
investing process, illustration of, 105106 dividend recapitalization
investment strategies, 106110 exit strategies, leveraged dividend
distressed for control, 106108 recapitalization, 222, 231, 233t
loan-to-own, 108 leveraged buyouts (LBOs), 7677, 81
substantive consolidation, 109 dividends
turnaround investing, 109110 misconceptions regarding, 28
junk bond market, 100, 102, 104 private investment in public equity (PIPE),
outlook for distressed private equity investing, investor protections, 405
117118 document management systems, 518
PrEQIn database, 101, 103 Dodd-Frank Act, 38, 39, 381, 510512, 519
TargetCo, debt capital structure of, 105106, adviser registration, requirements, 511512
107, 108, 109, 110, 116 firm registration requirements and exemptions,
distressed debt private equity market SEC, Title IV of Dodd-Frank, 39
current state of market, 110117 foreign private adviser, defined, 39
acquisition debt multiples, 111, 113f fraud prevention efforts, 512
buyout private equity firms, 116117 gatekeepers, 512
coercive exchange offer, 116 liquidity issues during global financial crisis,
distressed private equity funds raised in 174175
North America, 111112 proposed approaches for future financial
European market, 113116 stability, 512
geographic expansion of distressed investing, reporting and recordkeeping of operational
113116 information, 512
leveraged finance default rates, 111112, 113f Section 113, nonbanks systemically important
low rated high yield issuance, 19952009, financial institutions, 39
111, 112f Volcker Rule (Section 619 of Dodd-Frank Act),
merger and acquisition transaction value 38, 39, 40, 381, 510511, 519
(19952009), 111, 112f dot-com bubble, 32
vulture investors, 116 benchmark biases in private equity
future outlook, 117118 performance, 268
historical evolution, 101104, 118 compensation structure, 374
Bankruptcy Act of 1978 (Chapter 11), 100, distressed debt private equity market, 110, 112
101102, 118 initial public offerings (IPOs), 208
early high yield market, size, 102 liquidity issues during global financial
median distressed fund return, 103104, 112 crisis, 169
tech bubble (dot-com bubble), 110, 112 private equity market and regulatory
Lehman Brothers bankruptcy, 109, 112 developments, post-financial crisis, 32
distressed for control investment strategy, 106108 value creation, 331
distressed (turnaround) investments, defined, 3 venture capital in Europe, 58
distributions, leveraged buyout (LBO) funds, 70 dry powder (uncalled committed capital), 221,
distribution to paid-in-capital (DPI), 261, 275 233t, 361362, 505
distribution waterfalls. See waterfall provisions defined, 18
diversification distressed debt investments, 113, 117
funds-of-funds, 14, 463481 leveraged buyouts (LBOs), 70, 72, 81
global private equity as diversifying agent, due diligence, 11, 290306
252254 accountants, 293
institutional investors, 320321 angel investors, 290, 293
private equity returns, cross-country evidence banks, 293
of diversification benefits, 243 banks and bank lawyers, 293
diversification management, 192195 by general partners, 292
average fund performance per vintage year capital expenditures (CAPEX), 303
(1990 to 2000), 193f challenges of due diligence, 304305
i n d e x 567

current process, 304305 structural framework of process, 294295, 295


limited partners, challenges faced in sub-sector, finding, 295
conducting, 305 target evaluation, 300
conducting due diligence, 300 types of, 295304, 296t
conducting through rough due diligence, 295 commercial due diligence, 296297
current due diligence process, 304305 cultural due diligence, 301302
definition, 290 financial due diligence, 300301
documentation, drafting, 298 human due diligence, 301
environmental auditors, 293 information technology due diligence, 303
financing a buyout, 293294 legal due diligence, key issues, 298299
finding target, 300 operational due diligence, 297
identification of potential target, 295 special due diligence, 303304, 304t
incorporation, 300 strategic due diligence, 299300
individualism vs. collectivism, 302 tax due diligence, 303
indulgence vs. self-restraint, 302 uncertainty avoidance, 302
investment bankers, 293 understanding of business and, 298
lawyers, managements, 293 valuing target company, 298
limited partners
challenges faced in conducting due
diligence, 305 economics of private equity business model
on general partners, 291292 structure, 7, 1628
long-term orientation, 302 avoidable costs, 1920
Madoff, Bernard, 297 background, 1922, 503, 518
management buyouts (MBOs) of public carried interest, 17
companies, 293 cumulative abnormal returns (CARs), 24
managements lawyers, 293 debt tax shield, 17, 19
management team, 292 dry powder, defined, 18
management team, finding, 292 financial crisis of 20072008, 18
masculinity vs. femininity, 302 firm value, 1920
meaning and importance of due diligence in poorly performing publicly listed companies
private equity, 291294 in different financial markets ( January
by general partners, 292 2014), 1920
environment, social, and governance (ESG), first formal private equity deal, 16
292 general partners, 2122
financing a buyout, 293294 going private, 21, 24, 28, 32
limited partners on general partners, 291 growth of private equity industry, 18
292 largest buyout deals, 16, 17t
parties involved in the buyout process, leveraged buyout (LBO) process, 21
292293 limited partners, 2122
national culture, dimensions, 302 Longitudinal Business Database (LBD), 24
organizational culture, 301302 management buyouts (MBOs) of public
organizational skills, 292 companies, 18f, 2324, 27
parties involved in the buyout process, role of, Obama administration, 17
292293 private equity firms
post-failure, 292 benefits of private equity firms, 2228
power distance, 302 compensation schemes, 18
private equity provider and providers criticisms of private equity firms, 1618
lawyers, 293 firm value, 1920
process of due diligence, summary, 304t poorly performing publicly listed
PR providers, 293 companies in different financial
sector, identification of, 295 markets ( January 2014), 1920
sourcing the deal, 292 private equity funds
steps in due diligence process, 294 background, typical structure, 2122
data analysis phase, 294 mixed evidence of fund performance,
data collection, 294 2324
planning phase, 294 structure of funds, 2122
report finalization, 294 top 10 largest private equity funds, 22, 23t
strategy formation, 299 typical structure, 2122
568 i n d e x

economics of private equity business model Ernst & Young survey


structure (continued) institutional investors, perceptions regarding
private equity industry, growth of value of private equity and VC-backed
background, 18 IPOs, 209
number of closed or effective leveraged estimation, cost of capital for private equity
buyouts (LBOs) or management decomposition of risk, 155161
buyouts (MBOs) (1970 to 2013), 18f buyout index (BO), 156, 158, 159t
top 10 largest private equity buyouts, ranked growth index (GRO), 156, 157t
by inflation-adjusted deal value in 2014 mezzanine index (MEZ), 158, 160t
dollars, 16, 17t private equity index, 158, 161t
top 10 largest private equity funds, 22, 23t Estonia, private equity attractiveness, 449f,
Edgar v. Mite Corp., 389 451f, 452
Egypt, private equity attractiveness, 444, 451f ethical framework, 392393
emerging markets. See also Country Attractiveness Ethiopia, private equity attractiveness, 444, 450f
Index; international and emerging markets Europe
globalization trends in private equity industry, economic growth and private equity
509510 investments, 505507
BRICs, 509 European debt crisis, impact, 506
investment benefits, 509510 exiting pre-financial crisis investments,
Southeast Asia, 509510 506507
Sub-Saharan Africa (East Africa), 509510 financial crisis of 20072008, 505506
Emerging Markets Private Equity Association innovation, new firm creation and patents, 506
(EMPEA), 504 leveraged buyouts (LBOs), impact on job
endogeneity creation and reduction and operational
leveraged buyouts (LBOs), 357 gains, 506
private equity firms, 26 macroeconomic uncertainties, 507
VC-backed IPOs, monitoring investments, 210 sovereign wealth funds, 507
endowments, 189190 leveraged buyouts (LBOs)
Harvard Management Corporation continental European leveraged buyouts
(HMC), 173 (LBOs), transaction levels, 345
Yale model (endowment model), 169, 171, history of, 331332
173174, 177 listed private equity. See listed private equity
Energy Future Holdings (TXU), 16, 17t (LPE) and institutional investors
enterprise-value-to-EBITDA (EV/EBITDA) Listed Private Equity Association (LPEQ), 423,
global exit activities, value generation before 424, 426, 427, 438, 484, 495
exit, 217 market developments, post-financial crisis, 35,
leveraged buyout (LBO) funds, 7273 3738, 43
median leveraged buyout (LBO) debt levels buyout boom 20032007, 37
and EBITDA multiples over time in European Central Bank, 2013 monetary
United States, 71, 72f policy, 38
leveraged buyouts (LBOs), 356, 357f recovery and statistics, European buyout
continental European leveraged buyouts market, 3738
(LBOs), transaction levels, 345 sovereign debt crisis, 33, 35, 36, 37, 38, 43
multiples in leveraged buyout deals, private equity attractiveness, 240241, 244,
348, 349t 246, 248
operating performance, impact on, 352 STOXX Europe Private Equity 20 index, 485
value creation, 336f337f Europe, venture capital, 7, 4963
valuing private equity, 123, 131132, 133 dot-com bubble, 58
trading multiples, 133, 134t, 140 European venture capital industry
transactions multiples, 133 establishing stable community of venture
entrepreneurial culture and deal opportunities, 447 capital investors, 5051
environment, social, and governance (ESG) European Private Equity and Venture Capital
due diligence, 292 Association (EVCA), 50, 62
private equity portfolio management, 188 internationalization of venture capital
environmental auditors, due diligence, 293 practices, 49
equity capital, regulation of, 380381 professionalization of European venture
equity crowdfunding. See crowdfunding capital practices, 5053, 6263
i n d e x 569

selection bias, 5354 private equity returns, cross-country


money-chasing deal phenomenon, 59 evidence, 242
performance comparisons, Europe and United regulatory developments, post-financial crisis,
States, 5962 38, 4142, 43
comparative performance of venture capital alternative investment fund (AIF),
funds in Europe and United States, defined, 38
5961 alternative investment fund manager
IPO exit rates, 60 (AIFM), 38, 41
toward a convergence of venture capital Alternative Investment Fund Managers
practices in Europe and United States, Directive (AIFMD), 33, 38, 4142
6162 UCITS Directive (Article 5 of Directive
performance of funds, 5458 2009/65/EC), 41
J-curve effect, 5455 asset stripping, prohibition, 42
mature funds, 57 EU Takeover Directive, 42
principal-agent problem, 55 systemic risk, 41, 42
smoothing bias, 57 transparency and control issues, 4142
Thomson Venture Economics (TVE), 5758, European venture capital industry
268269, 281t establishing stable community of venture
performance of investments, 5354 capital investors, 5051
persistence of funds performance, 5859 European Private Equity and Venture Capital
professionalization of European venture capital Association (EVCA), 50, 62
practices, 5053, 6263 internationalization of venture capital practices,
public market equivalents (PMEs), 56 49
selection bias, 5354 professionalization of European venture capital
successful exits, 54 practices, 5053, 6263
venture capital funding, generally, 5359 increasing professionalism, 6162
European Central Bank, 2013 monetary policy, 38 negative aspects of investors stability, 51
European debt crisis, impact, 506 positive aspects of investors stability, 5051
European market professionalizing investors practices, 5253
developments, post-financial crisis, serial entrepreneurship in Europe, 5152
3738, 43 selection bias, 5354
distressed debt private equity market, 113116 Euro Zone, financial assets held by institutional
center of main interest (COMI), 115 investors, 323t
collateralized loan obligation (CLO), 115 EU Takeover Directive, 42, 385, 387, 389391
forum shopping, 115 EV/EBITDA
fundraising for European distressed-asset leveraged buyouts (LBOs), 356, 357f
private equity, 115116 continental European leveraged buyouts
geographic expansion of distressed investing, (LBOs), transaction levels, 345
113116 multiples in leveraged buyout (LBO) deals,
leveraged buyout (LBO) volume, 113114 348, 349t
role of government, cultural distinctions, operating performance, impact on, 352
114115 evolution of private equity, geography and
European Private Equity and Venture Capital globalization trends, 503, 518
Association (EVCA), 50, 62, 173, 186, ex-ante performance evaluation, 285286
465, 505 exchange-traded funds (ETFs), historical
European-style waterfall provisions, 366 performance, 248
European Union exit strategies, 10, 215233
Alternative Investment Fund Managers agency costs, 222, 232
Directive (AIFMD), 33, 38, 4142 asset stripping, prohibition, 42
acquisition regulation, 390, 391 California Public Employees Retirement
global regulatory and ethical framework, 378, Systems (CalPERS), 173
380, 381382 choice of exit route, factors affecting, 230233
liquidity issues during global financial crisis, financial economics approach, 232, 233t
174175, 177 information asymmetry and pecking order
new marketing rules, 515, 518 theory, 232
UCITS Directive (Article 5 of Directive macroeconomic conditions, 231
2009/65/EC), 41 risk management theory, 232
570 i n d e x

exit strategies (continued) leveraged dividend recapitalization, 222, 231,


using representation and warranty insurance, 233t
231232 number of global buyout-backed exits by
decline in exit activity, 172173 channel (1995 to 2013), 218, 219f
Europe private equity-backed exits by type (2006 to
exiting pre-financial crisis investments, 2013), 219220
506507 secondary buyouts (SBOs) (sponsor-to-
venture capital Europe, successful exits, 54 sponsor buyout), 37, 221222, 233t
examples of exiting private equity investments, trade sale, 220, 233t
173
financial economics approach, choice of exit
route, 232, 233t Facebook IPO, 225
information asymmetry and pecking order failure, post-failure due diligence, 292
theory, 232 fallen angels, 102, 117
risk management theory, 232 Fama-French three-factor model, 149, 150, 155,
global exit activities, 216218 156, 157t, 158, 159t, 160t, 161t, 163t
EBITDA/earnings growth, 217 Federal Deposit Insurance Corporation
free cash flow generation, 217 (FDIC), 510
multiple expansion, 217218 Federal Reserve, 2013 monetary policy, 36
number of global buyout-backed exits by fees. See also compensation structure of private
region (1995 to 2013), 216 equity funds
number of global buyout exits determinants of fee structure, 371372
(2009 to 2013), 216 distortion of fees and incentive distortion, 373
types of financial instruments, 218 management fees, 361362
value generation before exit, 217218 measuring gross or net of, 267
Harvard Management Corporation (HMC), 173 relationship between fees and fund
information asymmetry, 215, 231, 232, 233t performance, 373
initial public offerings (IPOs) time series dynamics and the future of fee
PE-backed IPOs. See initial public offerings structure, 373374
(IPOs) Financial Crisis Inquiry Commission (2011),
VC-backed IPOs. See VC-backed IPOs 8788
institutional investors exiting private equity financial crisis of 20072008, 18, 3233, 35, 36, 37
investments, examples, 173 global crisis and liquidity. See liquidity issues
leveraged buyouts (LBOs) during global financial crisis
choice of exit channels by (19902013), housing market crash, 32
7879 international and emerging markets, private
exit phase, 7780 equity growth, 442, 443
investment phase categories, cash multiples leveraged buyouts (LBOs), 67, 72
of leveraged buyouts (LBOs) by exit post-financial crisis, 3435
channel (19902013), 79 post-financial crisis
leveraged dividend recapitalization, 222, 231, leveraged buyouts (LBOs), 3435
233t liquidity issues. See liquidity issues during
macroeconomic conditions, 231 global financial crisis
pecking order theory, 232, 233 market developments. See market
post-financial crisis, 35, 42 developments, post-financial crisis
regulatory developments, post-financial crisis, regulatory developments. See regulatory
35, 42 developments, post-financial crisis
representation and warranty (R&W) insurance, financial due diligence, 300301
231232 Financial Industry Regulatory Authority
risk management theory, 232 (FINRA), 514
sales, post-financial crisis, 33, 36 financial institution, defined, 314
secondary buyouts (SBOs) (sponsor-to- financing a buyout, due diligence, 293294
sponsor buyout), 37, 221222, 233t Finland, private equity attractiveness, 451f, 452, 460
trade sale, 220, 233t firm value, 1920
traditional exit routes, 218222, 233t first formal private equity deal, 16
initial public offering (IPO), 221, Foreign Account Tax Compliance Act (FATCA),
222230, 233t 515, 518
i n d e x 571

foreign exchange, 194 limited partnership agreement (LPA), fee


foreign private adviser, defined, 39 structuring and, 467, 468, 470, 478
Form F-3 (2008 amendments), 413, 414t catch-up clause, 468, 469t, 470
Form PF, adviser registration, 511512 multiples, funds-of-funds performance
Form S-3 (2008 amendments), 413, 414t measurement, 470
fraud prevention efforts, Dodd-Frank Act, 512 net of fees, 469t, 470
free cash flow (FCF), 330 pooled fund-of-funds fees, 468
FTSE EPRA/NAREIT Global Real Estate Index, 2 and 20 management compensation
250251, 252, 253, 254f framework, 360, 467, 468
funding liquidity risk, 169 general partners, 465, 466468, 470
fund managers. See also compensation structure of investment in funds-of-funds, 463464
private equity funds managerial compensation, criticism, 463
compensation, regulation of capital assembly, principal-agent problem, defined, 463
381382
private benefits, 494
fund performance, mixed evidence of, 2324 general partners (GPs), 2122
fundraising, leveraged buyout (LBO) funds capital calls, exclusive right to make, 172173
levels in 2013 by fund type, 67, 68f due diligence, 291292
volume, statistics, 67, 6970 funds-of-funds - fee structure, 465,
funds-of-funds, double layer of fees, 14, 463481 466468, 470
data set, computation of certainty levels of liquidity, generally, 168169
wealth, 471479 liquidity issues during global financial crisis,
descriptive analysis, 471474 171, 172173
downward risk, 471472 private equity funds, structure of, 2122
geographical distribution, 472t, 473, 474t profit-sharing between limited partners and
kernal density estimations, 472, 473t, 475t general partners, illustration of, 365f
liquidated funds-of-funds, 471, 473t, generic buyouts, 465
478t, 479 Germany
mature funds, 473, 474t, 476, 477t, private equity attractiveness, 240, 242, 244t
478t, 479 Super Return German Private Equity
mature funds-of-funds, 478t, 479 Summit, 506
risk-return ratio, 427t, 471, 473, 474t, 476, globalization trends in private equity industry,
477t, 479, 480t 505510
PrEQIn database, 471, 472t China, 507508
results, 478479 key concerns, structural weaknesses,
levels of indifference, 477479, 481 507508
net present value (NPV), 478 renminbi (RMB) funds, 508
simulating funds-of-funds, 474478 socialist capital, role of, 508
bootstrap technique, 474477, 479 socialist economy, capitalist characteristics,
certainty equivalent level of final wealth 507508
concept, 471, 477478, 481 venture capital, 507508
methods, 474 emerging markets, 509510
upward bias, 475t BRICs, 509
double layer of fees, 466, 467, 468, 468t, 470, investment benefits, 509510
471, 479 Southeast Asia, 509510
fees-on-fees, 464, 467, 471, 481 Sub-Saharan Africa (East Africa), 509510
fee structure of private equity funds and funds- Europe, economic growth and private equity
of-funds, 467470 investments, 505507
calculation of fees, 467470 European debt crisis, impact, 506
carried interest, 468, 470, 478 exiting pre-financial crisis investments,
determining management fees, calculation 506507
methods, 467468 financial crisis of 20072008, 505506
fixed and variable calculation components, innovation, new firm creation and patents,
467 506
gross of fees, 468, 470, 478 leveraged buyouts (LBOs), impact on job
hurdle rate, 468, 470 creation and reduction and operational
internal rate of return (IRR), 468 gains, 506
572 i n d e x

globalization trends in private equity industry estimation of cost, decomposition of risk,


(continued) 156, 157t
macroeconomic uncertainties, 507 S&P 500 index vs., 152153
sovereign wealth funds, 507
Latin America, 509
sovereign wealth funds, 507508, 519 Harrahs Entertainment, 17t, 116
Global Listed Private Equity (GLPE), 248 Harvard Management Corporation (HMC), 173
global private equity herding behavior, institutional investors, 318
activity, data on, 246248 Hiring Incentives to Restore Employment Act
buyout deals by geographic region (HIRE Act), 515
(20062013), 247f history of private equity
buyout deals by geographic regions evolution of private equity, geography and
(2006 vs. 2013), 248f globalization trends, 503, 518
buyouts as a proportion of their 2006 value, first formal private equity deal, 16
247f fundraising (20082013), 171172
as diversifying agent, 252254 largest buyout deals, 16, 17t
listed private equity (LPE) correlation mega deals, 36
coefficients, 2008-2013, 253t top 10 largest private equity buyouts, ranked by
PowerShares Global Listed Private Equity inflation-adjusted deal value in 2014
Portfolio (PSP), 4, 248, 249250 dollars, 16, 17t
correlation coefficient compared with H.J. Heinz, 36, 71
other assets, 254f holding period, 34
correlation coefficients (20082013), 253t home bias, listed private equity (LPE) and
Global Real Estate Index, FTSE EPRA/NAREIT, institutional investors, 420
250251, 252, 253, 254f Hong Kong, private equity attractiveness, 244t,
global regulatory framework, 12, 377393. See also 250, 449f, 451f, 452
acquisition regulation hostile takeovers, 208, 383384
accredited investors, 380381 hot issue markets, 207208
Alternative Investment Fund Managers human due diligence, 301
Directive (AIFMD), 378, 380, 381382 hybrid mezzanine investment, 93
Bank Holding Company Act of 1956, 381
capital assembly, regulation of, 379382
equity capital, 380381 identification of potential target, 295
fund managers and compensation, 381382 incentives
income taxation and, 378379 compensation structure, size of GP
issuance of debt, 382 incentives, 372
Sun Capital Partners III, LP v. New England distortion, 373
Teamsters & Trucking Indus., 379 income taxation. See taxation
disposition of acquired companies, 391392 India
ethical framework, 392393 globalization trends in private equity
EU Takeover Directive, 385 industry, 509
investment advisers, 380381 private equity attractiveness, 245, 441, 443,
Investment Advisers Act (Advisers Act), 380 444f, 449f, 451f, 452456, 460
Investment Company Act of 1940, 380 private equity returns, cross-country evidence,
Regulation D (Rule 501), 380 241, 242, 245
restructuring the acquired companies, 391 individual investors, distinguished from
tender offers and regulatory balancing, institutional investors, 318319
384385 individualism vs. collectivism, 302
Volcker Rule, 381 Indonesia
Williams Act, 384385 private equity attractiveness, 441, 444, 451f,
going private, 21, 24, 28, 32 452, 453455, 456f, 460
golden age of private equity, 35 indulgence vs. self-restraint, 302
grandstanding hypothesis, 205206, 229 information asymmetry
greenmail, 388 VC-backed IPOs
greenmarker, 388 monitoring investments, 203, 209
Grobow v. Perot, 389 underpricing issued securities, 206, 208
growth capital investments, 313 information cascades argument, 207
growth index (GRO) information technology due diligence, 303
i n d e x 573

initial empirical evidence on return persistence, determinants of institutional investment,


279281 319322
initial public offerings (IPOs), 221, 222230, 233t, agency costs, 320
241. See also VC-backed IPOs agency problems, 319320
arguments favoring IPO as exit route, commitment, 321
229230 committed capital, 321
Europe, venture capital concentrated ownership, 319
performance comparisons, Europe and controlling shareholders and dilution of
United States IPO exit rates, 60 control, 319320
successful exits, 54 diversification, 320321
exit strategies, post-financial crisis, 33, 34, institutional affiliation of private equity fund,
36, 38 321322
Facebook IPO, 225 performance of funds, variation, 322
IPO defined, 221 PrEQIn database, 322
IPO process, 225, 227 recommitment, 321
lock-up periods, 221, 231 sophistication differences, 322
offering market outlook, 223225 direct investments, access to funds, 314
aggregate deal value of PE-backed IPOs by drawdowns (capital calls), 311, 316, 326
region of portfolio companies (2005 to Ernst & Young survey, perceptions regarding
June 2014), 223, 226t value of PE and VC-backed IPOs, 209
aggregate value of PE-backed IPOs by region financial institution, defined, 314
of portfolio company ( January 2005 to growth capital investments, 313
June 2014), 223, 227f herding behavior, 318
PE-backed IPOs and stock market evolution individual investors, distinguished from
(2005 to June 2014), 223, 224t institutional investors, 318319
PE-backed IPOs (volume) and market investor behavior, 318319
conditions ( January 2005 to June herding behavior, 318
2014), 223, 225f individual investors, distinguished from
volume and aggregate value of PE-backed institutional investors, 318319
IPOs ( January 2005 to June 2014), syndicating investments, 319
223, 225f investor trends, post-financial crisis of
yearly aggregate deal value at the completed 20072008, 516517
date per portfolio company by co-investment, defined, 517
industrial sector (2005 to June 2014), leveraged buyouts (LBOs), 313
223, 225, 228t limited liability companies (LLCs), 316
yearly deal value at completed date per limited partnerships, 316317
industry sector, 223, 225, 229f listed private equity (LPE), allocation of capital
post-financial crisis, 33, 34, 36, 38 to. See listed private equity (LPE) and
reasons to initiate an IPO, 227, 229230 institutional investors
VC-backed IPOs. See VC-backed IPOs mezzanine capital investments, 313
innovation model investment structure of private
at-the-market (ATM) offering, 414415 equity, 315f
benefits of private equity firms, 28 number of private equity deal flows
confidentially marketed public offerings (20042013), 312
(CMPOs), 415416 pension funds, 311, 314, 323325
entrepreneurial culture and deal opportunities, defined benefit plans, 323234
447 defined contribution plans, 323234
new firm creation and patents, 506 largest retirement funds, 324
small firms, alternative financing, 412, 412t PrEQIn database, 326
20 percent rule, 416 primary motivation to invest in private equity,
institutional investors, 1112, 311326 325326
accredited investors, 312 principal-agent problem, 317
American Research and Development Private Equity Growth Capital Council
(ARD), 311 (PEGCC), 314
assets under management (AUM), 317 private equity market, growth statistics,
Bank Holding Company Act of 1956, 325 311312
committed capital, 311, 316317 private equity portfolio management, 182, 183,
comparison of private and public equity, 313 184, 186187, 188, 190, 198
574 i n d e x

institutional investors (continued) historic private equity returns, 444, 451452,


role and influence of, 323 457459
Small Business Act of 1956, 325 large populations, 443, 453, 460
structure of private equity, 314317 new millennium, private equity activity
assets under management (AUM), 317 investment volume, 442444
blind pools, 314 global investment volume, 442
channeling investments, 315316 investment volume differentiated by region,
committed capital, 316317 442443
direct investments, access to funds, 314 investment volume in BRICs, 443, 444f
drawdowns (capital calls), 311, 316, 326 private equity business model, ability of
financial institution, defined, 314 countries to support, 444448, 459460
limited liability companies (LLCs), 316 chicken and egg problem, causality
limited partnerships, 316317 determinations, 447
model investment structure of private equity, key drivers of private equity maturity,
315f determinants, 445448, 459460
principal-agent problem, 317 depth of the capital market, bank centered
private equity funds, typical formation of, and stock market centered, 445
316 economic activity, 445
Private Equity Growth Capital Council entrepreneurial culture and deal
(PEGCC), 314 opportunities, innovation, 447
restrictive covenants, 316317 human and social environment, 446447
signaling model, 317 investor protection and corporate
syndicating investments, 319 governance, quality of legal
types of funds available, 321 system, 446
types of institutional investors in private tax regimes, impact of, 446
equity, 323326 private placement memoranda (PPM),
banks and holding companies, 325 457458
endowments and foundations, 325 traditional private equity markets, 441
financial assets of institutional investors by upward bias, 458
country, 323t International Association of Risk and Compliance
investment corporation, distinguished from Professionals (IARCP), 510
pension funds, 324325 investment advisers
life insurance policies, 324 Bank Holding Company Act of 1956, 381
VC-backed IPOs, 203, 204, 209 foreign private adviser, defined, 39
venture capital equity, 313 Investment Advisers Act (Advisers Act), 380
Institutional Limited Partners Association Investment Company Act of 1940, 380
(ILPA), 374 Principal Advisers Act, 511
international and emerging markets, 13, 441460. registration, requirements under Dodd-Frank
See also Country Attractiveness Index Act, 511512
BRICs Investment Advisers Act (Advisers Act), 39, 380,
country attractiveness, current situation, 511, 514
452453 investment bankers, due diligence, 293
Indonesia, Mexico, the Philippines, and Investment Company Act of 1940, 380, 490, 511
Turkey, following the path of BRICs, investment phase categories, leveraged buyouts
453455 (LBOs), 7477
key drivers of BRICs private equity cash multiples of leveraged buyouts (LBOs) by
attractiveness, 452, 453f exit channel (19902013), 79
new millennium, private equity activity dividend recapitalization, 7677, 81
investment volume, 443, 444f governance intervention and management
catch-up potential, 443, 460 monitoring mechanisms, 76
conditions required to increase transaction operational engineering (operational
volume, 441, 442444 excellence), 7677
country private equity maturity, 444, 451452, pre-investment phase, 74
457459 restructuring strategies in leveraged buyouts
development patterns, general statistics, 441 (LBOs), 77, 78t
financial crisis of 20072008, impact of, value creation principle, 75
442, 443 investment strategies, distressed debt, 106110
i n d e x 575

distressed for control, 106108 lawsuits from shareholders, reduction of


loan-to-own, 108 likelihood, 207
substantive consolidation, 109 lawyers, managements; due diligence, 293
turnaround investing, 109110 legal due diligence, 298299
investment trusts, 490 Lehman Brothers bankruptcy, 109, 112
investment volume. See international and emerging lending model, crowdfunding, 514
markets Lesotho, private equity attractiveness, 450f
investor protections leveraged buyout (LBO) funds
private investment in public equity (PIPE), carried interest, 69, 333
405406 debt capital markets and, 7173, 81
investor trends, post-financial crisis of 20072008, defined, 66
516518 distributions, 70
institutional investment in private equity, enterprise-value-to-EBITDA (EV/EBITDA),
516517 7273
co-investment, defined, 517 financial crisis of 20072008, impact on global
investor demands, 517518 fundraising, 67
IPO On-Ramp provision ( JOBS Act), 201, 514 focus on performance, impact, 6869
issuance of debt fundraising levels in 2013 by fund type,
capital assembly, regulation of, 382 67, 68f
fundraising volume, statistics, 67, 6970
hurdle rate (preferred return), 69
Japan life cycle, 6771
institutional investors, financial assets held by, management fees, compensation structures
323t (19902013), 69
private equity attractiveness, 244t, 249, 250 management of fund, 66
J-curve median debt levels and EBITDA multiples over
liquidity issues during global financial crisis, 170 time in United States, 71, 72f
publicly traded private equity (PTPE), net asset operation of, 332333
value (NAV) discounts, 494495, 496 relationship banking, 71, 75
venture capital in Europe, 5455 secondary buyouts (SBOs) (sponsor-to-
job creation and unemployment sponsor buyout), 70
benefits of private equity firms, 2728 syndicated loans used, total volume and average
Hiring Incentives to Restore Employment Act tranche size of, 71, 73f
(HIRE Act), 515 top quartile performance, 6869
leveraged buyouts (LBOs), Europe, 506 leveraged buyouts (LBOs), 78, 6681. See also
Jumpstart Our Business Startup Act ( JOBS Act) leveraged buyout (LBO) funds
crowdfunding, 201, 513514 advantages and disadvantages, 3435
new marketing rules, 200201, 209, 513 carried interest, 34
junk bond market, 100, 102, 104 continental European leveraged buyouts
(LBOs), transaction levels, 345
debt, primary effect on equity returns, 6667
Key Elements of Legal and Tax Regimes Optimal defined, 34, 66
for the Development of Private Equity Dell Inc., 36, 71
(EMPEA), 504 distribution waterfalls, regulation of, 80
Kohlberg, Kravis, and Roberts (KKR), 16, 17t, dry powder, 70, 72, 81
420, 486, 487f economic value creation, 74
as PEGCC member, 504 endogeneity, 357
creation of, 36, 204 Europe
first purchase, 240 continental European leveraged buyouts
new liquidity sources, 507 (LBOs), transaction levels, 345
history of, 331332
history of leveraged buyouts (LBOs), 331332
Lake Wobegon fallacy (overconfidence bias), 197 impact on job creation and reduction and
Latin America, globalization trends in private operational gains, 506
equity industry, 509 EV/EBITDA, 356, 357f
Latin American Private Equity and Venture continental European leveraged buyouts
Capital Association (LAVCA), 505, 509 (LBOs), transaction levels, 345
576 i n d e x

leveraged buyouts (LBOs) (continued) value drivers, 73


multiples in leveraged buyout deals, Warner Music Group (case study), 8081
348, 349t leveraged dividend recapitalization, 222, 231, 233t
operating performance, impact on, 352 limited liability corporations (LLCs), 5
exit phase, 7780 institutional investors, 316
choice of exit channels by (19902013), limited partnership agreement (LPA)
7879 catch-up clause, 468, 469t, 470
financial crisis of 2007-2008, 67, 72 fee structuring, 467, 468, 470, 478
post-financial crisis, 3435 limited partnerships
free cash flow (FCF), 330 capital assembly and income taxation, 378
free cash flows, 74 challenges faced in conducting due
funds-of-funds, fee structure terms and diligence, 305
characteristics, 465 institutional investors, 316317
generating returns, 66 listed private equity (LPE) and institutional
H.J. Heinz, 36, 71 investors, 419
holding period, 34 on general partners, due diligence, 291292
institutional investors, 313 limited partners (LPs), 2122
investment phase categories, 7477 benefits of private equity firms, value creation,
cash multiples of leveraged buyouts (LBOs) 2425
by exit channel (19902013), 79 capital assembly and income taxation, 378
dividend recapitalization, 7677, 81 challenges faced in conducting due diligence, 305
governance intervention and management due diligence
monitoring mechanisms, 76 challenges, 305
operational engineering (operational on general partners, 291292
excellence), 7677 economics of private equity business model
pre-investment phase, 74 structure, 2122
restructuring strategies in leveraged buyouts institutional investors, 316317
(LBOs), 77, 78t Institutional Limited Partners Association
value creation principle, 75 (ILPA), 374
leveraged buyouts process, 21 limited partnership agreement (LPA), fee
low price, 74 structuring and, 467, 468, 470, 478
mortgage effect, 67 catch-up clause, 468, 469t, 470
one-time liquid private equity funds, 80 liquidity, generally, 168
operating performance, impact on, 351356 liquidity issues during global financial
post-financial crisis, 3435 crisis, 172
pre-investment phase, 74 on general partners, due diligence, 291292
private equity market and regulatory private equity funds, structure, 2122
developments, post-financial crisis, 3435 profit-sharing between limited partners and
regulatory guidance, crackdown on risky general partners, illustration of, 365f
banking practices, 516 liquidity, 9, 166169
relationship banking, 71, 75 capital calls, 169
reversed leveraged buyout (RLBO) commonality-in-liquidity effect, 166
literary survey, 370 cost of capital for private equity
value creation, 335, 352, 354t, 355t illiquidity and the cost of equity in private
special purpose acquisition companies equity, 162, 163t
(SPACs), 7980, 491 liquidity premium, 150, 155, 156, 158, 162
3G (Brazilian private equity firm), 36, 80 general partners, 168169
transaction data, 334345 global financial crisis. See liquidity issues during
data panel and descriptive key statistics, 335, global financial crisis
337, 338t340t institutional investors and listed private equity
data panel key statistics, 337, 341t344t (LPE), 426427
key data statistics of sample, 336f337f, 345 Kohlberg, Kravis, and Roberts (KKR), new
reversed leveraged buyout (RLBO), 335, liquidity sources, 507
352, 354t, 355t limited partners, 168
value creation, 346351 liquidated funds-of-funds, 471, 473t, 478t, 479
reversed leveraged buyout (RLBO), 335, listed private equity (LPE) and institutional
352, 354t, 355t investors, 426427
i n d e x 577

portfolio management challenges, RRTTLU Pastor-Stambaugh liquidity measure, 175


list, 187 public market equivalents (PMEs), 176
private equity secondary markets, 168169, regulatory changes, 174175
172, 173, 176 W-curve, 170
risk adjustments. See liquidity and risk Listed Private Equity Association (LPEQ), 423,
adjustments 424, 426, 427, 438, 484, 495
special purpose acquisition companies listed private equity (LPE). See also publicly traded
(SPACs), one-time liquid private equity private equity (PTPE)
funds, 80 defined, 4
standard definition of liquidity, 168 institutional investors. See listed private equity
types of liquidity risk, 167169 (LPE) and institutional investors
funding liquidity risk, 169 listed PE firms, 420
market liquidity risk, 168169 listed PE funds, 420
liquidity and risk adjustments popularity of, statistics, 420
benchmark biases in private equity private nature of private equity, 419
performance, 262267 synonyms for listed private equity, 483
liquidity adjustments, 264265 types of, 420
measuring gross or net of fees, 267 listed private equity (LPE) and institutional
risk adjustments, 263264 investors, 13, 419439
stale pricing problem, 265266 abnormal returns, 420, 424
valuation challenges, 266267 beta of listed private equity funds,
liquidity issues during global financial crisis, 9, 420421, 424
167, 169176 data and summary statistics, 427436
alternative investment fund (AIF), 175 comparisons of descriptive statistics between
Alternative Investment Fund Managers LPE as part and not part of investment
Directive (AIFMD), 174175, 177 mandate, 432436
annual median net internal rates of return by definitions and summary statistics, 427432
vintage, 170 fund-of-funds, 420, 424, 432
California Public Employees Retirement home bias, 420
Systems (CalPERS), 173 limited partnerships, 419
capital asset pricing model (CAPM), 173174 liquidity-time preference, literature and
capital calls, 171, 172173, 176177 hypotheses, 426427
denominator effect (portfolio composition), literature and hypotheses, 423427
172 liquidity-time preference, 426427
Dodd-Frank Act, 174175 search costs, 423425
dot-com bubble (tech bubble), 169 specific human capital, 425426
empirical evidence, 175176 logit regressions, 436437
endowment model (Yale model), 169, 171, market capitalization, 1993 to 2013, 420, 422f
173174, 177 multivariate analyses, 436438
European Private Equity and Venture Capital logit regressions, 436437
Association (EVCA), 173 regression analyses of percentage allocations,
exit activity, decline, 172173 437438
exiting private equity investments, examples, performance
173 absolute return performance (2004 to 2014),
general partners, affect on, 171, 172173 421, 422f
general partners, exclusive right of to make statistics, 420423
capital calls, 172173 post initial public offering performance of
Harvard Management Corporation Blackstone and Fortress (2007 to 2014),
(HMC), 173 420, 421f
historical private equity fundraising regression analyses of percentage allocations,
(20082013), 171172 437438
institutional investors exiting private equity search costs, literature and hypotheses,
investments, examples, 173 423425
investors, affect on, 171 smaller institutional investors, 421, 423,
J-curve, 170 424, 438
lessons learned, 173175 specific human capital, literature and
limited partners, affect on, 172 hypotheses, 425426
578 i n d e x

listed private equity (LPE) and institutional carried interest, proposed tax change, 515, 519
investors (continued) crowdfunding, 201, 513514
United Kingdom, institutional investors, 421f, Foreign Account Tax Compliance Act
423, 424, 427, 438 (FATCA), 515, 518
unlisted funds, 420, 424, 425 IPO On-Ramp provision ( JOBS Act), 201, 514
venture capital trusts (VCTs), 424 Jumpstart Our Business Startup Act ( JOBS
worldwide distribution of LPE companies Act), 200201, 209, 513
(December 2013), 420, 421f leveraged buyouts (LBOs), risky banking
literature survey, 368374 practices, 516
compensation structure in practice, 369371 retail investors, prohibitions, 512
determinants of fee structure, 371372 market liquidity risk, 168169
fees and incentive distortion, 373 market timing, 193194
relationship between fees and fund Markowitz portfolio optimization, 184
performance, 373 masculinity vs. femininity, 302
size of GP incentives, 372 mature funds, 57, 266267, 268, 271, 278, 473,
summary statistics of fee terms, 370t 474t, 476, 477t, 478t, 479
reversed leveraged buyout (LBO) benchmark biases in private equity
(RLBO), 370 performance, 266268, 271
time series dynamics and the future of fee return persistence, 278
structure, 373374 return persistence and quartile managers, 278
Lithuania, private equity attractiveness, 449f, sample selection bias, 268
451f, 452 venture capital in Europe, 57
loan default and pubic sale (under UCC median distressed fund return, 103104, 112
definition), 9596 Mexico, private equity attractiveness, 441, 444,
loan-to-own investment strategy, 108 449f, 451f, 452, 453455, 456f, 460
Longitudinal Business Database (LBD), 24 mezzanine commercial real estate (CRE)
long-term orientation, 302 financing, 8, 8498
fundamentals, 8486, 9798
authorities, 86
Madoff, Bernard, 297, 512 consumers of mezzanine financing, 85
Malaysia, private equity attractiveness, 448, 449f, hallmark of mezzanine investments, 8485
451f, 452, 460 primary categories of mezzanine
management buy-ins, 34 financing, 86
management buyouts (MBOs), 200, 230 sources/providers of mezzanine capital, 85,
due diligence, 293 8889, 9798
of public companies, 18f, 2324, 27, 34, 37 history of, 8688
value creation, private equity and, 331, 351 commercial mortgage-backed securities
management team, finding, 292 (CMBS), 8688, 9091, 9395, 9798
managers. See also top-quartile managers debt securitization, 87
carried interest, proposed tax change, 515, 519 Financial Crisis Inquiry Commission (2011),
compensation. See Compensation structure of 8788
private equity funds; funds-of-funds, financial crisis of 20072008, 8788, 89
double layer of fees legal factors, distinguishing forms of mezzanine
Marine Corps, Code Red, 239 financing, 9497
market developments, post-financial crisis, enforceability, legal remedies, 9596
3538, 43 fiduciary duty, 9697
Europe, 35, 3738 loan default and pubic sale (under UCC
Federal Reserve, 2013 monetary policy, 36 definition), 9596
general trends, 3637 Revenue Procedure 200365, 94
golden age of private equity, 35 standardization of documentation, 97
private equity mega deals, 36 tax considerations, 9495
recovery and statistics, U.S. buyout market, 36 mezzanine capital
United States, 35, 3637 collateralized debt obligation (CDO), 88
marketing/advertising, new rules, 512514 cost of capital for private equity, mezzanine
accredited investors, 512513 index (MEZ), 153, 158, 160t
Alternative Investment Fund Managers defined, 3, 8485
Directive (AIFMD), 515, 518 demand for, 88
Basel III Accord, banking industry, 514515 hedge funds, 8889, 98
i n d e x 579

institutional investors, 313 J-curve, 494495, 496


pension plans, 8889, 93, 94 market timing and investor sentiment, 495498
sources/providers of, 85, 8889, 9798 private equity, specific explanations, 494495
users of, 8889, 9798 qualitative behavior of premiums, 495498
modes of investments, 8994, 9798 Netherlands, private equity attractiveness,
A-B loans, 9394 240, 244t
debt obligations, 9091 net present value (NPV), 35, 261, 276, 478, 557
hybrid mezzanine investment, 93 NewCo SPEs, 465
preferred equity, comparison of mezzanine new firm creation and patents, 506
loan and, 92t newly listed firms, 203, 206207, 211
preferred equity, distinction between New Zealand
mezzanine debt and, 91, 92t private equity attractiveness, 449f, 451f, 452
rating agencies, 86, 87, 90, 93, 94, 97, 98 return on capital (ROC) performance, 20
structural subordination, 90 Nigeria, private equity attractiveness, 444, 450f
providers and users of mezzanine capital, North America, private equity attractiveness, 241,
8889, 9798 246, 248
real estate investment trust (REIT), 9495
special purpose entity (SPE), 90, 92t
types of mezzanine financing, 86 Obama, Barack, 513
mezzanine index (MEZ) Obama administration, 17, 200
estimation of cost, decomposition of risk, Oman, private equity attractiveness, 451f,
158, 160t 452, 460
S&P 500 index vs., 153 operating performance, impact of leveraged
minority shareholders, eliminating, 386387 buyouts (LBOs), 351356
modern portfolio theory (MPT), 181, 184185, operational due diligence, 297
189190, 196197 operational efficiency benefits of private equity
modified internal rate of return (MIRR), 148 firms, 28, 518, 519
money-chasing deal phenomenon, 59 organizational culture, 301302
monitoring investments organizational skills, due diligence, 292
VC-backed IPOs, 203, 209210, 229 outliers, sample selection bias, 270, 272
board of directors and management outsourcing, 519
team, 210 overconfidence bias (Lake Wobegon fallacy), 197
endogeneity challenges, corporate
governance and, 210
information asymmetry, reduction of, Pakistan, private equity attractiveness, 444
203, 209 Pan-European focus, private equity attractiveness,
Moran v. Household International, Inc., 388 241
Morocco, private equity attractiveness, 449f, 451f, partnership or limited liability corporation
452, 460 (LLC), 5
mortgage effect, leveraged buyouts (LBOs), 67 Pastor-Stambaugh liquidity measure, 175
MSCI Europe Index, 250, 251t, 252, 253t, 254 patents, 506
MSCI World Index, 250, 251t, 252, 253, 254 PE-backed IPOs. See initial public offerings
mutual funds, 402, 404t, 412 (IPOs)
PEGCC. See Private Equity Growth Capital
Council (PEGCC)
nave investor hypothesis, 204 pension funds
NASDAQ Composite Index, 265, 271 California Public Employees Retirement
national culture, dimensions, 302 Systems (CalPERS), 173
natural hedges, 194 defined contribution plans, 323234
net asset value (NAV), 262, 491498 institutional investors, 311, 314, 323325
cash-on-cash multiples, 275 largest retirement funds, 324
closed-end fund puzzle, 492, 493494 mezzanine capital, 8889, 93, 94
fund managers, private benefits, 494 performance evaluation, top quartile managers,
tax timing hypothesis, 494 285286
compensation structure of private equity funds, performance of funds
362, 367 venture capital in Europe, 5458
hot issue markets, 496 J-curve effect, 5455
internal rate of return (IRR), 277 mature funds, 57
580 i n d e x

performance of funds (continued) post-financial crisis


performance comparisons, Europe and liquidity issues. See liquidity issues during global
United States, 5962 financial crisis
comparative performance of venture market developments. See market developments,
capital funds in Europe and United post-financial crisis
States, 5961 net present value (NPV), 35
IPO exit rates, 60 Private Equity Growth Capital Council
toward a convergence of venture capital (PEGCC), 36
practices in Europe and United regulatory developments. See regulatory
States, 6162 developments, post-financial crisis
principal-agent problem, 55 potential private equity investment, valuing. See
smoothing bias, 57 valuing potential private equity investment
Thomson Venture Economics (TVE), 5758, (valuation example)
268269, 281t power distance, 302
performance of investments, venture capital in PowerShares Global Listed Private Equity
Europe, 5354 Portfolio (PSP), 4, 248, 249250
Perlman v. Feldman, 383 correlation coefficient compared with other
persistence in private equity fund returns assets, 254f
transition matrix, probability of transition from correlation coefficients (20082013), 253t
one quarter to another, 282f PrEQIn database, 101, 103, 246, 260, 274, 281t,
persistence of funds in private equity fund returns 282, 322, 326, 369, 427, 471, 472t
venture capital in Europe, 5859 benchmark biases in private equity
Peru, private equity attractiveness, 449f, 451f, 452 performance, 260
philanthropy model, crowdfunding, 514 institutional investors, 322, 326
Philippines, private equity attractiveness, 441, 444, Principal Advisers Act, 511
450, 451f, 452, 453455, 456f, 460 principal-agent problem, 55, 204, 317, 356
placement agents, role in PIPE market, defined, 463
407411, 417 institutional investors, 317
agent compensation value creation, private equity and, 356
good faith retainer fee, 409 VC-backed IPOs, 204
private placement agreement (engagement venture capital in Europe, 55
letter), 409 private equity, generally, 5, 181185
types of compensation, 409 appraisal value effect, 182
contract knowledge gap, bridging, 410411 as cash-flow assets, 184
private placement agreement (engagement buy-to-hold (buy-to-keep), 182, 183, 184
letter), 408 buy-to-sell, 182, 183, 184, 190
relationship between issuer and placement definition, 3, 182, 258
agent, 407408 intermediation through fund investment,
reputation of agents, 409410 183184
Carter and Manaster (C&M) ranking, portfolio management, 184185
409410 private equity intervention, 183
SEC enforcement actions, effect of, 412413 privately held companies and private equity,
specific tasks and advising function, 408409 distinction between, 182
poison pills, 387389, 390391 stale pricing, 182, 194
acquisition regulation, 387389, 390391 private equity associations, 504505
greenmail, 388 Emerging Markets Private Equity Association
Grobow v. Perot, 389 (EMPEA), 504
Moran v. Household International, Inc., 388 European Private Equity and Venture Capital
Revlon v. MacAndrews & Forbes Holdings, Inc., Association (EVCA), 50, 62, 173, 186,
388, 391 465, 505
Smith v. Van Gorkom, 388 Latin American Private Equity and Venture
Unocal Corporation v. Mesa Petroleum Co., 388 Capital Association (LAVCA),
pooled fund-of-funds fees, 468 505, 509
portfolio composition (denominator effect), Listed Private Equity Association (LPEQ), 423,
172, 188 424, 426, 427, 438, 484, 495
portfolio management. See private equity portfolio Private Equity Growth Capital Council
management (PEGCC), 314, 504
i n d e x 581

private equity business model, ability of countries compensation. See compensation structure of
to support, 444448, 459460 private equity funds
chicken and egg problem, causality debt-push-down mergers, 465
determinations, 447 defined, 4, 465
depth of the capital market, bank centered and general partners, 2122
stock market centered, 445 limited partners, 2122
economic activity, 445 mixed evidence of fund performance, 2324
entrepreneurial culture and deal opportunities, NewCo SPE, 465
innovation, 447 return persistence and quartile managers,
human and social environment, 446447 278284
investor protection and corporate governance, special purpose entity (SPE), management of
quality of legal system, 446 fund, 465
key drivers of private equity maturity, structure of funds, 2122
determinants, 445448, 459460 top 10 largest private equity funds, 22, 23t
tax regimes, impact of, 446 typical structure, 2122
private equity deals private equity funds-of-funds
first formal private equity deal, 16 defined, 465467
largest buyout deals, 16, 17t fund raising challenges, 465466
mega deals, 36 fund raising statistics, 465
top 10 largest private equity buyouts, ranked by make or buy private equity access
inflation-adjusted deal value in 2014 alternatives, 466467
dollars, 16, 17t pooled funds-of-funds, 466467
private equity due diligence. See due diligence proprietary funds-of-funds, 466
private equity firms. See also Kohlberg, Kravis, and Private Equity Growth Capital Council (PEGCC),
Roberts (KKR) 4, 36, 127, 314, 504, 512
benefits of private equity firms, 2228 private equity index
bankruptcy rate, default and recovery, 2627 estimation of cost, decomposition of risk,
corporate governance structure 158, 161t
improvements, 2526 S&P 500 index vs., 154
dividends, misconceptions regarding, 28 private equity industry
endogeneity, corporate governance and, 26 background, 18
fund performance, mixed evidence of, 2324 globalization trends in private equity industry,
innovation and operational efficiency, 28 505510
job creation and unemployment, 2728 number of closed or effective leveraged buyouts
value creation, 2324 (LBOs) or management buyouts (MBOs)
for limited partners, 2425 (1970 to 2013), 18f
buyout private equity firms, 116117 top 10 largest private equity buyouts, ranked by
compensation schemes, 18 inflation-adjusted deal value in 2014
criticisms of private equity firms, 1618 dollars, 16, 17t
firm value, 1920 top 10 largest private equity funds, 22, 23t
poorly performing publicly listed companies worldwide private equity growth trend, 463
in different financial markets ( January private equity intervention, 183
2014), 1920 buy-to-hold (buy-to-keep), 182, 183, 184
innovation buy-to-sell, 182, 183, 184, 190
at-the-market (ATM) offering, 414415 private equity performance, benchmark biases in.
benefits of private equity firms, 28 See Benchmark biases in private equity
confidentially marketed public offerings performance
(CMPOs), 415416 private equity portfolio management, 9, 181198
entrepreneurial culture and deal active portfolio management, 196
opportunities, 447 asset allocation, 188190
new firm creation and patents, 506 background risk, 189190
small firms, alternative financing, 412, 412t nave approaches, 189
20 percent rule, 416 buy-to-hold (buy-to-keep), 182, 183, 184
largest private equity firms, 505 buy-to-sell, 182, 183, 184, 190
operational efficiency benefits of, 28 challenges
private equity funds asset liability management (ALM), 187
background, typical structure, 2122 RRTTLU framework, 181, 185188
582 i n d e x

private equity portfolio management (continued) S&P 500 index, 242243, 250, 251, 252,
diversification management, 192195 253t, 254
endowment model (Yale model), 189190 S&P GSCI Commodity Total Return, 250,
generally, 184185 251t, 252, 253
institutional investors, 182, 183, 184, 186187, country-specific private equity attractiveness,
188, 190, 198 243246 (See also Country Attractiveness
Markowitz portfolio optimization, 184 Index)
modern portfolio theory (MPT), 181, BRICs country attractiveness index values
184185, 189190, 196197 (20102014), 245t
monitoring and rebalancing, 196 factors, 243244
outlook, 197 graph of the top three and bottom three
overconfidence bias (Lake Wobegon fallacy), countries for private equity
197 attractiveness, 245f
portfolio management frameworks, 190192 least attractive countries for private equity,
adaptive market hypothesis (AMH), 181, 244245, 450f
191 most attractive countries for private equity,
application to private equity portfolios, 244, 441
191192 Venture Capital & Private Equity Country
behavioral finance, 190191 Attractiveness Index, 243244
core-satellite method, 181, 190191 diversification benefits, 243
practical implementation, 195196 diversifying agent, global private equity as,
private equity intervention, 183 252254
privately held companies and private equity, global interest in private equity, growth of,
distinction between, 182 240241
return, risk, time horizon, taxes and fees, Global Listed Private Equity (GLPE) return
liquidity, legal, and unique circumstances analysis, 251252
(RRTTLLU) framework, 181, 185188 annual returns by year for various assets,
satisficing rather than optimizing, 185 2008-2013, 251t
sovereign wealth funds, 183, 187 correlation coefficient for Global Listed
stale pricing, 182, 194 Private Equity (GLPE) compared with
top-down and bottom-up approaches, 181, other assets, 252f
195196 inverse coefficient of variation, 20082013,
university endowments, 189190 252t
private equity return measures, 275278 global private equity activity, data on, 246248
asset class, characteristics of, 275 global buyout deals by geographic regions,
cash-on-cash multiples, 275276 2006 vs. 2013, 248f
internal rate of return, 276277 global private equity buyouts as a proportion
public market equivalent (PME), 275, 277278 of their 2006 value, 247f
return persistence and quartile managers, proportion of buyout deals by geographic
275278 region, 2006-2013, 247f
top-to-bottom quartile private equity internal global private equity as a diversifying agent,
rate of return ranges by vintage year, 277f 252254
private equity returns, cross-country evidence, 10, Global Listed Private Equity (GLPE)
239254 correlation coefficients, 2008-2013,
aggregate value of private equity-backed buyout 253t
deals by region, 20062013, 246t PowerShares Global Listed Private Equity
BRIC (Brazil, Russia, India, and China), 241, Portfolio (PSP), 4, 248, 249250
242, 245 correlation coefficient compared with
buyouts, existing strategy for, 241 other assets, 254f
Companies Act of 1981, 240 correlation coefficients (2008-2013), 253t
comparison of private equity investments historical performance of Global Listed Private
indexes, 250251 Equity (GLPE), 248249
FTSE EPRA/NAREIT Global Real Estate initial public offerings (IPOs), 241
Index, 250251, 252, 253, 254f literature review on private equity returns,
MSCI Europe Index, 250, 251t, 252, 241243
253t, 254 performance of private equity, assessment
MSCI World Index, 250, 251t, 252, 253, 254 issues, 242243
i n d e x 583

PowerShares Global Listed Private Equity structured equity line, 400, 401t, 402f, 405
Portfolio (PSP), 248, 249250, 253t, 254f toxic PIPE, 400
Red Rocks Global Listed Private Equity placement agents. See placement agents, role in
Index, 249 PIPE market
venture capital, define, 240 private equity (PE) funds, 398, 412, 417
private investment in public equity (PIPE), 13, Regulation D (Rule 501), 397398
397417 regulations, 413414
abnormal returns, 415 seasoned equity offering (SEO), 398, 405, 407,
at-the-market (ATM) offering, 414415 408, 409, 413, 414t, 415, 416, 417
buyouts, 402, 404t SECs Securities Offering Reform, registration
confidentially marketed public offerings procedure reforms, 413
(CMPOs), 415416 Securities and Exchange Commission (SEC)
contractual structure of PIPE transactions, enforcement actions, 411413
terms and provisions, 405407 Amro International, S.A. v. Sedona Corp., 411
company-forced conversion options, 407 death-spiral transactions, hedge funds,
company put option, 407 411413
investor protections, 405406 effect of enforcement actions, 412413
issuer rights, 407 lack of success, 412
lock-up period to PIPE transaction, post-action period compared with pre-action
406407 period, 413
optional redemption provisions, 407 SEC v. Badian, 411
placement agents, bridging contractual SEC v. Berlacher, 412
knowledge gap, 410411 SEC v. Rhino Advisors, Inc. and Thomas
SEC enforcement actions, effect of, 412 Badian, 411
structured PIPEs, 405, 411, 412 Sedona Corp. v. Ladenburg Thalmann & Co.,
trading restrictions, 406407 411
traditional PIPEs, 405, 412, 417 Form S-3 and Form F-3, 2008 amendments,
cumulative abnormal returns (CAR), 24, 403 413, 414t
death-spiral transactions and SEC enforcement, new regulations, 413
411413, 417 resale registration, 398, 407, 417
Form S-3 and Form F-3, 2008 amendments, Securities Offering Reform, registration
413, 414t procedure reforms, 413
hedge funds, 398, 402403, 404 structured PIPEs, 405, 411, 412
innovations, 414416 traditional PIPEs, 405, 412, 417
issuance of PIPE securities under Regulation D 20 percent rule, 416
(Rule 501), 397398 venture capital funds, 398, 402403, 404f,
major investors in PIPE market, 402404 412
mutual funds, 402, 404t, 412 private investment in public equity (PIPE),
new regulations investors
Form S-3 and Form F-3, 2008 amendments, anti-dilution protection, 406
413, 414t cumulative abnormal returns (CARs), 24, 403
SECs Securities Offering Reform, dividend, interest, and warrants, 405
registration procedure reforms, 413 investor call options, 406
new regulations and innovations investor protections, 405406
small firms, alternative financing, 412, 412t investor registration rights, 405406
offerings, statistics, 397 major investors in PIPE market, 402404
overview of PIPE market, 397402 private investment statistics, 403t
development of market, 417 financial sector investments
financial crisis of 2007-2008, impact on PIPE (19952012), 403t
market, 402 total amount (USD) invested, investor type
growth of PIPE market, statistics, 398399 (19952012), 404t
industry distributions, statistics, 398, 400t redemption rights, 406
popularity of PIPE market, 398 right of first refusal, 406
resale registration, SEC filing, 398 strategic investors, 403
security structure, variations, 398, 401t, 402t privately held companies and private equity,
security structure variations, 398, 401t, 402f distinction between, 182
size of PIPE market, 399t private placement memoranda (PPM), 286, 288
584 i n d e x

private placement memoranda (continued) special legal forms for public investments in
international and emerging markets, private private equity, 490491
equity growth, 457458 business development companies
private equity country maturity and historic (BDCs), 490
private equity returns, 457458 investment trusts, 490
return persistence and quartile managers, small business investment companies
286, 288 (SBICs), 490
professionalization of European venture capital special purpose acquisition companies
practices, 5053, 6263 (SPACs), 7980, 491
increasing professionalism, 6162 split capital trusts, 491
negative aspects of investors stability, 51 venture capital trusts (VCTs), 484, 490491
positive aspects of investors stability, 5051 popularity of, statistics, 420
professionalizing investors practices, 5253 portfolio companies, 483
serial entrepreneurship in Europe, 5152 private nature of private equity, 419
ProShares Global Listed Private Equity ETF, 4 publicly traded private equity, defined, 483
PR providers, due diligence, 293 synonyms for publicly traded private equity, 483
pubic sale (under UCC definition), 9596 types of, 420
public companies, management buyouts (MBOs), public market equivalents (PMEs), 56, 258, 260,
331, 351 261262, 275
due diligence, 293 cost of capital for private equity, 148
publicly traded corporation acquisition, hostile liquidity issues during global financial crisis, 176
takeovers, 383384 private equity return measures, 277278
publicly traded private equity (PTPE), 14, survivorship bias, 267
483500. See also listed private equity (LPE)
advantages, 483, 490
alternative investment market (AIM), quartile managers. See return persistence and
486, 490 quartile managers
current and historical market trends,
484485
indexes, 484485 rational discounting hypothesis, 204
disadvantages, 483484 real estate investment trust (REIT), 9495
dot-com boom, 484 rebranding private equity firms, 503
financial crisis of 2007-2008, 484, 499 recommitment, institutional investors, 321
institutional investors and. See listed private Red Rocks Global Listed Private Equity Index, 249
equity (LPE) and institutional investors registration
listed private equity firms, 420 advisers, requirements, 511512
listed private equity funds, 420 seasoned issuers, SECs Securities Offering
location of vehicles, statistics, 484 Reform, 413
market risk, 498499 Regulation D (Rule 501), 380, 397398
net asset value (NAV) discounts, 491498 regulatory developments, post-financial crisis,
closed-end fund puzzle, 492, 493494 3842, 43
fund managers, private benefits, 494 European Union, 38, 4142
tax timing hypothesis, 494 Investment Advisers Act (Advisers Act), 39
hot issue markets, 496 Volcker Rule (Section 619 of Dodd-Frank Act),
J-curve, 494495, 496 38, 39, 40
market timing and investor sentiment, regulatory framework. See global regulatory
495498 framework
private equity, specific explanations, relationship banking, 71, 75
494495 relative compounded return (RCR), 262
qualitative behavior of premiums, 495498 renminbi (RMB) funds, China, 508
organizational forms, vehicles, 486491 restructuring the acquired companies, 391
classification of, dimensions, 486 retail investors, 416, 419
funds, 486487 prohibitions, new marketing rules, 512
funds-of-funds, 487488 retirement funds. See pension funds
investment companies, 488489 return, risk, time horizon, taxes and fees, liquidity,
lock-up periods, 489 legal, and unique circumstances
management companies, 488, 489t (RRTTLLU) framework, 181, 185188
i n d e x 585

asset liability management (ALM), 187 Revlon v. MacAndrews & Forbes Holdings, Inc.,
denominator effect (portfolio composition), 388, 391
188 reward-based model, crowdfunding, 514
environment, social, and governance (ESG), 188 Risk Measurement Guidelines (EVCA ), 186
European Private Equity and Venture Capital RJR Nabisco, 16, 17t, 32
Association (EVCA), 186 Romney, Mitt, 16, 515, 519
legal issues, regulations and perceptions of Russia
regulations, 187188 globalization trends in private equity
liquidity, 187 industry, 509
return, 185186 private equity attractiveness, 245, 441, 443,
risk, 186 444f, 449f, 451f, 452456, 460
Risk Measurement Guidelines (EVCA ), 186 private equity returns, cross-country evidence,
tax and regulatory requirements, 187 241, 242, 245
time horizon, 186187
unique circumstances, 188
return persistence and quartile managers, 11, sales, exit strategies, post-financial crisis, 33, 36
274288 Santa Fe Industries v. Green, 386
distribution to paid-in-capital (DPI), 275 seasoned equity offering (SEO), 398, 405, 407,
evidence of quartile persistence 408, 409, 413, 414t, 415, 416, 417
reasons why returns do or do not persist, seasoned issuers
283284 SECs Securities Offering Reform, registration
return persistence post-2000, 282283 procedure reforms, 413
ex-ante performance evaluation, 285286 secondary buyouts (SBOs) (sponsor-to-sponsor
mature funds, 278 buyout)
net present value (NPV), 276 exit strategies, post-financial crisis, 37
PrEQIn database, 274, 281t, 282 leveraged buyout (LBO) funds, 70
private equity fund returns, persistence in, post-financial crisis, 34, 37
278284 exit strategies, 37
empirical evidence on private equity return SEC's Securities Offering Reform, registration
persistence, 281t procedure reforms, 413
evidence of quartile persistence, 279283 sector, identification of; due diligence, 295
initial empirical evidence on return Securities Act of 1933 (SA), 379
persistence, 279281 Securities and Exchange Commission (SEC)
top-quartile fund, defined, 279 private investment in public equity (PIPE)
transition matrix: probability of transition enforcement actions, 411413
from one quarter to another, 282f Amro International, S.A. v. Sedona
private equity return measures, 275278 Corp., 411
cash-on-cash multiples, 275276 death-spiral transactions, hedge funds,
characteristics of asset class, 275 411413
internal rate of return, 276277 effect of enforcement actions, 412413
public market equivalent (PME), 277278 lack of success, 412
top-to-bottom quartile private equity internal post-action period compared with
rate of return ranges by vintage pre-action period, 413
year, 277f SEC v. Badian, 411
top-quartile managers, 284287 SEC v. Berlacher, 412
due diligence, role of, 286287 SEC v. Rhino Advisors, Inc. and Thomas
ex-ante performance evaluation, 285286 Badian, 411
limitations in implementing a top-quartile Sedona Corp. v. Ladenburg Thalmann &
approach, 284286 Co., 411
private placement memoranda (PPM), Form S-3 and Form F-3, 2008 amendments,
286, 288 413, 414t
top-quartile classification, 285 new regulations, 413
Revenue Procedure 2003-65, 94 resale registration, 398, 407, 417
reverse causality explanation, 505 Securities Offering Reform, registration
reversed leveraged buyout (LBO) (RLBO) procedure reforms, 413
literary survey, 370 Regulation D (Rule 501), 380, 397398
value creation, 335, 352, 354t, 355t Securities Exchange Act of 1934 (SEA), 379
586 i n d e x

SEC v. Badian, 411 acquisition regulation, 389391


SEC v. Berlacher, 412 CTS Corp. v. Dynamics Corp. of Am., 389
SEC v. Rhino Advisors, Inc. and Thomas Badian, 411 Edgar v. Mite Corp., 389
Sedona Corp. v. Ladenburg Thalmann & Co., 411 STOXX Europe Private Equity 20 index, 485
selection bias, 267270 strategic due diligence, 299300
outliers and weighting, 270, 272 Sub-Saharan Africa (East Africa)
sample, 267270 globalization trends in private equity industry,
size bias, 269 509510
skill bias, 269272 sub-sector, due diligence in finding, 295
survivorship bias, 267268 Sun Capital Partners III, LP v. New England
update bias, 269 Teamsters & Trucking Indus., 379
venture capital in Europe, 5354 survivorship bias, 267268
vintage bias, 268269 swarm of locusts analogy, 392
serial entrepreneurship in Europe, 5152 Sweden, private equity attractiveness, 242,
Silver Lake Management, LLC, 36, 312 244t, 250
Singapore, private equity attractiveness, 244, 245f, Switzerland, private equity attractiveness, 242,
250, 449f 244t, 250
Singer v. Magnavox, 386 syndicated loans, total volume and average tranche
size bias, 269 size of, 71, 73f
skill selection bias, 269272 syndicating investments, institutional investors, 319
Small Business Act of 1956, 325 systemic risk, 33, 38, 39, 40, 42
small business investment companies (SBICs), 490
small firms, alternative financing, 412, 412t
Smith v. Van Gorkom, 388 Taiwan, private equity attractiveness, 448, 449f
smoothing bias, 57 TargetCo, debt capital structure of, 105106, 107,
Socit Gnrale Private Equity Index 108, 109, 110, 116
(PRIVEX), 485 taxation
sourcing the deal, 292 carried interest, 515, 519
Southeast Asia, emerging markets, 509510 Country Attractiveness Index, impact of tax
sovereign debt crisis, 33, 35, 36, 37, 38, 43 regimes, 446
sovereign wealth funds, 507508, 519 debt tax shield, 17, 19
private equity portfolio management, 183, 187 due diligence, 303
S&P 500 index, 242243, 250, 251, 252, 253t, 254 Foreign Account Tax Compliance Act
special due diligence, 303304, 304t (FATCA), 515, 518
special purpose acquisition companies (SPACs), global regulatory and ethical framework,
7980, 491 378379
special purpose entity (SPE) Key Elements of Legal and Tax Regimes Optimal
debt obligations, 90, 92t for the Development of Private Equity
NewCo SPEs, 465 (EMPEA), 504
S&P GSCI Commodity Total Return, 250, 251t, limited partners and partnerships, capital
252, 253 assembly and income taxation, 378
S&P index vs. mezzanine commercial real estate (CRE)
buyout (BO) index, 153 financing, 9495
growth index (GRO), 152153 private equity portfolio management, RRTTLU
mezzanine index (MEZ), 153 list, 187
private equity index, 154 publicly traded private equity (PTPE), tax
split capital trusts, 491 timing hypothesis, 494
sponsorship, defined under Volcker Rule, 40 valuing private equity
sponsor-to-sponsor buyout. See secondary earnings before interest, tax, depreciation,
buyouts (SBOs) (sponsor-to-sponsor and amortization (EBITDA), 123,
buyout) 131132
S&Ps Listed Private Equity Index, 485 enterprise-value-to-EBITDA (EV/
stale pricing problem EBITDA), 131, 133, 134t, 140
cost of capital for private equity, 149, 150, 162 VC-backed IPOs, underpricing issued
private equity portfolio management, 182, 194 securities; tax advantages, 207
risk and liquidity adjustments, 265266 tax due diligence, 303
state antitakeover laws, 388391 tax timing hypothesis, 494
i n d e x 587

tech bubble. See dot-com bubble greater underpricing, top-tier investment


technology improvements, 518, 519 bankers and underwriters, 208
tender offers and regulatory balancing, 384385, 391 hostile takeovers, reduction of threat, 208
EU Takeover Directive, 42, 385, 387, 389391 hot issue markets, 207208
Williams Act, 383, 384385, 389, 390391 information asymmetry, 206, 208
Thomson Reuters VentureXpert (TVEX), 24, 25 information cascades argument, 207
Thomson Venture Economics (TVE), 5758, lawsuits from shareholders, reduction of
268269, 281t likelihood, 207
3G (Brazilian private equity firm), 36, 80 newly listed firms, 203, 206207, 211
time series dynamics and the future of fee price stabilization, 207
structure, 373374 tax advantages, 207
top-quartile fund, defined, 279 top-tier investment bankers and greater
top-quartile managers, 284287. See also return underpricing, 208
persistence and quartile managers underpricing, defined, 203
due diligence, role of, 286287 VC-backed vs. non-VC-backed IPOs,
ex-ante performance evaluation, 285286 208209, 229
limitations in implementing a top-quartile winners curse, 206
approach, 284286 unemployment and job creation
private placement memoranda (PPM), 286, 288 benefits of private equity firms, 2728
top-quartile classification, 285 Hiring Incentives to Restore Employment Act
top quartile performance, 6869 (HIRE Act), 515
total value to paid-in capital (TVPI), 258, 261 leveraged buyouts (LBOs), Europe, 506
toxic PIPE, 400 United Kingdom
trends in private equity institutional investors, financial assets held by,
diversification benefits, private equity 323t
funds-of-funds, 14, 463481 leveraged buyouts (LBOs) in, history of,
future of private equity, global perspective, 14, 331332
503519 listed private equity (LPE) and institutional
listed private equity, 13, 419439 investors, 421f, 423, 424, 427, 438
private equity growth in international and private equity attractiveness, 240242, 244
emerging markets, 13, 441460 United States
private investment in public equity, 13, institutional investors, financial assets held by,
397417 323t
publicly traded private equity (PTPE), 14, market developments, post-financial crisis, 35,
483500 3637, 43
Turkey, private equity attractiveness, 441, 444, Federal Reserve, 2013 monetary policy, 36
451f, 452, 453455, 456f, 460 general trends, 3637
turnaround investing, 109110 mega deals, 36
20 percent rule, 416 recovery and statistics, U.S. buyout
2-and-20 standard compensation structure, 360, market, 36
467, 468 private equity attractiveness, 244
regulatory developments, post-financial crisis,
38, 3940, 43
UCITS Directive (Article 5 of Directive Dodd-Frank Act, 38, 39, 40
2009/65/EC), 41 Investment Advisers Act (Advisers Act), 39
uncalled committed capital. See dry powder Volcker Rule (Section 619 of Dodd-Frank
(uncalled committed capital) Act), 38, 39, 40
uncertainty avoidance, 302 U.S. Census Bureau, 24, 27
underpricing issued securities U.S. Steel Corporation, 16
VC-backed IPOs, 203, 206209, 229 university endowments, 189190
behavioral aspects, 207 Harvard Management Corporation
dot-com bubble, 208 (HMC), 173
Ernst & Young survey unlisted funds, 420, 424, 425
institutional investors, perceptions Unocal Corporation v. Mesa Petroleum Co., 388
regarding value of private equity and unseasoned issuers
VC-backed IPOs, 209 SECs Securities Offering Reform, registration
generation of excess demand, 208 procedure reforms, 413
588 i n d e x

update bias, 269 valuing private equity, 89, 123145. See also
upward bias, 458 valuing potential private equity investment
(valuation example)
adjusted present value (APV) method
valuation, risk and liquidity adjustment, 266267 cash flows to estimate value, 130131
value creation generally, 133, 134t
leveraged buyouts (LBOs), 346351 capital asset pricing model (CAPM), 133, 135
private equity firms, benefits of, 2324 capital expenditures (CAPEX), 125, 136t138t
for limited partners, 2425 cash flow
reversed leveraged buyout (LBO) (RLBO), complex whole-firm cash flow measure,
335, 352, 354t, 355t 125, 133
value creation, private equity and, 12, 330357 difficulties of valuation process, 124126
data panel and descriptive key statistics, 335, estimating value, 126131
337, 338t340t major cash-flow definitions, components,
data panel key statistics, 337, 341t344t 126
dot-com bubble, 331 measures of cash flow, 124126
Europe, history of leveraged buyouts (LBOs) in, net cash flow, 124
331332 net income, 124126
key data statistics of sample, 336f337f, 345 simple cash flow (SCF), 124, 125, 126
leveraged buyout (LBO) funds, operation of, discounted cash flow (DCF) valuation method,
332333 133, 134t
leveraged buyout (LBO) transaction data, alternative, multiples-based valuation
observations method, 131132
data panel and descriptive key statistics, 335, cash flows to estimate value, 127129
337, 338t340t earnings before interest, tax, depreciation, and
data panel key statistics, 337, 341t344t amortization (EBITDA), 123, 131132
key data statistics of sample, 336f337f, 345 enterprise-value-to-EBITDA (EV/EBITDA),
leveraged buyouts (LBOs) 131, 133
and value creation, generally, 333334 trading multiples, 133, 134t, 140
impact on operating performance, 351356 transactions multiples, 133
value creation, and leverage, 346351 enterprise value, 133, 134t, 135
management buyouts (MBOs) of public estimation of value, use of cash flows, 126131
companies, 331, 351 adjusted present value (APV), 130131
principal-agent problem, 356 capital cash flow (CCF) method, 131
reversed leveraged buyout (LBO) (RLBO), cost of capital, 129130
335, 352, 354t355t discounted cash flow (DCF) valuation
transaction multiples (EV/.EBITDA), method, 127129
336f337f steps involved, 126127
transaction value leverage in continental increasing value of acquisition, private equity
European leveraged buyouts (LBOs), 345 firms turnaround actions, 132
valuing potential private equity investment measures of cash flow, 124126
(valuation example), 133145 multiples-based valuation method, 133,
adjusted present value (APV) valuation method 134t, 140
generally, 133, 134t as alternative to discounted cash flow (DCF)
private equity ownership assumptions, 140, valuation method, 131132
143t144t earnings multiple method (relative value
discounted cash flow (DCF) valuation method method), 132
generally, 133, 134t generally, 133, 134t
private equity ownership assumptions, 138t valuation methods, generally, 133, 134t
139t, 140 weighted average cost of capital (WACC),
multiples-based valuation method, 133135, 129130, 133, 135
141t142t, 145 VC-backed IPOs, 10, 200211, 229
current structure and management American depository receipt (ADR), 202
assumptions, 135, 136t137t certification hypothesis, 203, 205206, 229
generally, 133, 134t grandstanding hypothesis, 205206, 229
private equity ownership assumptions, 138t reputable and less reputable VC firms,
139t, 140 205206
i n d e x 589

conflicts of interest, 203205 underpricing, defined, 203


economic agents with conflicting interests, VC-backed vs. non-VC-backed IPOs,
203 208209, 229
nave investor hypothesis, 204 winners curse, 206
principal-agent problem, 204 venture capital funding, 5359
rational discounting hypothesis, 204 venture capital funds, 242243
conflicts of interest, VC-backed IPOs Venture Capital & Private Equity Country
Kohlberg, Kravis, and Roberts (KKR), Attractiveness Index, 243244
creation of, 204 venture capital trusts (VCTs), 424, 484, 490491
crowdfunding, 201 venture capital (VC). See also Europe, venture
institutional investors, 203, 204, 209 capital; VC-backed IPOs
IPO On-Ramp provision ( JOBS Act), 201 benchmark biases in private equity
Jumpstart Our Business Startup Act ( JOBS performance, 258
Act), 200201, 209 China, private equity industry, 507508
monitoring investments, 203, 209210, 229 defined, 3, 464
board of directors and management team, performance comparisons, Europe and United
210 States, 5962
endogeneity challenges, corporate comparative performance of venture capital
governance and, 210 funds in Europe and United States,
information asymmetry, reduction of, 203, 5961
209 IPO exit rates, 60
Obama administration, 200 toward a convergence of venture capital
performance comparisons, Europe and United practices in Europe and United States,
States 6162
IPO exit rates, 60 venture capital (VC) equity, 313
reasons to exit, 229 Venture Economics, 243
statistics, 200203 Vietnam, private equity attractiveness, 444, 450f
comparison of short and long term returns of vintage bias, 268269
VC-backed and non-VC-backed IPOs Volcker Rule (Section 619 of Dodd-Frank Act),
in United States (19802013), 202 38, 39, 40, 381, 510511, 519
financial buyer exits, different geographic vulture funds, 18
regions and industries, 201 vulture investors, 116
number of VC-backed IPOs in United States
(19802013), 201, 202f
time, average time to exit IPO, 229 Warner Music Group (case study), 8081
unanswered research questions, 211 waterfall provisions
underpricing issued securities, 203, American-style waterfall provisions, 365367
206209, 229 automation technologies, operational
behavioral aspects, 207 improvement opportunities, 518
dot-com bubble, 208 European-style waterfall provisions, 366, 367
Ernst & Young survey fees and expenses, 367
institutional investors, perceptions fees and incentives, distortion, 373
regarding value of PE and regulation of, 80
VC-backed IPOs, 209 waterfall provisions, 363, 365367, 374
generation of excess demand, 208 W-curve, 170
greater underpricing, top-tier investment Weavering Macro Fixed Income Fund
bankers and underwriters, 208 Limited, 297
hostile takeovers, reduction of threat, 208 weighted average cost of capital (WACC),
hot issue markets, 207208 129130, 133, 135
information asymmetry, 206, 208 weighting, sample selection bias, 270, 272
information cascades argument, 207 Weinberger v. UOP, 386
lawsuits from shareholders, reduction of well-known seasoned issuers (WKSI), 413
likelihood, 207 Williams Act, 383, 384385, 389, 390391
newly listed firms, 203, 206207, 211 winners curse, 206
price stabilization, 207
tax advantages, 207
top-tier investment bankers and greater Yale model (endowment model), 169, 171,
underpricing, 208 173174, 177

Potrebbero piacerti anche