Beyond Inflation Targeting

Beyond Infation Targeting
Beyond Infation
Targeting
Assessing the Impacts and Policy Alternatives
Edited by
Gerald A. Epstein
Political Economy Research Institute (PERI), University of
Massachusetts Amherst, USA
A. Erin Yeldan
Bilkent University, Ankara, Turkey and International
Development Economics Associates (IDEAs), India
Edward Elgar
Cheltenham, UK Northampton, MA, USA
Gerald A. Epstein and A. Erin Yeldan 2009
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, electronic,
mechanical or photocopying, recording, or otherwise without the prior
permission of the publisher.
Published by
Edward Elgar Publishing Limited
The Lypiatts
15 Lansdown Road
Cheltenham
Glos GL50 2JA
UK
Edward Elgar Publishing, Inc.
William Pratt House
9 Dewey Court
Northampton
Massachusetts 01060
USA
A catalogue record for this book
is available from the British Library
Library of Congress Control Number: 2009933402
ISBN 978 1 84720 938 2
Printed and bound by MPG Books Group, UK
v
Contents
List of contributors vii
List of abbreviations xii
Preface xiii
PART I INTRODUCTION AND THEORETICAL
FRAMEWORKS
1 Beyond infation targeting: assessing the impacts and policy
alternatives 3
Gerald Epstein and A. Erin Yeldan
2 Real exchange rate, monetary policy and employment:
economic development in a garden of forking paths 28
Roberto Frenkel and Lance Taylor
3 Infation targeting and the real exchange rate in a small
economy: a structuralist approach 44
Jose Antonio Cordero
PART II THEMATIC ISSUES: CLASS RELATIONS AND
GENDER IMPACTS OF INFLATION TARGETING
4 Income, class and preferences towards anti-infation and
anti-unemployment policies 71
Arjun Jayadev
5 The gendered political economy of infation targeting:
assessing its impacts on employment 93
Elissa Braunstein and James Heintz
6 Infation and economic growth: a cross-country non-linear
analysis 116
Robert Pollin and Andong Zhu
PART III INFLATION TARGETING: CRITIQUES AND
COUNTRY-SPECIFIC ALTERNATIVES
7 Infation targeting in Brazil: 19992006 139
Nelson H. Barbosa-Filho
vi Beyond infation targeting
8 Alternatives to infation targeting in Mexico 158
Luis Miguel Galindo and Jaime Ros
9 Five years of competitive and stable real exchange rate in
Argentina, 200207 179
Roberto Frenkel and Martn Rapetti
10 A general equilibrium assessment of twin-targeting in Turkey 203
Cagatay Telli, Ebru Voyvoda and A. Erin Yeldan
11 Employment targeting central bank policy: a policy proposal
for South Africa 227
Gerald Epstein
12 Infation targeting and the design of monetary policy in India 248
Raghbendra Jha
13 Towards an alternative monetary policy in the Philippines 271
Joseph Anthony Lim
14 Monetary policy in Vietnam: alternatives to infation targeting 299
Le Anh Tu Packard
Index 315
vii
Contributors
Nelson H. Barbosa-Filho is Professor of Macroeconomics and Public
Finance at the Institute of Economics of the Federal University of Rio de
Janeiro, Brazil and holds a PhD in economics from the New School for
Social Research. Since 2003, Barbosa has been working in President Lulas
Administration and he is currently the Secretary of Economic Policy at the
Brazilian Ministry of Finance.
Elissa Braunstein is Assistant Professor of Economics at Colorado State
University, Fort Collins, Colorado. She received her PhD in economics
from the University of Massachusetts Amherst and a Master of Pacifc
International Afairs from the School of International Relations and
Pacifc Studies at the University of California San Diego. Prior to joining
CSU, Braunstein was an assistant research professor at the Political
Economy Research Institute (PERI) at the University of Massachusetts
Amherst. Her research focuses on feminist analyses of international eco-
nomics, including the role of foreign direct investment in development.
Recent work involves analysing the impact of foreign investment on gen-
der-based wage inequality in mainland China, the gender-diferentiated
employment impacts of defationary monetary policy in developing coun-
tries, the linkages between trade liberalization and reproductive health
and rights, and the impact of US state tax policy on womens employment.
Among her recent publications are Trading Womens Health & Rights?
Trade Liberalization and Reproductive Health in Developing Countries (Zed
Books, 2006, co-edited with Caren Grown and Anju Malhotra).
Jose Antonio Cordero is Professor of Economics at the Escuela de Economia,
Universidad de Costa Rica, San Jose, Costa Rica, and Center for Economic
and Policy Research (CEPR). Portions of his chapter were written during
an academic year as visiting professor at Mount Holyoke College. Cordero
received his PhD from the University of Notre Dame in 1995. His research
focuses on the macroeconomics of growth and distribution in open-mone-
tary economies, and on the efects of FDI on economic development.
Gerald Epstein is Professor of Economics and founding Co-Director of
the Political Economy Research Institute (PERI) at the University of
Massachusetts Amherst. His major research and teaching areas are the
viii Beyond infation targeting
political economy of central banking, the political economy of inter-
national fnance and egalitarian macroeconomic policy reform. He has
published widely in these areas. He has also served as a consultant to
the United Nations Development Programme (UNDP), International
Labor Organization (ILO), United Nations Conference on Trade and
Development (UNCTAD) and is a staf economist with the Center for
Popular Economics (CPE) in Amherst, Massachusetts.
Roberto Frenkel is Principal Research Associate at CEDES (since 1977)
and Professor at the University of Buenos Aires, Argentina (since 1984).
Presently he is also Director of the Graduate Program on Capital Markets,
University of Buenos Aires and teaches graduate courses at the Di Tella
and FLACSO-San Andrs Universities in Argentina and the University
of Pavia in Italy. He is a member of the UNDP Advisors Group, member
of the Board of the World Institute for Development Economic Research
(WIDER), United Nations University and member of the Academic
Council of CEFID-AR. He has published numerous books and articles
in academic journals on macroeconomic theory and policy, money and
fnance, infation and stabilization policies, and labor market and income
distribution, with special focus on Argentina and Latin America.
Luis Miguel Galindo is Professor of Economics at the Faculty of Economics
at the National Autonomous University of Mexico in Mexico City. He
received his PhD in economics from Newcastle University in Newcastle
upon Tyne in the UK. His main research interests are monetary economics
and environmental economics. His recent research is concentrated in the
transmission mechanism of monetary policy in emerging economics and
the economics of climate change. He has worked as an advisor to several
fnancial institutions and international organizations.
James Heintz, an Associate Research Professor and Associate Director
of the Political Economy Research Institute (PERI at the University
of Massachusetts, Amherst), holds a PhD from the University of
Massachusetts and a Masters degree from the University of Minnesota.
He has written on a wide range of economic policy issues, including job
creation, global labor standards, egalitarian macroeconomic strategies
and investment behavior. He has worked as an international consultant
on projects in Africa sponsored by the International Labor Organization
(ILO) and the United Nations Development Program (UNDP), that
focus on employment-oriented development policy. From 1996 to 1998
he worked as an economist at the National Labour and Economic
Development Institute in Johannesburg , a policy think tank af liated with
the South African labor movement.
Contributors ix
Arjun Jayadev is Assistant Professor of Economics at the University of
Massachusetts Boston. He has also been a research fellow at Columbia
Universitys Committee on Global Thought. His work focuses on
development and inequality. He received his PhD from the University
of Massachusetts Amherst in 2005.
Raghbendra Jha is Professor of Economics and Executive Director of
the Australia South Asia Research Centre, Arndt-Corden Division of
Economics, Australian National University, Canberra. He obtained his
PhD in economics in 1978 from Columbia University, New York, and
has previously taught at Columbia University and Williams College in
the US, Queens University in Canada, the University of Warwick in the
UK, and the Delhi School of Economics and Indira Gandhi Institute of
Development Research in India. His work focuses on macroeconomics
and public economics.
Joseph Anthony Lim is currently a Professor in the Economics Department
of the Ateneo de Manila University, the Philippines. He was also pro-
fessor in economics at the University of the Philippines from 1978 to
2004. He was a Rockefeller scholar and his PhD and Master of Science
graduate degrees were obtained from the University of Pennsylvania and
Massachusetts Institute of Technology, respectively. His felds of expertise
are macroeconomics and development economics. His research works
include analyses of the Asian crises, growth analyses and diagnostics of
the Philippine macroeconomy, the debt burden and its adverse impact
on social and economic spending of developing countries, and the gender
aspects of employment generation. He was a macroeconomics, fnance
and debt advisor to the Bureau for Development Policy (BDP) division of
the United Nations Development Programme (UNDP) New York head-
quarters from 2002 to 2004. He is currently a board member of the Journal
of the Asia Pacifc Economy published by Francis and Taylor.
Le Anh Tu Packard is Senior Economist at Moodys Economy.com, a divi-
sion of Moodys Analytics. She is also a Research Fellow and Convener
of the Research and Study Group on Vietnamese Social and Economic
Reform, Center for Vietnamese Philosophy, Culture and Society, Temple
University, Philadelphia, PA, US, and has spent many years helping the
Vietnamese government on various aspects of economic reform. Her
current research focus is on the international business cycle, trade and
capital fows, and fnancial crises.
Robert Pollin is Professor of Economics and founding Co-Director of the
Political Economy Research Institute at the University of Massachusetts
Amherst. He received a PhD in economics from the New School for Social
x Beyond infation targeting
Research in New York City in 1982. His research centers on macroeco-
nomics, conditions for low-wage workers in the US and globally, and the
analysis of fnancial markets. Has also written and consulted in the area of
labor market policies, including the viability of living wage policies in the
US. He has directed United Nations Development Programme sponsored
projects on employment targeted macroeconomic policies for South Africa
and Kenya, and most recently has focused his research on the analysis of job
creation in the US in the transition to a renewable energy based economy.
Martn Rapetti is Assistant Researcher at CEDES and a PhD student
at the University of Massachusetts Amherst. His feld of specialization
is macroeconomics for developing countries and his current research
focuses on the efects of competitive real exchange rates on economic
development.
Jaime Ros is Professor of Economics at the University of Notre Dame,
Notre Dame, Indiana and Fellow of the Kellogg Institute for International
Studies. Ros specializes in development economics with special reference
to Latin America. His most recent book is Development Theory and the
Economics of Growth (University of Michigan Press, 2000). His articles
have appeared in the Cambridge Journal of Economics, World Development,
Journal of Development Studies, The Manchester School of Economics and
Social Studies, El Trimestre Economico, Desarrollo Economico and other
scholarly journals and edited books. Current projects include an edited
handbook with Amitava Dutt on development economics as well as a
book on the history of Mexicos economic development with Juan Carlos
Moreno Brid.
Lance Taylor is the Arnhold Professor of International Cooperation and
Development at the New School University, New York. He received a PhD
in economics from Harvard University in 1968. He has been Professor in
the economics departments of Harvard and the Massachusetts Institute of
Technology, as well as a visiting professor at the University of Minnesota,
the Universidade da Brasilia, Delhi University and the Stockholm School
of Economics. He moved to the New School for Social Research in 1993.
Taylor has published widely in the areas of macroeconomics, develop-
ment economics and economic theory. He has served as a visiting scholar
or policy advisor in over 25 countries, including Chile, Brazil, Mexico,
Nicaragua, Cuba, Russia, Egypt, Tanzania, Zimbabwe, South Africa,
Pakistan, India and Thailand.
Cagatay Telli is a Researcher at the State Planning Organization, Ankara,
Turkey. He is a graduate of Bilkent University and his areas of expertise
are on general equilibrium modelling and computational methods.
Contributors xi
Ebru Voyvoda is Assistant Professor of Economics at the Department of
Economics, Middle East Technical University, Ankara, Turkey. She holds
a PhD from Bilkent University. Her areas of specialization include general
equilibrium modeling, economics of productivity growth and technical
change, and environmental economics.
A. Erin Yeldan is Professor of Economics at Bilkent University, Ankara,
Turkey. He was a Fulbright Scholar and Visiting Professor at the University
of Massachusetts Amherst and Amherst College while this book was being
written. He is also one of the directors of the International Development
Economics Associates (IDEAs), New Delhi. Yeldan received his PhD
from the University of Minnesota in 1988. His recent work focuses on
development macroeconomics and on applied general equilibrium models
with emphasis on the Turkish economy.
Andong Zhu is Assistant Professor of Economics at Tsinghua University,
Beijing, China. He is a Research Associate of PERI at the University of
Massachusetts Amherst. Zhu received his PhD from the University of
Massachussetts Amherst in 2005. His recent work focuses on political
economy and the imbalances in the world economy and China.
xii
Abbreviations
BCB Brazilian Central Bank
CBRT Central Bank of Turkey
CGE computable general equilibrium
CMN National Monetary Council
CPE Center for Popular Economics
CPI consumer price index
ET employment targeting
FX foreign exchange
GEAR growth, employment and redistribution
ILO International Labour Organization
IMF International Monetary Fund
ISSP International Social Survey Program
IT infation targeting
MDG Millennium Development Goals
MLE medium to large enterprise
PERI Political Economy Research Institute
PPP purchasing power parity
RER real exchange rate
RBI Reserve Bank of India
SAPs structural adjustment programs
SBV State Bank of Vietnam
SCRER stable and competitive real exchange rate
SME small- to medium-sized enterprise
SOCB state-owned commercial bank
SOE state-owned enterprise
UNCTAD United Nations Conference on Trade and Development
UNDP United Nations Development Programme
VAR vector auto-regression
WTO World Trade Organization
xiii
Preface
The chapters in this edited volume report on the results of a three year
international research project designed to evaluate the impacts of the infa-
tion targeting approach to central banking which has swept the global eco-
nomic landscape in the last decade or so. Moving beyond critique, though,
our main motivation for this project has been to develop socially useful
alternatives to infation targeting. This task has become only more crucial
over the years since we began the project, because, as we describe more
fully in the following pages, we believe infation targeting has largely had
negative consequences for economic development, equality and poverty
reduction, especially in the developing world, even as more and more
countries adopt this fawed monetary policy regime. This has become
especially apparent as a global fnancial crisis engulfs much of the world
while many central banks have been slow to respond, partly due to their
obsession with keeping commodity infation in the low single digits.
In structuring the project, we have been anxious to avoid the one size fts
all approach that underlies not only infation targeting itself but also much
of the policy approach of the neoliberal Washington Consensus of which
infation targeting has become a key part. To that end, the authors of the
country study chapters were tasked with designing concrete, country-
specifc alternatives to infation targeting. In addition, other authors have
written studies of long neglected themes in central banking, including the
gender impacts and class aspects of monetary policy. While most of the
chapters are historical and empirical in nature, a few lay out a basic mac-
roeconomic modeling framework to help us understand the key issues at
stake.
Initiated and directed by Gerald Epstein of the Political Economy
Research Institute (PERI) and Economics Department of the University
of Massachusetts and Erin Yeldan of Bilkent University in Turkey,
this Alternatives to Infation Targeting project has benefted from the
numerous presentations and discussions we have had over these several
years in various parts of the world with many economists, activists and
practitioners. More generally, a multi-author, multi-country project such
as ours requires many other things to be successful, including, frst and
foremost, creative, skilled and dedicated researchers, which we are fortu-
nate to have found. The work of many of them appears in these pages.
1
In
xiv Beyond infation targeting
addition it takes fnancial resources to mount a project such as this, and we
thank the Ford Foundation, the Rockefeller Brothers Fund, the United
Nations Department of Economic and Social Afairs (UN-DESA) and
PERI for fnancial support. The project has also benefted from the help
and encouragement of many people and we especially thank K.S. Jomo of
UN-DESA, Manuel Montes and Michael Conroy previously of the Ford
Foundation and Rockefeller Brothers Fund respectively, and Jo-Marie
Greisgraber and Jamie Baker of New Rules for Global Finance. We are
also grateful to Roberto Frenkel and CEDES for hosting an important
project conference in Buenos Aires. Thanks are also due to our graduate
students, Hasan Comert and Luis Rosero, who have contributed frst rate
research assistance, and to Judy Fogg of PERI who has provided untold,
key support along the way. Finally, we are indebted to Alan Sturmer,
Acquisitions Editor at Edward Elgar Publishing, for his help, support and
encouragement along the way.
NOTE
1. Many of these chapters appeared in shorter form in a Special Issue of the International
Review of Applied Economics (IRAE), March 2008: Infation targeting, employment
creation and economic development: assessing the impacts and policy alternatives.
Thanks very much to Malcolm Sawyer, editor of the IRAE, for his help with that issue
and his cooperation with this book project as well.
PART I
Introduction and theoretical frameworks
3
1. Beyond infation targeting: assessing
the impacts and policy alternatives
Gerald Epstein and A. Erin Yeldan
1
1.1 INTRODUCTION
Infation targeting (IT) is the new orthodoxy of mainstream macroeco-
nomic thought. The approach has now been adopted by 24 central banks,
and many more, including those in developing countries, are expressing
serious interest in following suit. Initially adopted by New Zealand in
1990, the norms surrounding the IT regime have been so powerful that
the central banks of both the industrialized and the developing economies
alike have declared that maintaining price stability at the lowest possible
rate of infation is their only mandate. It was generally believed that price
stability is a pre-condition for sustained growth and employment, and that
high infation is damaging the economy in the long run.
In broad terms, the IT policy framework involves the public announce-
ment of infation targets, coupled with a credible and accountable com-
mitment on the part of government policy authorities to the achievement
of these targets (Setterfeld, 2006, p. 653). As advocated, full fedged
infation targeting consists of fve components: absence of other nominal
anchors, such as exchange rates or nominal GDP; an institutional com-
mitment to price stability; absence of fscal dominance; policy (instrument)
independence; and policy transparency and accountability (Bernanke et
al., 1999; Mishkin and Schmidt-Hebbel, 2001, p. 3). In practice, while few
central banks reach the ideal of being full fedged infation targeters,
many others still focus on fghting infation to the virtual exclusion of
other goals.
For its proponents, the appropriate infation target is typically pre-
scribed as maintaining price stability, though there is less agreement on
the meaning of this term and on its precise measurement. Many prac-
titioners simply adopt the widely cited defnition of Alan Greenspan, the
former Governor of the US Fed, issued at the July 1996 meeting of the
Federal Open Market Committee, as a rate of infation that is suf ciently
low that households and businesses do not have to take it into account in
4 Beyond infation targeting
making every day decisions. For Feldstein (1997), however, price stability
meant a long-run infation rate of zero. In addition, IT is usually associ-
ated with appropriate changes in the central bank law that enhances the
independence of the institution (Bernanke et al., 1999, p. 102; Mishkin and
Schmidt-Hebbel, 2001, p. 8; see also Buiter, 2006 for an evaluation). Note
that this promotion of central bank independence often is inconsistent
with the above mentioned commitment to accountability, if by account-
ability, one means democratic accountability.
Ironically, employment creation has dropped of the direct agenda of
most central banks just as the problems of global unemployment, under-
employment and poverty are taking center stage as critical world issues
(Heintz, 2006). The International Labour Organization (ILO) estimates
that in 2003 approximately 186 million people were jobless, the highest
level ever recorded (ILO, 2004a). The employment to population ratio a
measure of unemployment has fallen in the last decade, from 63.3 percent
to 62.5 percent (ILO, 2004). And as the quantity of jobs relative to need has
fallen, there is also a signifcant global problem with respect to the quality
of jobs. The ILO estimates that 22 percent of the developing worlds
workers earn less than $1 a day and 1.4 billion (or 57 percent of the devel-
oping worlds workers) earn less than $2 a day. To reach the Millennium
Development Goal of halving the share of working poor by 2015, sus-
tained, robust economic growth will be required. The ILO estimates that
on average, real GDP growth has to be maintained at 4.7 percent per year
to reduce the share of $1 a day poverty by half by 2015, and signifcantly
more than that to reduce the share of $2 a day poverty by half.
Moreover, Chinas and Indias opening up to the global markets and the
collapse of the Soviet system together have added 1.5 billion new workers
to the worlds economically active population (Freeman, 2004, 2005;
Akyuz, 2006). This means almost a doubling of the global labor force and
a reduction of the global capital-labor ratio by half. Concomitant with the
emergence of the developing countries in the global manufacturing trade,
about 90 percent of the labor employed in world merchandise trade is low
skilled and unskilled, sufering from marginalization and all too frequent
violation of basic worker rights in informalized markets (see, for example,
Akyuz, 2003, 2006; Akyuz et al., 2006).
Under these conditions, a large number of developing countries have
sufered de-industrialization, serious informalization and consequent
worsening of the position of wage-labor, resulting in a deterioration of
income distribution and increased poverty. Many of these phenomena
have occurred in tandem with the onset of neoliberal conditionalities
2

imposing rapid liberalization of trade and premature deregulation of the
indigenous fnancial markets.
Impacts and policy alternatives 5
The key problem is that the ongoing fnancial globalization appears
primarily to redistribute shrinking investment funds and limited jobs
across countries, rather than to accelerate capital accumulation across a
global scale (Adelman and Yeldan, 2000; Akyuz, 2006). Simply put, the
world economy is growing too slowly to generate suf cient jobs and it is
allocating a smaller proportion of its income to fxed capital formation.
In addition, asset price bubbles and crashes, with their attendant fnancial
fall-out are plaguing the system. Under these conditions, it ought to be
clear that price stability, on its own, will not suf ce to maintain macro-
economic stability, as it cannot suf ce to secure fnancial stability and
employment growth. In the words of Akyuz (2006, p. 46), the source of
macroeconomic instability now is not instability in product markets but
asset markets, and the main challenge for policy makers is not infation, but
unemployment and fnancial instability (emphasis added).
Yet, surprisingly, despite a disappointing record, this almost single minded
focus on commodity infation is gaining a more secure foothold in monetary
policy circles and the circles are widening to include an increasing number
of developing countries. According to a recent report by the International
Monetary Fund (IMF), an increasing number of central banks in emerging
markets are planning to adopt IT as their operating framework (Batini et
al., 2006; Table 1.1). An IMF staf survey of 88 non-industrial countries
found that more than half expressed a desire to move to explicit or implicit
quantitative infation targets. More relevant to our concerns, nearly three-
quarters of these countries expressed an interest in moving to full-fedged
IT by 2010. To support and encourage this movement, the IMF is provid-
ing technical assistance to many of these countries and is willing to provide
more (Table 1.1 and further discussion below). In addition, the IMF is
considering altering its conditionality and monitoring structures to include
infation targets. In short, despite little evidence concerning the success of
IT in its promotion of economic growth, employment creation and poverty
reduction, and mixed evidence at best that it actually reduces infation itself,
a substantial momentum is building up for full-fedged IT in developing
countries. Promotion eforts by the IMF and Western-trained economists
are at least partly responsible for this increasing popularity.
While it might seem obvious that stabilization focused monetary policy
represents the only proper role for central banks, in fact looking at history
casts serious doubt on this claim. Far from being the historical norm, in
many of the successful currently developed countries, as well as in many
developing countries in the post-Second World War period, pursuing
development objectives was seen as a crucial part of the central banks
tasks (Epstein, 2007). Now, by contrast, development has dropped of the
policy agenda of central banks in most developing countries.
Table 1.1 Infation targeting countries: initial conditions and modalities
Developing countries
(in order of adoption)
IT
adoption
date
Infation
rate at
start
(% per
annum)
Current
infation
target
(% per
annum)
Of cially
declared
policy
instrument
Israel 1997Q2 8.5 13 headline O/N
rate
Czech Rep. 1998Q1 13.1 3 (1/1) 2 week repo
Poland 1998Q4 9.9 2.5 (1/1) 28 day
intervention
Brazil 1999Q2 3.3 4.5 (1/2) selic O/N rate
Chile 1999Q3 2.9 24 O/N rate
Colombia 1999Q3 9.3 5 (1/0.5) repo
South Africa 2000Q1 2.3 36
Thailand 2000Q2 1.7 03.5 14 day repo
Korea 2001Q1 3.2 2.53.5 O/N call rate
Mexico 2001Q1 8.1 3 (1/1) 91-day Cetes
Hungary 2001Q2 10.5 3.5 (1/1) 2 week deposit
Peru 2002Q1 0.8 2.5 (1/1)
The Philippines 2002Q1 3.8 56 reverse repo
Slovak Republic 2005Q1 3.2 3.5 (1/1)
Indonesia 2005Q3 7.8 5.5 (1/1) 1-month SBI
Romania 2005Q3 8.8 7.5 (1/1)
Turkey
a
2006Q1 7.8 5 (1/2) CB O/N rate
Turkey
b
2001Q2 82.0 n.a CB net
domestic assets
Industrial Countries
New Zealand 1990Q1 7.0 13 cash rate
Canada 1991Q1 6.2 13 O/N funding
rate
United Kingdom 1992Q4 3.6 2 repo
Sweden 1993Q1 4.8 2 (1/1) repo
Australia 1993Q2 1.9 23 cash rate
Iceland 2001Q1 3.9 2.5
Norway 2001Q1 3.7 2.5
Candidate Countries
Costa Rica, Egypt,
Ukraine
Near Term
(12 years)
Albania, Armenia,
Botswana,
Dominican
Medium
Term (35
years)
The theme of this book is that modern central banking ought to have
more policy space in balancing out various objectives and instruments.
In particular, employment creation, poverty reduction and more rapid
economic growth should join infation stabilization and stabilization more
generally as key goals of central bank policy. In introducing this volume,
this chapters outlines why a shift away from IT, the increasingly fashion-
able, but problematic approach to central bank policies, and a move
toward a more balanced approach is both desirable and feasible.
The rest of the chapter is organized as follows. In the next section, we
briefy survey the macroeconomic record of IT and its current structure.
Section 1.3 focuses on the role of the exchange rate as one of the key macro
prices, and discusses alternative theories of its determination. We also
make remarks on the issue of IT in the context of the so-called trilemma
of monetary policy. In Section 1.4 we discuss various alternatives to infa-
tion focused central banks, concentrating on the results of a multi-country
research project undertaken with the support of UN-DESA, among other
organizations. This section shows that there are viable, socially productive
Table 1.1 (continued)
Rep., Gutemala,
Mauritius, Uganda,
Angola, Azerbaijan,
Georgia, Moldova,
Serbia, Sri Lanka,
Vietnam, Zambia
Medium
Term
(35 years)
Belarus, China,
Kenya, Kyrgyz Rep.,
Moldova, Serbia,
Sri Lanka, Vietnam,
Zambia
Bolivia, Honduras,
Nigeria, Papua New
Guinea, Sudan,
Tunisia, Uruguay,
Venezuela
Long Term
(. 5 years)
Notes:
a. Of cial adoption date for Turkey.
b. Turkish central bank declared disguised infation targeting in the aftermath of the 2001
February crisis.
Source: Batini et al. (2006).
alternatives to IT, including those that focus on employment generation,
and makes the case that these alternatives should be further developed.
Section 1.5 concludes the chapter.
1.2 MACROECONOMIC RECORD OF INFLATION
TARGETING
Much of the existing literature on the record of IT has focused mostly on
whether systemic risks and accompanying volatility has been reduced in the
IT economies, and whether infation has come down actually in response to
adoption of the framework itself or due to a set of exogenously welcome
factors. On the one side, there is a fair amount of agreement that IT has
been associated with reductions in infation. Furthermore, exchange rate
pass-through efects were reportedly reduced and consumer prices have
become less prone to shocks (Edwards, 2005; Mishkin and Schmidt-
Hebbel, 2001). Yet existing evidence also suggests that IT has not yielded
infation below the levels attained by the industrial non-targeters that have
adopted other monetary regimes (Ball and Sheridan, 2003; Mishkin and
Scmidt-Hebbel, 2001; Roger and Stone, 2005). Moreover, even if domestic
monetary policy has reduced infation, the hoped for gains in economic
growth and employment have, generally, not materialized.
On the qualitative policy front, it is generally argued that with the
onset of central bank independence, communication and transparency
have improved and that the central banks have become more account-
able. Yet in practice, central bank independence means that central
banks have become less accountable to their governments, and, arguably,
more accountable to fnancial elites and international organizations such
as the IMF.
Moreover, little is known about the true costs of IT on potential output
growth, employment, and on incidence of poverty and income distri-
bution. Bernanke et al. (1999) and Epstein (2007), for instance, report
evidence that IT central banks do not reduce infation at any lower cost
than other countries central banks in terms of foregone output. That is,
IT does not appear to increase the credibility of central bank policy and
therefore does not appear to reduce the sacrifce ratio. Per contra, based
on an econometric study of a large sample of infation targeters and non-
targeters, Corbo et al. (2001) concluded that sacrifce ratios have declined
in the emerging market economies after adoption of IT. They also report
that output volatility has fallen in both emerging and industrialized
economies after adopting IT. This position is recently complemented by
a study of the IMF economists, who, using a complex econometric model
and policy simulations, report fndings that infation targeting economies
experience reductions in the volatility in infation, without experiencing
increased volatility in real variables such as real GDP (Batini et al., 2006).
According to these estimates, IT central banks do enhance economic sta-
bility relative to other monetary rules, such as pegged exchange rates and
monetary rules.
However, in the assessments of stability, these papers do not consider
the issue of the stability of asset prices, including exchange rates, stock
prices and other fnancial asset prices. As we discuss further below, asset
price stability may need to be included in a full analysis of the impact of IT
on overall economic stability. Asset price stability aside, while intriguing,
these results are only as strong as the simulation model on which they are
based and are only as relevant as the relevance of the questions they pose.
Moreover, they are only as broad as the alternatives they explore. On all
these scores, these results are problematic. First, they do not simulate the
impact of IT relative to other possible policy regimes, such as targeting the
real exchange rate as discussed below. Second, the model is based on esti-
mates of potential output that are themselves afected by monetary policy
(see, for example, Michl, 2007; Tobin, 1980). Hence, if monetary policy
slows economic growth, it also lowers the rate of growth of potential
output and, therefore, reduces the gap between the two, thereby appearing
to stabilize the economy.
Equally, if not more important, is the practical problematique of setting
the targeted rate of infation itself. Even if the advocated requisites of the
IT regime are taken for granted, it is not yet clear what the practically
targeted rate of infation should be. Even though there appears to be a
consensus among the advocates of the IT regime that the infation target
has to be as low as possible, there is no theoretical justifcation of this
assertion; and as such, it sounds more of a recommendation than a careful
calculation. Most disturbing is the common belief that what is good for
the industrialized/developed market economies should simply be repli-
cated by the developing countries as well. That this may not be the case
is forcefully argued in Pollin and Zhu (2006). Based on their non-linear
regression estimates of the relationship between infation and economic
growth for 80 countries over the period 19612000, Pollin and Zhu report
that higher infation is associated with moderate gains in GDP growth up
to a roughly 1518 percent infation threshold. Furthermore they report
that there is no justifcation for IT policies as they are currently being
practiced throughout the middle- and low-income countries, that is, to
maintain infation with a 24 percent band.
Moreover, we have other evidence on the negative consequences of
monetary policy designed to produce extremely low levels of infation in
developing countries. Braunstein and Heintz show that contractionary
monetary policy used to fght infation often has a diferentially negative
impact on the employment rates of women relative to men (Braunstein
and Heintz, Chapter 5, this volume). Given the possible negative costs
of IT on output and employment, there should be some direct survey
evidence indicating peoples preferences with respect to infation and
unemployment. While some studies have indicated that people have an
absolute preference for low infation, Arjun Jayadev (Chapter 4, this
volume) reports on survey results asking people in diferent countries
and income levels what is their bigger concern, high infation or high
unemployment. His main result is that, in his sample, poorer people are
concerned more about high unemployment than high infation, while
richer respondents have the opposite preferences. Hence, concerns over
employment and infation probably have an important class dimension
to them, something that economists and historians have suspected for
many years.
Finally, is the issue of the role of IT in the context of supply shocks,
which periodically afect individual economies and, as we have seen
recently, the world economy as a whole. Rigid IT rules can prove highly
problematic in the context of supply shocks, where the main problem
facing countries is too little supply, not too much demand. The solution
is to help the economy absorb the loss of real income associated with the
supply shock, without incurring collateral damage associated with greater
income loss than absolutely necessary, while at the same time making the
investments necessary to increase the supply of the key commodities or
fnd substitutes for them over the medium term. Contractionary monetary
policy is a decidedly highly inef cient tool for carrying out this complex
task. This problem is exacerbated by rigid infation targets and is only
slightly ameliorated by the use of core infation indices which usually
ostensibly exclude the frst round costs of volatile energy and food prices.
This is only a partial solution because in the medium term the increased
costs of key commodities must flter their way through the input output
structure and, even without real wage resistance on the part of workers
which the central banks are presumably concerned with, will raise the core
infation rate temporarily. Hence, rigid adherence to IT in this context will
lead to possibly large unnecessary costs in slower economic growth and
income.
An overall picture on the selected macroeconomic indicators of the
infation targeters can be obtained from Tables 1.2 and 1.3. In Table
1.2 we provide information on the observed behavior of selected macro
aggregates as the annual average of fve years before the adoption of the
IT versus the annual average after the adoption date to the current period.
Table 1.3 keeps the same calendar frames and reports data on key macro
prices, viz. the exchange rate and the interest rates.
Evidence on the growth performance of the IT countries is mixed.
Taking the numbers of Table 1.2 at face value, we see that seven of the 21
countries report a decline in the average annual rate of real growth, while
three countries (Canada, Hungary and Thailand) have not experienced
much of a shift in their rates of growth. Yet clearly it is quite hard to dis-
entangle the efects of the IT regime from other direct and indirect efects
on growth. One such factor is the recent fnancial glut in the global asset
markets and the associated surge of the household defcit spending bubble,
which is now bursting. The Institute of International Finance data reveal,
for instance, that the net capital infows to the developing economies as
a whole has increased from $47 billion in 1998, to almost $400 billion in
2006, surpassing their peak before the Asian crisis of 1997. As the exces-
sive capital accumulation in telecommunications and the dot.com high
tech industries phased out in late 1990s, the global fnancial markets seem
to have entered another phase of expansion, and external efects such as
these make it hard to isolate the growth impacts of the IT regimes.
Despite the inconclusive verdict on the growth front, the fgures on unem-
ployment indicate a signifcant increase in the post-IT era. Only three coun-
tries on our list (Chile, Mexico and Switzerland) report a modest decline
in their rates of unemployment in comparison to the pre-IT averages. The
deterioration of employment performance is especially pronounced (and
puzzling) in countries such as the Philippines, Peru and Turkey where rapid
growth rates were attained. The increased severity of unemployment at the
global scale seems to have afected the IT countries equally strongly, and
the theoretical expectation that price stability would bring growth and
employment in the long run seems quite far from materializing yet.
The adjustment patterns on the balance of foreign trade have been
equally diverse. Ten of the 21 countries in Table 1.2 achieved higher
(improved) trade surpluses (balances). While there have been large defcit
countries, such as Turkey, Mexico, the Philippines and Australia, there
were also sizable surplus generators such as Brazil, Korea, Thailand,
Canada and Sweden. Not surprisingly much of the behavior of the trade
balance could be explained by the extent of over-valuation of the exchange
rates. This information is tabulated in Table 1.3.
Table 1.3, like Table 1.2, calculates the annual averages of the fve-year
period before the IT versus annual averages after IT to date. Focusing on
the infation-adjusted real exchange rate movements, we fnd a general
tendency towards appreciated currencies in the aftermath of adoption
of the IT regimes. Mexico, Indonesia, Korea and Turkey are the most
signifcant currency appreciating countries, while Brazil, and to some
Table 1.2 Selected macroeconomic aggregates in the infation targeting
countries
Before : Annual average of 5 years prior to adoption of IT; After : Annual average of
adoption of IT to current
Growth
rate
Unemploy-
ment rate
Trade Balance
(external
balance on
goods and
services)
/GDP (%)
Central bank
foreign reserves
(US$ Million)
Year IT
started
Before After Before After Before After Before After
New Zealand 1990 2.7 3.0 4.2 6.9 0.4 1.3 2 897.9 4 623.2
Canada 1991 2.9 2.8 8.4 8.7 0.5 2.7 11 964.0 24 256.0
United
Kingdom 1992 2.2 2.7 7.4 5.2 2.5 1.6 39 666.5 37 408.5
Sweden 1993 0.8 2.7 2.8 6.1 1.3 6.2 15 399.0 18 521.8
Australia 1994 2.2 3.9 8.6 7.3 0.6 1.3 13 777.9 20 337.1
Israel 1997 5.8 3.1 8.5 9.4 14.6 7.2 7 567.3 24 421.1
Czech Rep.
a
1998 4.5 3.2 4.0 8.9 3.4 1.8 9 172.5 21 686.5
Poland 1998 7.9 3.7 14.3 16.7 0.0 4.1 12 591.8 31 581.8
Brazil
b
1999 3.2 2.3 7.0 9.8 1.7 1.0 47 701.3 42 304.5
Colombia 1999 3.3 2.3 11.1 15.8 6.0 0.5 7 567.3 24 421.1
Mexico 1999 1.7 4.8 2.7 1.9 0.5 1.9 20 630.9 51 396.6
South Africa
c
2000 2.6 3.8 n.a 27.7 0.0 0.0 15 860.0 9 580.0
Switzerland 2000 1.4 1.7 4.1 3.1 0.1 0.1 38 277.1 40 646.5
Thailand 2000 1.5 1.7 1.9 2.4 0.0 0.1 32 556.1 40 474.8
Korea 2001 4.6 4.5 4.4 3.7 0.0 0.0 55 299.5 157 739.2
Hungary 2001 4.2 4.2 8.0 6.1 1.3 2.9 9 918.1 13 652.1
Peru
d
2002 2.0 5.2 7.8 10.2 3.2 0.4 9 264.8 11 222.9
Philippines 2002 3.1 5.1 10.2 11.5 3.6 0.7 11 281.6 14 006.6
Indonesia 2005 4.6 5.6 6.5 10.3 7.3 4.6 31 326.7 32 989.2
Turkey
e
2006 4.5 7.8 9.9 10.4 9.8 11.0 33 237.4 56 990.4
Turkey
e
2001Q2 4.0 4.5 6.6 10.0 7.5 9.8 20 083.4 33 237.4
Notes:
a. The period before the infation targeting refers to the period of 199497 for Growth and
CPI infation for the Czech Republic.
b. The period before the infation targeting refers to the period of 199698 for reserves in
Brazil.
c. The period before the infation targeting refers to the period of 199497 and after infa-
tion targeting refers to the period of 19992004 for unemployment rate in South Africa.
Note that due to change in methodology and data coverage, unemployment fgures are
not directly comparable before and after apartheid.
d. The period after the infation targeting refers to the period of 200304 for unemployment
rate in Peru.
e. Of cial adoption date for Turkey is 2006. However, Turkish central bank declared dis-
guised infation targeting in the aftermath of the 2001 February crisis.
Source: IMF (2008).
Table 1.3 Macroeconomic prices in the infation targeting countries
Before: Annual average of 5 years prior to adoption of IT; After: Annual average of
adoption of IT to current
Infation rate
(variations
in CPI)
Exchange
rate real
depreciationg,
h
CB real
interest
rate
h,i
Public
assets real
interest rate
i,j
Year IT
started
Before After Before After Before After Before After
New Zealand 1990 11.6 2.2 7.6 0.6 7.0 5.5 2.1 5.1
Chile
a
1991 19.7 7.2 6.0 4.0 .. 0.0 16.0 4.6
Canada 1991 4.5 2.1 7.5 1.7 6.0 2.6 5.8 2.5
United
Kingdom 1992 6.4 2.6 2.4 2.2 5.4 3.0 5.0 2.8
Sweden 1993 6.9 1.5 8.5 1.2 2.8 1.7 5.0 2.9
Australia
b
1994 4.2 2.5 6.9 1.1 7.1 3.2 6.3 4.0
Israel 1997 11.3 3.1 4.2 0.9 2.0 5.0 1.5 5.0
Czech
Republic
c
1998 9.1 3.1 6.6 6.2 1.9 0.7 0.0 0.9
Poland
d
1998 24.1 4.7 4.5 4.6 1.6 6.2 1.8 11.6
Brazil 1999 819.2 7.9 428.0 5.5 782.6 15.7 786.9 12.4
Colombia 1999 20.4 7.5 9.5 0.5 18.4 6.6 1.5 2.0
Mexico 1999 24.5 7.2 2.8 4.6 7.5 5.0 3.2 3.8
Thailand 2000 5.1 2.2 4.5 1.0 4.9 1.6 4.7 3.1
South Africa
e
2000 7.3 5.1 4.3 2.5 8.6 4.4 7.3 4.2
Switzerland 2000 0.8 1.0 1.6 3.7 0.2 0.1 0.9 0.3
Korea 2001 4.0 3.3 6.0 5.0 0.2 1.0 6.5 2.1
Hungary 2001 15.2 5.9 2.5 12.4 2.0 3.4 2.3 3.4
Peru 2002 5.0 1.9 1.6 1.4 9.3 2.0 3.8 0.5
Philippines 2002 6.3 5.0 8.7 3.0 5.1 0.9 5.2 1.2
Indonesia 2005 8.0 10.5 6.2 1.9 4.2 2.3 4.1 2.4
Turkey
f
2006 28.3 10.5 6.3 1.2 11.7 7.5 14.8 10.5
Turkey
f
2001Q2 74.1 28.3 9.9 6.3 13.3 12.7 23.9 15.5
Notes:
a. The period after the infation targeting period refers to the period of 19932005; the
period before the infation targeting refers to the period of 198790.
b. Treasury Bill: the period after the infation targeting refers to the period of 19942000;
CB Rate: the period after the infation targeting refers to the period of 199495.
c. The period before the infation targeting refers to the period of 199497.
d. Treasury Bill rates; the period after the infation targeting refers to the period of 19982000.
e. Treasury Bill: the period before the infation targeting refers to the period of 19942000.
f. Of cial adoption date for Turkey is 2006. However, Turkish central bank declared dis-
guised infation targeting in the aftermath of the 2001 February crisis.
g. A rise in value indicates depreciation. Annual average market rate is used for: UK,
Canada, Turkey, Australia, New Zealand, Brazil, Peru, Israel, Indonesia, Korea and
Philippines. Annual average Of cial Rate is used for: Colombia, Thailand, Hungary,
Poland and Switzerland. Principle rate is used for: South Africa, Mexico and Czech
Republic.
extent Columbia, have pursued active export promotion strategies and
maintained real depreciation. The Czech Republic, Switzerland and
Hungary are observed to have experienced nominal currency apprecia-
tion, and Poland seems to have maintained an appreciating path for its
real exchange rate.
Clearly much of this generalized trend towards appreciation can be
explained by reference to the increased expansion of foreign capital
infows due to the global fnancial glut mentioned above. With the IT
central bankers announcing a no-action stance against exchange rate
movements led by the markets, a period of expansion in the global asset
markets has generated strong tendencies for currency appreciation. What
is puzzling, however, is the rapid and very signifcant expansion in the
foreign exchange reserves reported by the IT central banks. As reported
in the last two columns of Table 1.2, foreign exchange reserves held at the
central banks rose signifcantly in the aftermath of the IT regimes. The rise
of reserves was especially pronounced in Korea, the Philippines and Israel
where almost a fve-fold increase had been witnessed. Of all the countries
surveyed in Table 1.3, the UK and Brazil are the only two countries that
had experienced a fall in their aggregate reserves.
3
This phenomenon is puzzling because the well-celebrated fexibility
of the exchange rate regimes were advocated precisely with the argument
that, under the IT framework, the central banks would gain freedom in
their monetary policies and would no longer need to hold reserves to
defend a targeted rate of exchange. In the absence of any of cially stated
exchange rate target, the need for holding such sums of foreign reserves
at the central banks should have been minimal. The proponents of the IT
regimes argue that the central banks need to hold reserves to maintain
price stability against possible shocks. Yet, the acclaimed defense of price
stability at the expense of such large and costly funds that are virtually
kept idle at the IT central banks reserves is questionable at best in an era
of prolonged unemployment and slow investment growth, and needs to be
justifed economically as well as socially.
We now turn to the issue of exchange rate policy more formally.
h. Nominal values are defated by the corresponding infation averages (CPI column).
i. Sweden, New Zealand, Canada: Bank Rate; Mexico: Bankers Acceptance.
j. Colombia: Interbankaria TBS; Peru and Chile: Saving Rate; New Zealand Newly
issued three months Treasury bill rates; Indonesia: three Months Deposit Rate; Korea:
National Housing Bond Rate; Thailand: Government Bond Yield Rate.
Source: IMF (2008).
1.3 THE ROLE OF THE EXCHANGE RATE UNDER
INFLATION TARGETING
As stated above, part of the broader requirements surrounding the IT
system is often argued to be the implementation of a foating/fexible
exchange rate system in the context of free mobility of capital. Thus,
exchange rate fexibility and foating exchange rate system became the
new motto, and to many advocates, central bank policy has typically been
reduced to mean merely setting the policy interest rate. The exchange
rate and macro prices are, in theory at least, thereby left to the unfettered
workings of the global fnance markets. The role of the exchange rate as
an adjustment variable has clearly increased over the last decade since the
adoption of the foating exchange rate systems. In the meantime, however,
the role of the interest rates and reserve movements has declined substan-
tially as counter-cyclical instruments available to be used against shocks
4

(see Table 1.2).
Against this background a number of practical and conceptual ques-
tions are inevitable: what is the role of the exchange rate in the overall
macroeconomic policy when an explicit IT regime is adopted? Under
what conditions should the central bank, or any other authority, react to
shocks in the foreign exchange market? And perhaps more importantly,
if an intervention in the foreign exchange market is regarded necessary
against, say, the disruptive efects of an external shock, what are the
proper instruments?
To the proponents of IT, the answer to these questions is simple and
straightforward: the central bank should not have any objective in mind
with regards to the level of the exchange rate, yet it might interfere against
the volatility of the exchange rate in so far as it afects the stability of
prices. However nuances remain. To what may be grouped under strict
conformists, the central bank should be concerned with the exchange rate
only if it afects its ability to forecast and target price infation. Any other
response to the foreign exchange market represents a departure from the
IT system. Advocated in the seminal works by Bernanke et al. (1999) and
Fischer (2001), the approach argues that attending to IT and reacting to
the exchange rate are mutually exclusive. Beyond this assertion, the con-
formist view also holds that intervention in the foreign exchange market
could confuse the public regarding the ultimate objective of the central
bank with respect to its priorities, distorting expectations. In a world of
credibility game, such signals would be detrimental to the central banks
authority.
Yet, while maintaining the IT objective, one can also distill a more
active role for the exchange rate in the literature. As outlined by Debelle
(2001), this fexible IT view proposes that the exchange rate can also be
a legitimate policy objective alongside the infation target. More formally,
an operational framework for the fexible IT view was envisaged within
an expanded Taylor rule. Taylor (2000) argued, for instance, that an
exchange rate policy rule can legitimately be embedded in a Taylor rule
that is consistent with the broad objectives of targeted infation rate and
the output gap.
In contrast to all this, the structuralist tradition asserts that irrespective
of the conditionalities of foreign capital and boundaries of IT, it is very import-
ant for the developing economies to maintain a stable and competitive real
exchange rate (SCRER) (see, for example, Cordero, Chapter 3, this volume;
Frenkel and Taylor, Chapter 2, this volume; Galindo and Ros, Chapter 8,
this volume; Frenkel and Ros, 2006; Frenkel and Rapetti, Chapter 9, this
volume). They argue that the real exchange rate can afect employment,
and the economy more generally, through a number of channels: (1) by
afecting the level of aggregate demand (the macroeconomic channel); (2)
by afecting the cost of labor relative to other goods and thereby afecting
the amount of labor hired per unit of output (the labor intensity channel);
and (3) by afecting employment through its impact on investment and eco-
nomic growth (the development channel). While the size and even direction
of these channel efects might difer from country to country, maintaining a
competitive and stable real exchange rate is likely to have a positive employ-
ment impact through some combination of these efects.
The gist of the structuralist case for SCRER rests on a recent (and
unfortunately not well understood or appreciated) paper by Taylor
(2004). Resting his arguments on the system of social accounting ident-
ities, Taylor argues that the exchange rate cannot be regarded as a simple
price determined by temporary macro equilibrium conditions. The main-
stream case for exchange rate determination rests on the well-celebrated
Mundell (1963) and Fleming (1962) model where the model rests on an
assumed duality between reserves (fxed exchange rate system) versus
fexible exchange rate adjustments. The orthodox mainstream model,
according to Taylor, presupposes that a balance of payments exists with
a potential disequilibrium that has to be cleared. This, however, is a false
presumption. The exchange rate is not an independent price and has no
fundamentals such as a given real rate of return (or a trade defcit) that can
make it self-stabilizing. In Taylors words, the balance of payments is at
most an accumulation rule for net foreign assets and has no independent
status as an equilibrium condition. The Mundell-Fleming duality is irrel-
evant, and in temporary equilibrium, the exchange rate does not depend
on how a country operates its monetary (especially international reserve)
policy (Taylor, 2004, p. 212).
The preceding discussions clearly underscore that the real world behav-
ior of exchange rates is quite complex and the focus of the infation target-
ing regime for foating exchange rates (in expectation of dropping it from
the policy agenda altogether) is a mirage. This view of exchange rates helps
to explain why many believe that there are no viable alternatives to IT as a
mode of central bank policy. However, as this book tries to demonstrate,
and as we briefy discuss in the next section, this view of no viable alterna-
tives to infation targeting is not correct.
1.4 SOCIALLY RESPONSIBLE ALTERNATIVES TO
INFLATION TARGETING CENTRAL BANK
POLICIES
One reason that infation-focused monetary policy has gained so many
adherents is the common perception that there is no viable alternative
monetary policy that can improve growth and employment prospects.
There are two main factors accounting for this perception. First, as we dis-
cussed in the previous section, in an internationally fnancially integrated
economy with high levels of international capital fows, monetary policy
can be extremely challenging. In particular it might be very dif cult to
gear monetary policy by targeting monetary aggregates, or by pegging an
exchange rate along with trying to promote employment growth. This is
often seen as the so-called trilemma which commands that central banks
can only have two out of three of the following: open capital markets, a
fxed exchange rate system and an autonomous monetary policy geared
toward domestic goals. While this so-called trilemma is not strictly true
as a theoretical matter, in practice it does raise serious issues of mon-
etary management (see the above arguments cited from Taylor, 2004 and
Frenkel and Taylor, 2006). From our perspective, the real crux of the
problem turns out to be the very narrow interpretation of the constraints
of the trilemma: central banks are often thought to be restricted to choose
two points out of three. Yet, the constraints of the trilemma could as well
be regarded as the boundaries of a continuous set of policies, as would
emerge out of a bounded, yet continuous depiction of a policy triangle.
Thus even within the boundaries of the trilemma a menu of choices does
exist, ranging from administered exchange rate regimes to capital manage-
ment/control techniques. In fact, many successful developing countries
have used a variety of capital management techniques to manage these
fows in order, among other things, to help them escape the rigid con-
straints of the so-called tri-lemma (Epstein et al., 2005; Ocampo, 2002).
In this section we report on a series of country studies undertaken by a
team of researchers working on a Political Economy Research Institute
(PERI) (University of Massachusetts, Amherst)/Bilkent project on alter-
natives to infation targeting, as well as a United Nations Development
Programme (UNDP) sponsored study of employment targeting economic
policy for South Africa. A range of alternatives were developed by the
researchers, all the way from modest changes in the IT framework to
allow for more focus on exchange rates and a change in the index of infa-
tion used, to a much broader change in the overall mandate of the central
bank to a focus on employment targeting, rather than IT. Some of the
alternative policies focus exclusively on changes in central bank policy,
while for other countries changes in the broad policy framework and in the
interactions of monetary, fnancial and fscal policy are proposed. Some
incorporate explicit goals and targets, while others prefer more fexibility
and somewhat less transparency.
It has to be noted at the outset that infation control is revealed among
the ultimate objectives in all country studies summarized below. Thus
there is a clear consensus among the country authors that controlling
infation is important and desirable. However all agree that the current
prescription insisting on very low rates of infation at the 24 percent
band is not warranted, and that responsibilities of the central banks, par-
ticularly in developing countries, must be broader than that. Accordingly,
the policy matrix of the central banks should include other crucial real
variables that have a direct impact on employment, poverty and economic
growth, such as the real exchange rate and/or investment allocation. They
also agree that in many cases, central banks must broaden their available
policy tools to allow them to reach multiple goals, including, if necessary,
the implementation of capital management techniques.
Table 1.4 presents a summary of the alternatives proposed in the PERI/
Bilkent project and is discussed further in what follows.
1.4.1 Modest but Socially Responsible Adjustments to the Infation
Targeting Regime
Some of the country studies in the PERI/Bilkent project proposed only
modest changes to the IT regime. In the case of Mexico, for example, the
authors argue that the IT regime has allowed for more fexible monetary
policy than had occurred under regimes with strict monetary targets or
strict exchange rate targets (Galindo and Ros, Chapter 8, this volume).
They suggest modifying the IT framework to make it somewhat more
employment friendly. In the case of Mexico, Galindo and Ros fnd that
monetary policy was asymmetric with respect to exchange rate movements
tightening when exchange rates depreciated, but not loosening when
19
T
a
b
l
e

1
.
4

P
E
R
I
/
B
i
l
k
e
n
t

a
l
t
e
r
n
a
t
i
v
e
s

t
o

i
n
f

a
t
i
o
n

t
a
r
g
e
t
i
n
g

p
r
o
j
e
c
t
,

s
u
m
m
a
r
y

o
f

p
o
l
i
c
y

r
e
c
o
m
m
e
n
d
a
t
i
o
n
s
C
o
u
n
t
r
y
U
l
t
i
m
a
t
e

t
a
r
g
e
t
s
I
n
t
e
r
m
e
d
i
a
t
e

t
a
r
g
e
t
s
S
t
r
i
c
t

t
a
r
g
e
t

o
r

d
i
s
c
r
e
t
i
o
n
T
o
o
l
s
/
i
n
s
t
r
u
m
e
n
t
s
C
e
n
t
r
a
l

b
a
n
k
:

i
n
d
e
p
e
n
d
e
n
t
,

i
n
t
e
g
r
a
t
e
d

o
r

c
o
o
r
d
i
n
a
t
e
d
?
A
r
g
e
n
t
i
n
a
I
n
f

a
t
i
o
n

c
o
n
t
r
o
l
,

a
c
t
i
v
i
t
y

l
e
v
e
l

a
n
d

e
m
p
l
o
y
m
e
n
t

e
x
p
a
n
s
i
o
n
,

e
x
t
e
r
n
a
l

s
u
s
t
a
i
n
a
b
i
l
i
t
y
S
C
R
E
R
,

i
n
t
e
r
e
s
t

r
a
t
e
D
i
s
c
r
e
t
i
o
n
S
t
e
r
i
l
i
z
a
t
i
o
n
,

r
e
s
e
r
v
e

r
e
q
u
i
r
e
m
e
n
t
s

(
o
t
h
e
r

p
r
u
d
e
n
t
i
a
l

r
e
q
u
i
r
e
m
e
n
t
s
)
,

c
a
p
i
t
a
l

m
a
n
a
g
e
m
e
n
t

t
e
c
h
n
i
q
u
e
s
C
o
o
r
d
i
n
a
t
e
d
B
r
a
z
i
l
I
n
f

a
t
i
o
n

c
o
n
t
r
o
l
,

e
x
p
o
r
t

p
r
o
m
o
t
i
o
n
,

i
n
v
e
s
t
m
e
n
t

e
x
p
a
n
s
i
o
n
S
C
R
E
R
D
i
s
c
r
e
t
i
o
n
I
n
t
e
r
e
s
t

r
a
t
e
,

a
s
y
m
m
e
t
r
i
c

m
a
n
a
g
e
d

f

o
a
t

(
m
o
v
i
n
g

f

o
o
r

o
n

e
x
c
h
a
n
g
e

r
a
t
e
)
,

b
a
n
k

r
e
s
e
r
v
e
s
,

b
a
n
k

c
a
p
i
t
a
l

r
e
q
u
i
r
e
m
e
n
t
s
,

b
a
n
k

c
a
p
i
t
a
l

r
e
q
u
i
r
e
m
e
n
t
s
I
n
t
e
g
r
a
t
e
d
/
c
o
o
r
d
i
n
a
t
e
d

w
i
t
h

t
h
e

f

s
c
a
l

a
n
d

a
n
t
i
-
p
o
v
e
r
t
y

o
b
j
e
c
t
i
v
e
s
I
n
d
i
a
G
D
P

g
r
o
w
t
h
,

i
n
f

a
t
i
o
n

c
o
n
t
r
o
l
,

e
x
p
o
r
t

p
r
o
m
o
t
i
o
n
S
l
i
g
h
t
l
y

u
n
d
e
r
v
a
l
u
e
d

(
c
o
m
p
e
t
i
t
i
v
e
)

e
x
c
h
a
n
g
e

r
a
t
e
D
i
s
c
r
e
t
i
o
n
I
n
t
e
r
e
s
t

r
a
t
e
,

c
a
p
i
t
a
l

m
a
n
a
g
e
m
e
n
t

t
e
c
h
n
i
q
u
e
s
,

i
f

n
e
c
e
s
s
a
r
y
I
n
t
e
g
r
a
t
e
d
20
T
a
b
l
e

1
.
4

(
c
o
n
t
i
n
u
e
d
)
C
o
u
n
t
r
y
U
l
t
i
m
a
t
e

t
a
r
g
e
t
s
I
n
t
e
r
m
e
d
i
a
t
e

t
a
r
g
e
t
s
S
t
r
i
c
t

t
a
r
g
e
t

o
r

d
i
s
c
r
e
t
i
o
n
T
o
o
l
s
/
i
n
s
t
r
u
m
e
n
t
s
C
e
n
t
r
a
l

b
a
n
k
:

i
n
d
e
p
e
n
d
e
n
t
,

i
n
t
e
g
r
a
t
e
d

o
r

c
o
o
r
d
i
n
a
t
e
d
?
M
e
x
i
c
o
I
n
f

a
t
i
o
n
,

S
C
R
E
R
I
m
p
l
e
m
e
n
t
a
t
i
o
n

o
f

d
o
m
e
s
t
i
c

i
n
f

a
t
i
o
n

m
e
a
s
u
r
e
,

S
C
R
E
R
,

s
l
i
d
i
n
g

f

o
o
r

o
n

e
x
c
h
a
n
g
e

r
a
t
e
D
i
s
c
r
e
t
i
o
n
C
a
p
i
t
a
l

m
a
n
a
g
e
m
e
n
t

t
e
c
h
n
i
q
u
e
s
I
n
t
e
g
r
a
t
e
d
S
o
u
t
h

A
f
r
i
c
a
E
m
p
l
o
y
m
e
n
t

g
e
n
e
r
a
t
i
o
n
,

i
n
f

a
t
i
o
n

c
o
n
t
r
o
l
R
e
a
l

G
D
P

g
r
o
w
t
h
S
t
r
i
c
t

e
m
p
l
o
y
m
e
n
t

t
a
r
g
e
t

(
c
o
o
r
d
i
n
a
t
e
d

w
i
t
h

o
t
h
e
r

i
n
s
t
i
t
u
t
i
o
n
s
)
,

l
o
o
s
e
r

i
n
f

a
t
i
o
n

c
o
n
s
t
r
a
i
n
t
I
n
t
e
r
e
s
t

r
a
t
e
;

c
r
e
d
i
t

a
l
l
o
c
a
t
i
o
n

t
e
c
h
n
i
q
u
e
s

(
e
.
g
.

a
s
s
e
t
-
b
a
s
e
d

r
e
s
e
r
v
e

r
e
q
u
i
r
e
m
e
n
t
s
,

l
o
a
n

g
u
a
r
a
n
t
e
e
s

e
t
c
.
)
,

c
a
p
i
t
a
l

m
a
n
a
g
e
m
e
n
t

t
e
c
h
n
i
q
u
e
s
I
n
t
e
g
r
a
t
e
d
T
u
r
k
e
y
I
n
f

a
t
i
o
n

c
o
n
t
r
o
l
;

e
m
p
l
o
y
m
e
n
t

g
e
n
e
r
a
t
i
o
n
;

s
o
l
v
e
n
c
y

o
f

p
u
b
l
i
c

d
e
b
t
;

c
o
n
s
o
l
i
d
a
t
e

a
n
d

e
x
p
a
n
d

s
o
c
i
a
l

i
n
f
r
a
s
t
r
u
c
t
u
r
e
R
e
a
l

i
n
t
e
r
e
s
t

r
a
t
e
,

n
o
n
-
a
p
p
r
e
c
i
a
t
e
d

e
x
c
h
a
n
g
e

r
a
t
e
D
i
s
c
r
e
t
i
o
n
C
a
p
i
t
a
l

m
a
n
a
g
e
m
e
n
t

t
e
c
h
n
i
q
u
e
s

i
f

n
e
c
e
s
s
a
r
y
,

l
a
b
o
r
-
t
a
x

r
e
f
o
r
m
,

i
n
c
r
e
a
s
e
d

p
u
b
l
i
c

i
n
v
e
s
t
m
e
n
t
s

i
n

s
o
c
i
a
l

c
a
p
i
t
a
l
I
n
t
e
g
r
a
t
e
d
/
c
o
o
r
d
i
n
a
t
e
d

w
i
t
h

t
h
e

f

s
c
a
l

a
n
d

e
m
p
l
o
y
m
e
n
t

o
b
j
e
c
t
i
v
e
s
21
P
h
i
l
i
p
p
i
n
e
s
I
n
f

a
t
i
o
n

c
o
n
t
r
o
l
;

s
o
l
v
e
n
c
y

o
f

p
u
b
l
i
c

d
e
b
t
S
C
R
E
R
D
i
s
c
r
e
t
i
o
n
I
n
t
e
r
e
s
t

r
a
t
e
,

c
a
p
i
t
a
l

m
a
n
a
g
e
m
e
n
t

t
e
c
h
n
i
q
u
e
s
,

p
r
u
d
e
n
t
i
a
l

s
u
p
e
r
v
i
s
i
o
n

o
f

b
a
n
k
s
,

t
a
r
g
e
t
e
d

c
r
e
d
i
t
;

i
n
c
o
m
e
s

p
o
l
i
c
i
e
s
I
n
t
e
g
r
a
t
e
d
V
i
e
t
n
a
m
I
n
c
o
m
e
s

g
r
o
w
t
h
,

e
x
p
o
r
t

p
r
o
m
o
t
i
o
n
,

i
n
f

a
t
i
o
n

c
o
n
t
r
o
l
S
C
R
E
R
D
i
s
c
r
e
t
i
o
n
I
n
t
e
r
e
s
t

r
a
t
e
,

c
a
p
i
t
a
l

m
a
n
a
g
e
m
e
n
t

t
e
c
h
n
i
q
u
e
s
,

p
r
u
d
e
n
t
i
a
l

s
u
p
e
r
v
i
s
i
o
n

o
f

b
a
n
k
s
,

t
a
r
g
e
t
e
d

c
r
e
d
i
t
,

i
n
c
o
m
e
s

p
o
l
i
c
i
e
s
I
n
t
e
g
r
a
t
e
d
N
o
t
e
:

S
C
R
E
R
:

S
t
a
b
l
e

a
n
d

C
o
m
p
e
t
i
t
i
v
e

R
e
a
l

E
x
c
h
a
n
g
e

R
a
t
e
;

C
e
n
t
r
a
l

B
a
n
k
s
:

i
n
t
e
g
r
a
t
e
d

m
e
a
n
s

i
n
t
e
g
r
a
t
e
d

i
n
t
o

g
o
v
e
r
n
m
e
n
t
a
l

m
a
c
r
o
e
c
o
n
o
m
i
c

p
o
l
i
c
y

m
a
k
i
n
g

f
r
a
m
e
w
o
r
k
;

c
o
o
r
d
i
n
a
t
e
d

m
e
a
n
s

i
n
d
e
p
e
n
d
e
n
t

b
u
t

c
o
m
m
i
t
t
e
d

t
o

c
l
o
s
e

c
o
o
r
d
i
n
a
t
i
o
n

w
i
t
h

o
t
h
e
r

m
a
c
r
o
e
c
o
n
o
m
i
c

p
o
l
i
c
y

m
a
k
i
n
g

i
n
s
t
i
t
u
t
i
o
n
s
.
S
o
u
r
c
e
:

I
M
F

(
2
0
0
8
)
.
exchange rates appreciated. This lent a bias in favor of an over-valued
exchange rate in Mexico. So they propose a neutral monetary policy so
that the central bank of Mexico responds symmetrically to exchange rate
movements and thereby avoids the bias toward over-valuation without
fundamentally changing the IT framework. In their own words, the
central bank would promote a competitive exchange rate by establishing a
sliding foor to the exchange rate in order to prevent excessive appreciation
(an asymmetric band. . .). This would imply intervening in the foreign
exchange market at times when the exchange rate hits the foor (i.e. an
appreciated exchange rate) but allows the exchange rate to foat freely
otherwise. They point out that such a foor would work against excessive
capital infows by speculators because they would know the central bank
will intervene to stop excessive appreciation. If need be, Galindo and Ros
also propose temporary capital controls, as do some of the other authors
from the PERI/Bilkent project.
In his study of Brazil Nelson Barbosa-Filho 2008 Chapter 7, (this
volume) also proposed extending the IT framework, but in a more dra-
matic way. According to Barbosa-Filho: because of Brazils past experi-
ence with high infation, the best policy is to continue to target infation
while the economy moves to a more stable macroeconomic situation.
However, the crucial question is not to eliminate infation targeting, but
actually make it compatible with fast income growth and a stable public
and foreign fnance.
Given Brazils large public debt, Barbosa-Filho proposes that the tar-
geted reduction in the real interest rate would reduce the Brazilian debt
service burdens and help increase productive investment. In terms of the
familiar targets and instruments framework, he proposes that the Brazilian
central bank choose exports, infation and investment as ultimate targets,
and focus on the infation rate, a competitive and stable real exchange
rate and the real interest rate as intermediate targets. Furthermore, in
order to achieve these goals, the central bank can use direct manipulation
of the policy interest rate, bank reserve requirements and bank capital
requirements.
Brazil is not the only highly indebted country in our project sample.
Turkey is another case with that problem. Here too the authors raise con-
cerns to the conformist straightjacket of IT, and develop an alternative
macroeconomic framework. Using a fnancial-linked computable general
equilibrium model (CGE) for the case of Turkey, Telli et al. (Chapter
10, this volume) illustrate the real and fnancial sectorial adjustments of
the Turkish economy under the conditionalities of the twin targets: on
primary surplus to GNP ratio and on the infation rate. They utilize their
model to study the impact of a shift in policy from a strict IT regime, to
one that calls for revisions of the primary fscal surplus targets in favor
of a more relaxed fscal stance on public investments on social capital,
together with a direct focus on the competitiveness of the real exchange
rate. They further study the macroeconomics of a labor tax reform imple-
mented through reduction of the payroll tax burden on the producers,
and an active monetary policy stance via reduction of the central banks
interest rates. They report signifcant employment gains due to a policy of
lower employment taxes. They also fnd that the economys response to
the reduction of the central banks interest rate is positive in general; yet
very much dependent on the path of the real exchange rate, thus they also
call for maintaining a stable real exchange rate path la Frenkel, Ros and
Taylor.
Frenkel and Rapetti (Chapter 9, this volume), in the case of Argentina,
show that targeting a stable and competitive real exchange rate has been
very successful in helping to maintain more rapid economic growth
and employment generation. In the case of India, Jha (Chapter 12, this
volume) also argues against an IT regime, and in favor of one that errs on
the side of undervaluation of the exchange rate with possible help from
temporary resort to capital controls. Jha argues that, to some extent, such
a policy would be a simple continuation of policies undertaken in India in
the past. In Vietnam Packard concludes: a strict infation targeting regime
is not appropriate for Vietnam. Infation targetings rigid rules constrain
policy makers to operate in a framework that requires infation to take
priority over more pressing development objectives. Thus, a stable and
competitive real exchange rate is a superior alternative, precisely because it
sets as a target a key macroeconomic relative price that is realistic, sustain-
able and growth enhancing (Packard, Chapter 14, this volume).
1.4.2 More Comprehensive Alternatives to Infation Targeting
Other country case studies propose more comprehensive policy alterna-
tives to simple infation-focused monetary policy, including IT. Joseph
Lim (Chapter 13, this volume) proposes a comprehensive alternative to
IT for the case of the Philippines. He argues that the Philippines govern-
ment has been seeking to achieve a record of dramatically higher economic
growth, but that its monetary policy has been inadequate to achieving that
goal. He therefore proposes an alternative that clearly dictates much
more than just a move from monetary targeting to infation targeting with
the following proposals: (1) Maintenance of a competitive real exchange
rate, either by pegging the exchange rate or intensively managing it as in
South Korea. (2) Implementation of capital management techniques, as
in China and Malaysia, to help manage the exchange rates. This should
include strong fnancial supervision to prevent excessive undertaking of
short-term foreign debt, and tax based capital controls on short-term
capital fows, as was used, for example in Chile. (3) An explicit statement
of output and employment goals, as the central bank transits from a
purely IT regime. (4) Incomes and anti-monopoly policies to limit infa-
tion to moderate levels. (5) Targeted credit programs, especially for export
oriented and small and medium sized enterprises that can contribute to
productivity growth and employment.
These policy proposals in broad outlines are similar to those proposed
by Epstein (Chapter 11, this volume) for the case of South Africa, which,
in turn, have been developed in a much broader framework and in more
detail by Pollin et al. (2006). Pollin et al. developed an employment-
targeted economic program designed to accomplish this goal, with a
focus on monetary policy, credit policy, capital management techniques,
fscal policy and industrial policy. The purpose of the program is to reduce
unemployment rate by half in line with the governments pledge to reduce
the of cial unemployment rate to 13 percent by 2014.
5
Here, employment
targeting replaces IT as the proposed operating principle behind central
bank policy, and moderate infation becomes an additional formal con-
straint which the central bank must take into account when formulating
its policies.
1.5 CONCLUDING COMMENTS
In this introductory chapter we have argued that the current day ortho-
doxy of central banking namely, that the top priority goal for central
banks is to keep infation in the low single digits is, in general, neither
optimal nor desirable. This orthodoxy is based on several false premises:
frst, that infation, in any magnitude, has high costs; second, that in a
low infation environment economies will naturally perform best, and in
particular, will generate high levels of economic growth and employment;
and third, that there are no viable alternatives to this infation-focused
monetary policy.
In fact, moderate rates of infation episodes reveal to have very low or
no costs; and whether countries where central banks have adopted formal
or informal IT have not performed better in terms of economic growth or
employment generation is a matter of dispute. Per contra, there are viable
alternatives to IT, historically, currently, and looking forward.
Historically, countries both in the currently developed and develop-
ing worlds had central banks with multiple goals and tools, and pursued
broad developmental as well as stabilization goals. Currently, very
successful economies such as Argentina, China and India have central
banks that are using a broad array of tools to manage their economies
for developmental purposes. And looking forward, the PERI/Bilkent
project on alternatives to IT and PERIs UNDP work on South Africa
have developed an array of real targeting approaches to central banking
which we believe are viable alternatives to IT and, in particular, do a
better job than mere IT in balancing the developmental and stabilization
functions of central banks.
NOTES
1. We are indebted to Hasan Comert, Luis Rosero and Lynda Pickbourn for their
diligent research assistance, and to Roberto Frenkel, Jose Antonio Ocampo, K.S.
Jomo, Geofrey Woglom, Refet Grkaynak, James Heintz, Leonce Ndikumana, Arjun
Jayadev and Robert Pollin for their valuable comments and suggestions on previous
versions of the chapter. Research for this chapter was completed when Yeldan was a
visiting Fulbright scholar at the University of Massachusetts, Amherst for which he
acknowledges the generous support of the J. William Fulbright Foreign Scholarship
Board and the hospitality of the Political Economy Research Institute at the University
of Massachusetts, Amherst. We are also grateful to the funders of the PERI/Bilkent
Alternatives to Infation Targeting project, including UN-DESA, Ford Foundation,
Rockefeller Brothers Fund and PERI for their support. Needless to mention, the views
expressed in the chapter are solely those of the authors and do not implicate in any way
the institutions mentioned above.
2. Note that with the use of the term conditionality here we refer not to the IMFs stand-
by rules in the narrow sense of balance of payments stabilization, but to the broader set
of reforms and structural adjustment agenda as advocated by the international fnance
community and the transnational corporations. Often dubbed as the (post-) Washington
consensus, the warranted set of policies range from IT central banks and fexible foreign
exchange markets to broader institutional reforms such as fexible labor markets, priva-
tization and increased governance. See Williamson (1993) for the original deployment of
the term, and Rodrik (2003) for further discussion.
3. Brazils case is actually explained in part by the recent decision (late 2005) of the Lula
government to close its debt arrears with the IMF with early payments out of its
reserves.
4. Though note the one sided ever increase in the aggregate reserves of the central banks.
The social desirability and economic optimality of this phenomenon in the aftermath
of the adoption of foating exchange rate systems is another issue that warrants further
research.
5. As of March 2005, South Africa had an unemployment rate of anywhere from 26 percent
to 40 percent, depending on exactly how it is counted.
REFERENCES
Adelman, I. and E. Yeldan (2000), The minimal conditions for a fnancial
crisis: a multi-regional inter-temporal CGE model of the Asian crisis, World
Development, 28 (6), 1087100.
Akyuz, Y. (2003), The Developing Countries and World Trade. Performance and
Prospects, London: ZED Books.
Akyuz, Y. (2006), From liberalization to investment and jobs: lost in transla-
tion, paper presented at the Carnegie Endowment for International Peace
Conference, 14-15 April 2005, Washington, DC, pp. 456.
Akyuz, Y., H. Flassback and R. Kozul-Wright (2006), Globalization, inequal-
ity and the labour market, in A. Kose, F. Senses and E. Yeldan (eds),
Neoliberal Globalization as New Imperialism: Case Studies on Reconstruction of
the Periphery, New York: NOVA Science.
Ball, L. and N. Sheridan (2003), Does infation targeting matter?, International
Monetary Fund working paper no. 03/129.
Batini, N., P. Breuer, K. Kochhar and S. Roger (2006), Infation targeting and the
IMF, International Monetary Fund staf paper, Washington, DC, March.
Bernanke, B.S., T. Laubach, A.S. Posen and F.S. Mishkin (1999), Infation
Targeting: Lessons from the International Experience, Princeton, NJ: Princeton
University Press.
Buiter, W. (2006), Rethinking infation targeting and central bank independence,
paper presented at the Turkish Economic Association Conference, Ankara,
September.
Corbo, V., O. Landerretche and K. Schmidt-Hebbel (2001), Does infation target-
ing make a diference? Central Bank of Chile working paper no. 106, accessed 15
January 2007 at www.bcentral.cl/Estudios/DTBC/doctrab.htm.
Debelle, G. (2001), The case for infation targeting in East Asian countries, in
Future Directions for Monetary Policies in East Asia, Sydney: Reserve Bank of
Australia, pp. 6587.
Edwards, S. (2005), The relationship between exchange rates and infation tar-
geting revisited, National Bureau of Economic Research, working paper no.
W12163.
Epstein, G., I. Grabel and K.S. Jomo (2005), Capital management techniques in
developing countries: an assessment of experiences from the 1990s and lessons
for the future, in Gerald Epstein (ed.), Capital Flight and Capital Controls in
Developing Countries. Cheltenham, UK and Northampton, MA, USA: Edward
Elgar, pp. 30133.
Feldstein, M. (1997), Capital income taxes and the benefts of price stability,
National Bureau of Economic Research working paper no. 6200, September.
Fischer, S. (2001), Distinguished lecture on economics in government, Journal of
Economic Perspectives, 15 (2), 324.
Fleming, J. Marcus (1962), Domestic fnancial policies under fxed and under fex-
ible exchange rates, International Monetary Fund staf papers 9, p. 36979.
Freeman, R. (2004), Doubling the global workforce: the challenges of integrat-
ing China, India, and the former Soviet Bloc into the world economy, paper
presented at the conference on Doubling the Global Work Force, Institute of
International Economics, 8 November, Washington, DC.
Freeman, Richard B. (2007), The great doubling: is your job going to Bombay
or Beijing?, in America Works; Critical Thoughts an the Exceptional US Labor
Market, New York: Russell Sage Foundation, pp. 12840.
Frenkel, R. and J. Ros (2006), Unemployment and the real exchange rate in Latin
America, World Development, 34 (4), 63146.
Frenkel, R. and L. Taylor (2008), Real exchange rate, monetary policy, and
employment, United Nations Department of Economic and Social Afairs
(DESA) working paper no 19, February, New York.
Heintz, J. (2006), Globalization, economic policy and employment: poverty and
gender implications, Employment Strategy Department employment strategy
paper 2006/3, International Labour Organization, Geneva.
International Labour Organization (ILO) (2004), Global Employment Trends 2004,
Geneva: ILO.
International Monetary Fund (IMF) (2008) International Financial Statistics, CD
Rom, Washington, DC: IMF.
Michl, Thomas (2007), Tinbergen rules the Taylor rule, Levy Economics Institute
working paper no. 444.
Mishkin, F.S. and K. Schmidt-Hebbel (2001), One decade of infation targeting in
the world: what do we know and what do we need to know?, National Bureau
of Economic Research working paper no 8397, July accessed at www.nber.org/
papers/w8397.
Mundell, R.A. (1963), Capital mobility and stabilization policy under fxed and
fexible exchange rates, Canadian Journal of Economics and Political Science,
29, 47585.
Ocampo, J.A. (2002), Capital-account and counter-cyclical prudential regulations
in developing countries, World Institute for Development Economics Research
discussion paper, August.
Pollin, R. and A. Zhu (2006), Infation and economic growth: a cross-country
nonlinear analysis, Journal of Post Keynesian Economics, 28 (4) (Summer),
593614.
Pollin, R., G. Epstein, J. Heintz and L. Ndikumana (2006), An Employment-Targeted
Economic Program for South Africa, Cheltenham, UK and Northampton, MA,
USA: Edward Elgar.
Rodrick, D. (2003), Growth strategies, National Bureau of Economic Research
Roger, S. and M. Stone (2005), On target? The international experience with
achieving infation targets, International Monetary Fund working paper
WP/05/163, Washington, DC.
Setterfeld, M. (2006), Is infation targeting compatible with post keynesian eco-
nomics?, Journal of Post Keynesian Economics, 28 (4), 65371.
Taylor, J.B. (2000), Low infation, pass-through and the pricing power of frms,
European Economic Review, 7 (44), 1389408.
Taylor, L. (2004), Exchange rate indeterminacy in portfolio balance, Mundell-
Fleming and uncovered interest parity models, Cambridge Journal of Economics,
28, 20527.
Tobin, J. (1980), Stabilization policy ten years after, Brookings Papers on
Economic Activity, 1, 1971.
Williamson, J. (1993), Democracy and the Washington Consensus, World
Development, 21 (8), 132936.
28
2. Real exchange rate, monetary
policy and employment: economic
development in a garden of forking
paths
Roberto Frenkel and Lance Taylor
Dejo a los varios porvenires (no a todos) mi jardn de senderos que se bifurcan.
[I am leaving my garden of diverse paths to some (but not all)]
Jorge Luis Borges, El jardn de los senderos que se bifurcan
2.1 INTRODUCTION
The exchange rate afects any economy through many channels. It scales
the national price system to the worlds, infuences key macro price ratios
such as those between tradable and non-tradable goods, capital goods and
labor, and even exports and imports (via the costs of intermediate inputs
and capital goods, for example). The exchange rate is an asset price, par-
tially determines infation rates through the cost side and as a monetary
transmission vector, and can have signifcant efects (both short and long
run) on efective demand.
Correspondingly the exchange rate can be targeted toward many policy
objectives. In developing and transition economies fve have been of
primary importance in recent decades:
1. Resource allocation: through its efects on the price ratios just men-
tioned, the exchange rate can signifcantly infuence resource alloca-
tion, especially if it stays stable in real terms for an extended period
of time. Through efects on both resource allocation and aggregate
demand, a relatively weak rate can help boost employment, a point
of concern in light of stagnant job creation in many developing econ-
omies over the past 1015 years.
2. Economic development: often in conjunction with commercial
and industrial policies, the exchange rate can be deployed to
Real exchange rate, monetary policy and employment 29
enhance overall competitiveness and thereby boost productivity
and growth.
3. Finance: the rate shapes and can be used to control expectations and
behavior in fnancial markets. Exchange rate policy mistakes can
easily lead to highly destabilizing consequences.
4. External balance: the trade and other components of the current
account usually respond to the exchange rate, directly via substitu-
tion responses and (at times more importantly) to shifts it can cause
in efective demand.
5. Infation: the exchange rate can serve as a nominal anchor, holding
down price increases via real appreciation and/or maintenance by the
authorities of a consistently strong rate. As will be seen below, it can
also serve as an important transmission mechanism for the efects of
monetary policy.
All these objectives have fgured in recent policy experience. Use of
the exchange rate to try to improve external balance has been central
to countless stabilization packages over the decades, especially in small
poor economies. The infation objective became crucial in middle-
income countries in the last quarter of the twentieth century (and is
notably less urgent as of 2005). Along with capital market liberalization,
fxed rates were signifcant contributors to the wave of fnancial crises
in the 1990s.
But in many ways the resource allocation and developmental objec-
tives can be the most important in the long run the central point of this
chapter. We trace the reasons why in the following section on channels of
infuence. We then take up the policy implications, contrasting the use of
the exchange rate as a development tool in conjunction with its other uses
(often in coordination with monetary policy) to maintain external balance,
contain infation and stabilize asset markets,
2.2 RESOURCE ALLOCATION, LABOR
INTENSITY, MACROECONOMICS AND
DEVELOPMENT
Following Frenkel (2004), in this section we trace out three ways in which
the exchange rate can have medium- to long-term impacts on develop-
ment. We begin with overall resource allocation, and go on to the labor
market and macroeconomics.
2.2.1 Resource Allocation
The traditional 2 3 2 trade theory model is a useful starting point. It does
focus on the key role of relative prices. It does not take into consideration
important non-price components of industrial and commercial polices.
Both themes are woven into the following discussion.
The Lerner Symmetry Theorem (1936) is a key early result. Its basic
insight is that if only the import/export price ratio is relevant to resource
allocation, then it can be manipulated by either an import or an export
tax-cum-subsidy. There is symmetry between the two instruments, so
that under appropriate conditions (at hand in the textbooks) only one
need be employed.
A now obvious extension is to bring three goods into the discussion:
exportable, importable and non-tradable in a Ricardo-Viner model. Two
price ratios say importable/non-tradable and exportable/non-tradable
in principle guide allocation. The real exchange rate (RER or r) natu-
rally comes into play as the relative price between the non-tradable and a
Hicksian aggregate of the two tradable goods.
1
These observations lead to
two important policy puzzles.
The frst has to do with level playing felds. As applied in East Asia and
elsewhere, industrial policy often involved both protection of domestic
industry against imports by the use of tarifs and quotas, and promotion
of exports through subsidies or cheap credits. In the case of a tarif on
imports, the domestic price P
m
becomes
P
m
5 e(1 1 t)P*
m
(2.1)
with e as the nominal exchange rate (defned as units of local currency per
unit of foreign), t the tarif, and P*
m
the world price. Similarly if the internal
price P
e
for exports is set from abroad we have
P
e
5 eP*
e
/ (1 2 s) (2.2a)
with P*
e
as the world price and s as the subsidy rate.
The level playing feld rests on the trade theorists notion that inter-
nal and external relative prices of tradable goods should be equal,
P
m
/P
e
5 P*
m
/P*
e
. This situation can be arranged if t 5 s 5 0 or more
generally (1 1 t) 5 1/ (1 2 s). The mainstream argument asserts that if
all that industrial policy does is give more or less equal protection to both
imports and exports, then its costs, administrative complications, and
risks of rent-seeking and corruption are unjustifable. You might as well
set t 5 s 5 0 and go to a free trade equilibrium.
In a Ricardo-Viner set-up, with P
n
as a price index for non-tradables the
price ratios P
e
/P
n
and P
m
/P
n
become of interest. Positive values of t and s
move domestic relative prices in favor of tradable goods. From a more or
less mainstream perspective (Woo, 2005) this outcome can be interpreted
as a justifcation for industrial policy.
The world, however, is a bit more complicated. If the home country
is exporting a diferentiated product, for example, a more appropriate
version of Equation (2.2a) is
P*
e
5 P
e
(1 2 s) /e (2.2b)
so that the foreign price of home exports is set by the subsidy and exchange
rate. Presumably, a lower value of P*
e
stimulates sales abroad. Moreover,
if the economic bureaucracy has the requisite motivation and organiz-
ation, it can tie export subsidies to the attainment of export, productivity
and other targets and so pursue a proactive industrial policy. In such a
context, import protection and export promotion serve diferent purposes:
the former allows domestic production to get started along traditional
infant industry lines, while the latter enables national frms to break into
international markets.
2
Now focus on the exchange rate. An increase in the nominal rate e
would also switch incentives toward production of tradables, without the
need for extravagant values of s and t. This simple observation is in fact a
strong argument in support of the use of a depreciated RER as a develop-
mental tool. If we defne r as
r 5 [mP
m
1 (1 2 m)P
e
] /P
n
(2.3)
with m as the weight in a tradable goods price index, then a high value of e
means that the RER r will also be weak or depreciated.
Of course, a weak RER may not be a suf cient condition for long-term
development. For example, it may usefully be supplemented by an export
subsidy or tarif protection to infant industries with their additional
potential benefts as mentioned above. Even without an efective bureauc-
racy, generalizing Lerner symmetry to a Ricardo-Viner world suggests
that more than one policy instrument may be helpful because there are
two relative price ratios that can be manipulated (Lerner, 1936). The rub
is that a strong exchange rate implies that commercial/industry policy
interventions also have to be strong, with correspondingly high interven-
tion costs. A weak RER may be only a necessary condition for benefcial
resource reallocation to occur, but a highly appreciated RER is likely to
be a suf cient condition for excessive intervention in a situation in which
development cannot happen. It is hard to fnd examples of economies with
strong exchange rates that kept up growth for extended periods of time.
2.2.2 Labor Intensity
Continuing with the allocational theme, it is clear that the exchange rate
will afect relative prices of imported intermediates and capital goods on
the one hand, and labor on the other. Moreover, the RER largely deter-
mines the economys unit labor costs in terms of foreign currency.
To explore the implications, we can consider the efects of sustained
real appreciation on diferent sectors. Producers of importables will face
tougher foreign competition. To stay in business they will have to cut
costs, often by shedding labor. If they fail and close down, more jobs will
be destroyed. If homes export prices P*
e
are determined by a relationship
like Equation (2.2b), similar logic applies to that sector. In non-tradables,
which will have to absorb labor displaced from the tradable sectors,
jobs are less likely to open up insofar as cheaper foreign imports in the
form of intermediates and capital goods substitute for domestic labor.
On the whole, real appreciation is not likely to induce sustained job
creation and could well provoke a big decrease in tradable sector employ-
ment. Reasoning in the other direction, RER depreciation may prove
employment-friendly.
In both cases it is important to recognize that a new set of relative prices
must be expected to stay in place for a relatively long period if these efects
are going to work through. Changes in employment/output ratios will
not happen swiftly because they involve restructuring frms and sectoral
labor market behavior. This must take place via changes in the pattern of
output among frms and sectors, by shifts in the production basket of each
frm and sector, and adjustments in the technology and organization of
production. These efects arise from a restructuring process in which indi-
vidual frms and the organization of economic activity adapt to a new set
of relative prices. Gradual adjustment processes are necessarily involved.
Finally, in the long run if per capita income is to increase there will have
to be sustained labor productivity growth with employment creation sup-
ported by even more rapid growth in efective demand. Macroeconomics
comes into play.
2.2.3 Macroeconomics
The question is how a weak exchange rate (possibly in combination with
other policies aimed at infuencing resource allocation among traded
goods) fts into the macroeconomic system. Much depends on labor
market behavior in the non-traded sector. Following Rada (2005) we work
through one scenario here, to illustrate possible outcomes.
Assume that output in the tradable sector is driven by efective demand,
responding to investment, exports and import substitution as well as
fscal and monetary policy. The level of imports depends on economic
activity and the exchange rate (along with commercial/industrial policies).
A worker not utilized in tradable sectors must fnd employment in non-
tradables, become under- or unemployed, or leave the labor force.
For concreteness, we assume that almost all labor not employed in
tradables fnds something to do in non-tradable production as a means
of survival. Typical activities would be providing labor services in urban
areas or engaging in labor-intensive agriculture. If L
t
is tradable sector
employment and L is the economically active population, then employ-
ment in non-tradables is L
n
5 L 2 L
t
. With w
n
as the non-tradable wage,
the value of labor services provided is Y
n
5 w
n
L
n
. The tradable sector
wage rate w
t
is determined institutionally, at a level substantially higher
than w
n
.
The non-tradable sectors demand-supply balance thus takes the form
Y
n
2 w
n
L
n
5 Y
n
2 w
n
(L 2 L
t
) 5 0 (2.4)
Demand for Y
n
is generated from the value of tradable sector output P
t
X
t
.
At the same time, real output X
t
determines L
t
and thereby L
n
. Suppose
that P
t
is set by mark-up pricing on variable costs including labor and
imports. Then from both the demand and supply sides an increase in
X
t
leads to a tighter non-traded labor market which should result in an
increase in w
n
. Equation (2.4) becomes the upward-sloping Non-tradable
equilibrium schedule in Figure 2.1. Non-tradable labor services become
more valuable when economic activity rises. In national accounting terms
this signals a productivity increase in the sector because each worker pro-
ducers a higher value of output in terms of tradable goods, or a general
price index. In other words, an endogenous productivity level is built into
the specifcation.
If workers in both sectors dont save, then their behavior does not infu-
ence overall macroeconomic balance. Leaving aside a formal treatment of
fscal and monetary instruments for simplicity, the equation takes the form
I
t
1 E
t
2 spX
t
2 eP*
m
(1 1 t)aX
t
/P
t
5 0 (2.5)
Demand injections come from investment I
t
, exports E
t
and changes in
the magnitude of the import coef cient a via import substitution. Saving
leakages come from profts with p as the tradable sector proft share and
s the saving rate as well as from foreign saving in the form of imports.
Equation (2.5) is the vertical Macroeconomic equilibrium line in Figure
2.1. Together, the two schedules determine X
t
and w
n
. In the lower quad-
rant the trade defcit is assumed to be an increasing function of tradable
sector output in the short run.
Now consider the outcomes of a devaluation. It will have impacts
all over the economy, including a loss in national purchasing power if
imports initially exceed exports, redistribution of purchasing power away
from low-saving workers whose real wages decrease, a decline in the real
value of the money stock and capital losses on the part of net debtors in
international currency terms. Presumably exports will respond positively
to an RER depreciation but that may take time if J-curve and similar
efects matter. Another positive impact on the demand for tradables will
Trade deficit
Non-tradable
sector wage
w
n
Tradable sector
output X
t
Macroeconomic equilibrium
Non-tradable
equilbrium
Figure 2.1 Equilibrium between tradable and non-tradable sectors
come from import substitution, reducing the magnitude of the coef cient
a.
One implication is that for a given level of output, the trade defcit
should fall with devaluation, or the corresponding schedule should shift
toward the horizontal axis in the lower quadrant. If devaluation is con-
tractionary, the macroeconomic equilibrium schedule will shift leftward
in the upper quadrant, reducing X
t
, w
n
, and the trade defcit further still.
In this case real devaluation should presumably be implemented together
with expansionary fscal and monetary policies. As discussed in detail
below, exchange rate strategies must be coordinated with other policy
moves.
If export demand and production of import substitutes are stimulated
immediately or over time by a sustained weak RER, the macroeconomic
equilibrium curve should drift to the right, driving up economic activity
and employment in the medium to long run.
So far, the analysis has taken labor productivity as a constant. Medium-
and long-run considerations have to take into account the evolution of
productivity. For the tradable sector, this question can be analysed in
terms of Figure 2.2, sketched verbally but not actually drawn by Kaldor
in his 1966 Inaugural Lecture (published in Kaldor, 1978). To the tra-
ditional diagram we follow Rada and Taylor (2004) by adding dashed
Output growth rate
X

t
Labor productivity
growth rate

Lt
Kaldor-Verdoorn
Output growth
Employment
growth
contours
Figure 2.2 Output, labor productivity and employment growth in the
tradable sector
Employment growth contours with slopes of 45 degrees. Each one shows
combinations of the output growth rate (X
^
t
5 (dX
t
/dt) /X
t
5 X
#
t
/X
t
) and
labor productivity growth rate (x
Lt
) that hold the employment growth
rate (L
^
t
5 X
^
t
2 x
Lt
) constant. Employment growth is more rapid along
contours further to the south east.
Movements across contours show the efects on employment growth
of shifts in the diagrams two solid curves. The Kaldor-Verdoorn sched-
ule represents a technical progress function of the form proposed by
Verdoorn (1949) and Okun (1962),
x
Lt
5 x
Lt
1 gX
^
t
(2.5)
in which the productivity trend term x
Lt
could be afected by human capital
growth, industrial policy, international openness, population growth and
other factors.
The Output growth curve refects the assumption that more rapid pro-
ductivity growth can make output expand faster, for example, by reducing
the unit cost of exports. The diagram presupposes that this efect is rather
strong because the slope of the Output growth line is less than 45 degrees,
implying that 0X
^
t
/0x
Lt
. 1.
If a depreciated RER stimulates net export growth, the Output growth
curve will shift to the right, causing x
Lt
, X
$
t
and L
^
t
all to increase. One might
also imagine that the trend rate x
Lt
of productivity growth could rise in the
new regime. The Kaldor-Verdoorn schedule would shift upward, and with
a relatively fat output growth curve, all three growth rates would rise.
However, if the slope of the Output growth curve were to exceed 45
degrees, efective demand would not increase as rapidly as productivity so
that L
^
t
would have to fall.
What happens to wages and productivity in the non-tradable sector?
Let l 5 L
t
/L be the share of tradable sector employment in the total.
Then lL
^
t
1 (1 2 l) L
^
n
5 L
^
where L
^
is overall employment growth. Non-
tradable employment expansion becomes
L
^
n
5
1
1 2 l
[L
^
2 l(X
^
t
2 x
Lt
) ]. (2.6)
Let the elasticity of demand for non-tradables with respect to X
t
be u.
Diferentiating Equation (2.4) then gives
w^
n
5 uX
^
t
1
1
1 2 l
[l(X
^
t
2 x
Lt
) 2 L
^
] (2.7)
Even taking into account the favorable efects on employment of a weak
exchange rate that were mentioned above, a low demand elasticity u and
fast labor force growth L
^
could mean that a strong export performance
translates into weak or even negative wage and productivity growth in
the non-traded sector. A case like this calls for fscal and social policies
intended to foster demand for non-tradables and compensate for the nega-
tive efects on income distribution and employment.
2.3 MACROECONOMIC POLICY REGIMES FOR
A STABLE COMPETITIVE REAL EXCHANGE
RATE
For the reasons just indicated, a competitive and stable RER can make
a substantial contribution to economic growth and employment crea-
tion. Programming the RER, however, is no easy task. It is most directly
impacted by the nominal exchange rate, itself infuenced by many factors,
but also depends on the overall infation rate and shifting relative prices.
Nor can the RER be the only macro policy objective. In any economy
there are bound to be multiple and partially conficting objectives. And all
policies exchange rate, fscal, monetary and commercial/industrial are
interconnected and have to be coherently designed and implemented.
The following discussion focuses on these exchange rate coordination
issues in the context of middle income economies with at least sporadic
access to private international capital markets. Although they are not
addressed in detail here, somewhat similar questions can easily arise in low
income countries receiving of cial capital infows, especially if they jump
to levels of 1020 percent of GDP as suggested in the discussion of the
Millenium Development Goals (MDG).
So how can policy makers target the RER while at the same time con-
trolling infation, reducing fnancial fragility and risk, and aiming toward
full employment of available resources? Our focus necessarily has to shift
from the real economy to encompass monetary and expectational consid-
erations. The principal emphasis is on the degrees of freedom available to
the monetary authorities, if only because they have been at center stage in
recent policy debates.
2.3.1 What Determines the Nominal Exchange Rate?
To set the stage, a few observations about how the nominal exchange rate
fts into the macroeconomic system make sense.
Theories that can reliably predict the level of the rate and its changes
over time when it is not strictly pegged do not exist. (The fact that pegs not
infrequently break down means they do not have 100 percent predictive
power either.) In present circumstances in developing and transition
economies (especially those at middle income levels) it is not unreason-
able to assume that a more or less foating rate is determined in spot and
future asset markets; in efect the spot rate foats against its expected
future values. The quotation marks mean that we view expectations along
Keynesian lines as emerging from diverse opinions on the part of market
participants about how the rate may move. Beauty contests which
magnify small shifts in average market opinion and other sources of seem-
ingly capricious market behavior can easily come into play (Eatwell and
Taylor, 2000).
With regard to the level of the rate, it is useful to think about a simple
bond market equilibrium condition such as
i 5 f(e, e
#
exp
, M) (2.8)
with i as the local interest rate, e the spot exchange rate, e
#
exp
the expected
(as an aggregate of by market perceptions) change in the rate over time
and M an index of monetary relaxation. A high or depreciated value of
e means that national liabilities are cheap as seen from abroad. It should
be associated with high local bond prices or low interest rates. If expected
depreciation e
#
exp
rises, on the other hand, foreign wealth-holders will want
to shift away from local liabilities and i will increase. Open market bond
purchases will increase M and be associated with a reduction in i.
Over the past couple of decades under conditions of external liberaliz-
ation, most developing economies have been af icted by high local interest
rates and appreciated currencies. This unfavorable constellation of macro
prices is consistent with Equation (2.8).
The dynamics of the exchange rate will be infuenced by interest rates,
because it is an asset price. One crucial question is whether lower domes-
tic rates will tend to make the nominal rate depreciate or appreciate. If it
tends to rise (or depreciate) over time, then exchange rate dynamics can be
a powerful mechanism for transmitting the efects of expansionary mon-
etary policy into infation by driving up local production costs.
Standard arbitrage arguments as built into interest rate parity theorems
imply that the expected change in the spot rate e
#
exp
should be an increasing
function of the diference between domestic and foreign rates. If myopic
perfect foresight applies, the expected change will be equal to the observed
change (up to a small error term). Hence a lower local interest rate should
cause appreciation over time. On Wall Street, such an analysis of exchange
rate movements is called an operational view.
A speculative view is that the exchange rate will depreciate when the
local interest rate decreases.
3
This view makes intuitive sense insofar as low
interest rates should make national liabilities less attractive. It was perhaps
frst advanced macroeconomically by Minsky (1983) and can be made
consistent with the parity theorems if it is assumed that there is a relatively
strong positive feedback of expected exchange rate increases into the domes-
tic interest rate via the bond market equilibrium condition Equation (2.6).
Recent macroeconomic history (Frenkel, 2004) suggests that the specu-
lative view is the more accurate description of exchange rate behavior in
middle income economies.
2.3.2 Avoiding Catastrophes
The most fundamental justifcation for avoiding a persistently strong
exchange rate is that it is an invitation to disaster. Exchange appreciation
is always welcome politically because it may be expansionary (at least in
the short run), is anti-infationary and reduces import costs (including
foreign junkets for those who can aford them). However, for the reasons
discussed above, it can have devastating efects on resource allocation
and prospects for development. Moreover, fxed or quasi-fxed strong real
rates can easily provoke destabilizing capital fow cycles as perhaps frst
described analytically by Frenkel (1983) and re-enacted many times since.
The existence and severity of these cycles is in practice a powerful argu-
ment for a stable exchange rate regime built around some sort of managed
foat (details below). A foating rate does appear to moderate destabiliz-
ing capital movements in the short run, and is therefore a useful tool to
deploy. At the same time, the central bank has to prevent the formation of
expectations that there will be RER appreciation, which can easily become
self-fulflling along beauty contest lines. A commitment to a stable rate,
backed up by forceful intervention if necessary, is one way the bank can
orient expectations around a competitive RER.
2.3.3 Trilemmas
Possibilities for central bank intervention are often said to be constrained
by a trilemma among (1) full capital mobility, (2) a controlled exchange
rate and (3) independent monetary policy. Supposedly, only two of these
policy lines can be consistently maintained. If the authorities try to pursue
all three, they will sooner or later be punished by destabilizing capital
fows, as in the run-up to the Great Depression around 1930 and Britain
and Italys dif culties during the European Exchange Rate Mechanism
(ERM) crisis more than 60 years later.
The trilemma as just stated is a textbook theorem which is, in fact,
invalid. Even with free capital mobility, a central bank can undertake
transactions in both foreign and domestic bonds (not to mention other
monetary control maneuvers) to regulate the money supply, regardless of
whatever forces determine the exchange rate (Taylor, 2004).
Nevertheless, something like a trilemma can exist in the eye of a
beholder. There are practical limits to the volume of interventions that a
central bank can practice, along with complicated feedbacks. Possibilities
for sterilizing capital infows or outfows are bounded by available asset
holdings. Volumes of fows depend on exchange rate expectations which in
turn can be infuenced by central bank behavior and signaling.
So how does the market decide when a perceived trilemma is ripe to be
pricked? The fact that no single form of transaction or arbitrage opera-
tion determines the exchange rate means that monetary authorities have
some leeway in setting both the scaling factor between their countrys
price system and the rest of the worlds and the rules by which it changes.
However, their sailing room is not unlimited. A fxed rate is always in
danger of violating what average market opinion regards as a fundamental.
Even a foating rate amply supported by forward markets can be an invita-
tion to extreme volatility. Volatility can lead to disaster if asset preferences
shift markedly away from the home countrys liabilities in response to
shifting perceptions about fundamentals or adverse news. Unregulated
international capital markets are at the root of any perceived trilemma.
It is a practical problem that must be evaluated in each case, taking into
account the context and circumstances of policy implementation.
2.3.4 Monetary and Exchange Rate Policies and Capital Flows
The implication is that if it wishes to target the RER, the central bank
has to maintain tolerable control over the macroeconomic impacts of
cross-border fnancial fows in a world with relatively open foreign capital
markets. For the sake of clarity, it makes sense to analyse situations of
excess supply and excess demand for foreign capital separately.
Large capital infows can easily imperil macro stability. Indeed, central
bank attempts to sterilize them by selling domestic liabilities from its
portfolio may even bid up local interest rates and draw more hot money.
Preservation of monetary independence in this case may well require capital
market regulation. Measures are available for this task.
4
They do not work
perfectly, but can certainly moderate infows during a boom. Booms never
last forever; the point is that the authorities can use capital market inter-
ventions to slow one down to avoid an otherwise inevitable crash.
If there are capital outfows too large to manage with normal exchange
rate and monetary policies, the authorities certainly do not want to engage
in recession-triggering monetary contraction. If the exchange rate has been
maintained at a relatively weak level, the external defcit is not setting of
fnancial alarm bells, and infation is under control, then there are no fun-
damental reasons for market participants to expect a maxi-devaluation.
Under such circumstances the way for the authorities to maintain a policy
regime consistent with a targeted RER is to impose exchange controls and
restrictions of capital outfows.
Contrary to IMF-style opinion that all runs against a currency must
be triggered by poor fundamentals (even if they momentarily escape the
notice of the authorities and IMF of cials), it is perfectly clear that they
can arise for reasons extraneous to economic policy think of a politi-
cal crisis, the fallout from mismanagement of an important bank or the
impacts of fnancial contagion from a regional neighbor. In all such cases
outfow controls can be used to maintain an existing policy package in
place. They may not have to be utilized for very long.
5
2.3.5 Monetary Policy
In a developmental policy regime, monetary policy must be designed in
view of its likely efects on the RER, infation control and the level of eco-
nomic activity. There is nothing very surprising here in practice central
banks always have multiple objectives. In the USA, despite lip service to
controlling price infation, the Federal Reserve certainly responds to the
level of economic activity and fnancial turmoil (witness the 1990s stock
market bubble and the long-term capital management (LTCM near-crisis).
In many developing countries central banks intervene more or less system-
atically in the exchange markets. The proposal here is that these interven-
tions should help support a developmentally oriented RER for the reasons
presented above. That is, the nominal rate should move to hold the RER in
the vicinity of a stable competitive level for an extended period of time.
Infation targeting, on the other hand, is the current orthodox buzzword.
The nominal exchange rate and other policies should be programmed
to ensure a low, stable rate of infation. A trilemma-like argument is
involved. If exchange market interventions target the RER as opposed to
the nominal exchange rate and the central bank cannot manage the money
supply, there is no nominal anchor on infationary expectations. The infa-
tion rate cannot be controlled.
As we have seen, in practical terms the trilemma can be circumvented,
allowing the monetary authorities to bring developmental objectives into
their remit. But they have to take at least fve important considerations
into account in monetary management.
First, many developing countries now have low to moderate infation
rates, demoting infation control in the hierarchy of policy objectives.
Second, will low interest rates tend to set of infationary nominal depre-
ciation (under speculative exchange rate dynamics as discussed above)?
RER targeting can help the central bank steer away from this problem.
Third, shifts in aggregate demand likely to result from changes in the
exchange rate and monetary policy must be taken into account, and
appropriate ofsetting policies deployed.
Fourth, also as mentioned above, some mix of temporary capital infow
or outfow controls may be needed to allow the central bank to regulate
monetary aggregates and interest rates rather than be overwhelmed by
attempts at sterilization.
Finally, unstable money demand and other unpredictable factors mean
that the monetary authorities have to be alert and fexible. Indeed, infa-
tion targeting is a codeword for orthodox recognition that quantitative
monetary and even interest rate targets are impractical. It is a means for
giving more discretion in trying to attain a single target.
The point being made here is that discretion can and should serve other
ends. A stable competitive RER in coordination with sensible industrial
and commercial policies can substantially improve prospects for economic
development. Surely that should be the overriding goal of the mon-
etary and all other economic authorities in any developing or transition
economy.
NOTES
1. Just to be clear, we will treat the RER as the ratio of tradable to non-tradable price
indexes. Real devaluation or weakening the RER means that r increases.
2. Again, these arguments are old. Ocampo and Taylor (1998) provide a recent summary.
3. To be more precise, the change over time in the spot rate e
#
5 de/dt will turn negative
when i decreases if the operational view applies and positive when the speculative view is
true.
4. For an ample menu, see papers by Deepak Nayyer, Eric Helleiner and Gabriel Palma
in Eatwell and Taylor (2002) and Epstein et al. (2003). Salih Neftci and Randall Dodd
assess the possibilities of using fnancial engineering to circumvent controls.
5. Argentina, for example, successfully managed exchange controls and capital outfow
restrictions in mid 2002. The measures were transitory. They were gradually softened as
buying pressure in the exchange market diminished.
REFERENCES
Eatwell, J. and L. Taylor (2000), Global Finance at Risk: The Case for International
Regulation, New York: The New Press
Eatwell, J. and L. Taylor (eds) (2002), International Capital Markets: Systems in
Transition, New York: Oxford University Press.
Epstein, Gerald, Ilene Grabel and K.S. Jomo (2003), Capital management tech-
niques in developing countries, in Ariel Buira (ed.), Challenges to the World
Bank and IMF; Developing Country Perspectives, London: Anthem Press,
reprinted in Gerald Epstein (ed.) (2005), Capital Flight and Capital Controls in
Developing Countries, Cheltenham, UK and Northampton, MA, USA: Edward
Elgar.
Frenkel, R. (1983), Mercado Financiero, expectativas cambiales, y movimientos
de capital, El Trimestre Economico, 50, 204176.
Frenkel, R. (2004), Real exchange rate and employment in Argentina, Brazil,
Chile, and Mexico, paper prepared for the Group of 24, Washington, DC:
September.
Kaldor, N. (1978), Causes of the slow rate of growth of the United Kingdom, in
Further Essays on Economic Theory, London: Duckworth.
Lerner, A.P. (1936), The symmetry between import and export taxes, Economica,
3, 30613.
Minsky, H.P. (1983), Monetary policies and the international fnancial environ-
ment, Washington University Department of Economics mimeo, St Louis,
MO.
Ocampo, J.A. and L. Taylor (1998), Trade liberalization in developing econ omies:
modest benefts but problems with productivity growth, macro prices, and
income distribution, Economic Journal, 108, 152346.
Okun, A.M. (1962), Potential GNP: its measurement and signifcance, reprinted
in Joseph Pechman (ed.) (1983), Economics for Policy-Making, Cambridge, MA:
MIT Press.
Rada, C. (2005), A growth model for a two-sector open economy with endog-
enous employment in the subsistence sector, Schwartz Center for Economic
Policy Analysis, New School University, New York.
Rada, C. and L. Taylor (2004), Empty sources of growth accounting, and empiri-
cal replacements la Kaldor with some beef, Schwartz Center for Economic
Policy Analysis, New School University, New York.
Taylor, L. (2004), Exchange rate indeterminacy in portfolio balance, Mundell-
Fleming, and uncovered interest rate parity models, Cambridge Journal of
Economics, 28, 20527.
Verdoorn, P.J. (1949), Fattori che Regolano lo Sviluppo della Produttivita del
Lavoro, LIndustria, 1, 310.
Woo, W.T. (2005), Some fundamental inadequacies in the Washington Consensus:
misunderstanding the poor by the brightest, in Jan Joost Teunissen (ed.),
Stability, Growth, and the Search for a New Development Agenda: Reconsidering
the Washington Consensus, The Hague: Forum on Debt and Development,
(FONDAD).
44
3. Infation targeting and the real
exchange rate in a small economy:
a structuralist approach
Jose Antonio Cordero
3.1 INTRODUCTION
In the past few years several nations, especially encouraged by the
International Monetary Fund (IMF), have decided to move into a mon-
etary regime based on explicit infation targets. This regime has been
strongly recommended to economies that were struggling to bring their
infation rates down to the level of the more advanced countries. The
application of this regime has been very efective in reducing infation but,
as argued in Epstein and Yeldan (2008), the fnal efects on employment
and economic growth are not so clear. A rigorous analysis of the conse-
quences of infation targeting in open economies requires a formal model
exploring the interaction among money, the foreign sector, output and
employment.
Within the structuralist tradition several models have been developed
in order to examine growth and distribution in open economies. These
works, however, do not always include the monetary sector and, when
they do, they rarely establish the links between the latter and the balance
of payments; they are not designed to examine the association between
the monetary regime and the foreign exchange rate system. This chapter
attempts to fll that void by means of a framework that allows comparing
two alternative monetary regimes: one resulting from infation targeting,
and another resulting from real exchange rate targeting.
The model allows formalizing the adjustment process in three periods:
frst, output and employment vary with given prices, second, infation
clears an explicit monetary sector and, fnally, capital accumulation,
income distribution and the real exchange rate reach a stable steady state.
The growth of labor productivity, through learning efects, leads to endog-
enous growth.
Infation is explained here as a monetary phenomenon, but efective
Infation targeting and the real exchange rate in a small economy 45
demand remains a critical determinant of output, employment and eco-
nomic growth. The chapter thus also contributes to a better understanding
of the interaction between the real and the monetary sector.
In Section 3.2 the central bank adjusts the nominal exchange rate to hit
a real exchange rate target, with infation control playing an important
but secondary role. In this case the growth and employment performance
are favorable but the economy sufers the dif culties arising from the tri-
lemma. It is shown that these complications may be mitigated by reducing
the emphasis on open market operations, and utilizing the possibilities
provided by the legal reserve requirement or capital controls.
The model in Section 3.3 presents a monetary system in which the
nominal interest rate is chosen to hit an infation target, and the nominal
exchange rate foats freely to equilibrate the balance of payments. Under
this infation targeting scheme, price stability becomes the only goal of
monetary policy. We show that this arrangement is very efective in bring-
ing infation down, but overdetermination hurts both the employment and
growth potential of the economy.
For each of the models developed in Sections 3.2 and 3.3, the dynamic
properties of the long-run equilibrium are also analysed. Section 3.4
presents some concluding remarks.
3.2 A SMALL OPEN ECONOMY WITH A REAL
EXCHANGE RATE TARGET
Here a real exchange rate target is set by the central bank at a level
that is just the one required to maintain the competitiveness of exports.
Maintaining exports will help employment so that, in a way, a real
exchange rate target is analogous to an employment target.
The economy operates in the following way. The foreign exchange rate
system is better defned as one in which the nominal rate is adjusted (crawl-
ing peg) by the central bank to bring the actual real rate to the specifed
target. This framework analyses the dif culties the central bank faces when
attempting to control both the money supply and the foreign exchange rate
under an open capital account. This is the well known problem of the tri-
lemma of monetary policy: with a rigid nominal exchange rate, attempts
to reduce the money supply (and therefore infation) by raising the interest
rate cause an increase in international monetary reserves which, in turn,
ofsets the reduction in money supply (and thus infation) that authorities
were originally looking for.
This situation leads to infation rates that are higher than internat-
ionally accepted levels. But it also allows the central bank to make the
adjustments that are needed to maintain the competitiveness of exports
and the rate of economic growth.
3.2.1 Assumptions of the Model
Some assumptions below are rather general, but others reveal the structur-
alist approach that is followed in building the model.
1
1. The country produces a consumption good Q, which could be
exported or consumed domestically.
2. The country imports a composite good that may be used for consump-
tion or investment purposes. All capital used in the country is imported.
3. The small economy determines domestically the price of local produc-
tion P
Q
, but it cannot afect the international price of imports P
M
,
assumed exogenous. Local prices P
Q
, are assumed fxed in the short
run, but may change in the medium run. This assumption, in relation
to the price of local production, allows analysing the efect that infa-
tion may have on the competitiveness of exports.
4. The price of imports, in terms of foreign currency is P
M
, assumed exog-
enous as seen above. In terms of local currency, the price of imports is
eP
M
, where e represents the nominal exchange rate (number of units of
local currency that have to be paid for one unit of foreign currency).
5. There are two social classes: producers and workers. Workers spend
all their wage income on consumption of both the local good and the
imported good, while producers save a fxed portion s of their proft
income and the rest they spend on consumption of both the domestic
and imported goods.
6. Good Q is produced by means of a fxed-coef cient production func-
tion: Q 5 Min[
L
a
, uK] where L represents the labor input, a is a
technical coef cient, K is capital and u represents the output/capital
ratio. Capital does not depreciate, and frms are assumed to hold
excess capacity so that u is a variable that moves up (down) when
capacity utilization increases (decreases).
7. The interest rate is a policy variable fxed by the central bank for
stabilization purposes.
8. The capital stock K and the labor force N (and of course their ratio,
denoted here as k) are given in the short and medium run, but they
may change in the long run.
9. The nominal exchange rate e is a policy variable that remains fxed
in the short and medium run, but is adjusted in the long run by the
central bank. The adjustment process follows a rule that allows hitting
a real exchange target.
10. The real exchange rate, which determines the competitiveness of
exports, is defned as h 5
eP
M
P
Q .
11. The wage share is assumed given in the short and medium run, and is
defned as A 5
Wa
P
Q . In the short run W and the technical coef cient a
are also fxed, but they adjust in the long run.
12. The analysis is conducted in three stages: in the short run output
adjusts while prices, the real exchange rate and the distribution
of income remain given; in the medium run infation will move to
clear the monetary market; and in the long run the wage share, the
real exchange rate and the capital/labor force ratio are allowed to
vary.
3.2.2 General Overview
In this section we present the goods market, and defne the conditions for
macroeconomic equilibrium in the short run (equality between the invest-
ment and savings rates). With the previous assumptions we may fnd equi-
librium values for the rate of growth of the capital stock, the rate of proft,
the output/capital ratio and the employment rate; infation only adjusts in
the medium run, as seen below.
Within the foreign sector, in the short run, the given levels of the
nominal interest rate and real exchange rate help us determine the increase
in international monetary reserves; this then enters into the determination
of money supply growth. At this point we have already determined the
short run variables, and the monetary sector may then clear in the medium
run through adjustments in the infation rate.
Once the whole model is solved we examine the impact that open market
operations have on infation and economic growth. It is at this point that
the trilemma arises and the real exchange rate emerges as a leading char-
acter in the story. The relevance of this variable stems also from the fact
that it is, along with the interest rate, the only one capable of stimulating
the level of economic activity.
2
3.2.3 Equations of the Model
Goods market
g
s
5 sr 1 B (3.1)
g
d
5 b
0
1 b
1
r 2 b
2
i (3.2)
r 5
u
h
(1 2 A) (3.3)
P
Q
5 P
Q
(3.4)
g
d
5 g
s
(3.5)

L
N
5
L
Q
Q
K
K
N
5 l 5 auk (3.6)
Monetary market
M
^
d
5 h
1
p 1 h
2
u 2 h
3
i 2 h
4
p
E
(3.7)
M
^
S
5 R 1 t
0
2 t
1
i (3.8)
M
^
S
5 M
^
d
(3.9)
i 5 i
0
(3.10)
Foreign sector
B 5 F 2 R (3.11)
B 5 B(h); B
h
, 0 (3.12)
F 5 F(i); F
i
. 0 (3.13)
e^ 5 W[h
T
2 h] (3.14)
In following the structuralist tradition, several variables are measured
as ratios over the value of the capital stock eP
M
K: desired investment
( g
d
) , total saving (g
s
), the current account defcit (B), capital infows (F),
the increase in international monetary reserves (R) and output (u). The
nominal interest rate (i), the actual infation rate (p), the expected infation
rate (p
E
), and the rates of growth of money supply and money demand
(M
^
s
, M
^
d
) are written in percentage terms. The real exchange rate, the wage
share and the capital/labor force ratio are denoted by h, A and k respec-
tively, and the symbols h
1
, h
2
, h
3
, h
4
, t
0
, t
1
, b
0
, b
1
, b
2
all represent positive
parameters.
Equation (3.1) defnes total saving as domestic saving sr, plus foreign
saving B. In Equation (3.2) desired investment depends on the proft and
interest rates. Then Equation (3.3) shows the rate of proft in terms of the
fexible output/capital ratio, the real exchange rate and the wage share,
and Equation (3.4) indicates that local prices remain fxed in the short run.
Goods market equilibrium is defned in Equation (3.5), while Equation
(3.6) shows the employment rate (l) as a function of the output/capital
ratio (for given levels of K/N, denoted here as k).
In the monetary market equilibrium is attained when the rates of growth
of money supply and money demand (in nominal terms) are equal. In
Equation (3.7) money demand is determined by infation, the output/
capital ratio, the nominal interest rate and an exogenous expected infa-
tion term. Agents face a dual decision regarding price increases: on the one
hand, infation raises their demand for money as they attempt to maintain
the value of their real balances. But, on the other hand, agents will also try
to get rid of (that is, spend) their holdings of money if they expect price
increases, the public will attempt to win the race against infation.
The money supply in Equation (3.8) grows with the accumulation of
foreign exchange reserves, but decreases with the exogenous interest rate.
This latter component attempts to formalize central bank policy making:
in order to reduce money growth, the bank conducts open market opera-
tions and drives interest rates up, and the opposite would happen if policy
makers wished to increase money growth. This is more elaborate than the
Taylor rule that appears in the new consensus macroeconomics (Arestis
and Sawyer, 2003a; Romer, 2000): we argue here that changes in the
interest rate will afect the money supply and this may or may not afect
infation; while the Taylor rule (Taylor, 1993) is set up as a mechanical
relationship between infation and the central banks reaction. Finally, the
constant parameter t
0
captures the efect of monetary policy that is not
related to open market operations (changes in the legal reserve require-
ment or the discount rate, for example).
Equation (3.9) presents money market equilibrium, and Equation (3.10)
shows the exogenous nominal interest rate. Infation is here a monetary
phenomenon: excess money supply growth leads to excess spending, and
hence to higher infation. But also the lack of money supply results from
high interest rates, which hurts employment and output growth. Thus,
efective demand continues as a relevant determinant of economic activity.
Here money is not neutral, so, unlike several versions of the new consen-
sus macroeconomics (Meyer, 2001), money does matter. Our approach
also difers from the post-Keynesian view of infation as resulting from
cost-push and social confict (as presented, for example, in Arestis and
Sawyer (2006)). This model creates a bridge between monetarist notions
of the money market, and Keynesian views on the importance of efective
demand.
Finally, we need to keep in mind that, ceteris paribus, the use of the
interest rate to control infation, will only lead to a stable adjustment
process if money supply is more sensitive to changes in the interest rate
than money demand (that is, t
1
is bigger than h
3
). Only in this case will the
higher interest rate lead to a reduction in the excess money supply, and to
a lower infation rate.
3
The foreign sector is described in Equations (3.11) through (3.14). In
Equation (3.11) the current account defcit is fnanced by capital infows
and depletion of international monetary reserves. The current account
defcit depends negatively on the real exchange rate and net capital infows
depend positively on the interest rate. In Equation (3.14) we see how, in
the long run, the monetary authority devalues the nominal exchange rate
to bring the real rate to its target (h
T
); is a positive parameter.
3.2.4 Formal Solution of the Model
The process starts when the central bank defnes the nominal interest
rate and the target for the real exchange rate. These two policy variables
are utilized, respectively, to control the money supply, and to maintain
the competitiveness of net exports. The current account defcit and net
capital infows can then be determined, and with this we can solve for R in
Equation (3.11).
In the goods sector Equations (3.1), (3.2), (3.12) and (3.5) determine the
proft rate. In Equation (3.3), given A and h, we fnd the output/capital
ratio, and Equation (3.6) then leads to the employment rate (l). Again in
Keynesian fashion, employment is determined in the goods market (not in
the labor market as monetarist and new classical approaches would suggest).
The short-run solution for the goods sector is shown in Figure 3.1.
We may now move into the medium run and use Equations (3.7), (3.8),
(3.9) and (3.10) in the monetary sector to solve for infation, as shown
below:
p
0
5
R
0
1 t
0
1 (2 t
1
1 h
3
) i
0
2 h
2
u
0
1 h
4
p
E
h
1
(3.15)
where u
0
represents the level of the output/capital ratio found in the goods
market, and i
0
denotes the interest rate fxed by the central bank. The rela-
tionship between the interest rate and infation is shown in Figure 3.2.
The money market equilibrium schedule (MM line) in Figure 3.2 has a
negative slope when the conditions specifed in Note 3 hold. This schedule
provides the combinations of infation and interest rate that are consistent
with equilibrium in the monetary market. Here p
0
is the infation rate that
clears the money market when the central bank fxes the interest rate at i
0
.
In order to bring infation down, the central bank raises the interest rate to
reduce money supply growth; as the availability of money declines, aggre-
gate spending decreases and infation goes down (the economy moves to
the left along the MM line).
g
g
0
g
d
g
s
l
l
u
l = auk
u
r
r
u
0
u =
rh
1A
Figure 3.1 Goods market equilibrium in the short run
0
i
0
i
MM
Figure 3.2 Money market equilibrium schedule
But, in the foreign sector, a higher i will attract more capital fows,
making international monetary reserves go up. The money supply will
then increase (the MM line shifts out), compensating (at least partially) the
initial attempt to reduce infation pressure. Thus the monetary authorities
cannot control both the exchange rate and interest rate (and the money
supply) under an open capital account.
Second, in the goods market, as i goes up, investment demand goes
down; the g
d
line in Figure 3.1 shifts down and the rate of growth, the
rate of proft, the output/capital ratio and the employment rate decrease.
Stabilization is not too successful and hurts economic activity.
Fortunately, the central bank can decide to increase the target level
for h, which will cause foreign savings (the current account defcit) to
go down. The g
s
line in Figure 3.1 will shift to the right (down) leading
to faster growth, higher proft rate, and higher output/capital ratio and
employment rate.
We have thus confrmed the double edged character of the real exchange
rate: it is to blame for the dif culties the central bank faces in controlling
infation; had it had more fexibility in the short run, there would had been
no accumulation of international reserves and we would not have lost
ground in the fght against infation. But, on the other hand, it is the ability
of the central bank to move the real exchange rate to a targeted level,
which provides the economy with the possibility to reach faster growth
and higher employment rates. Arguments and evidence in favor of target-
ing the real exchange rate may be found in Frenkel (2004) and in Frenkel
and Ros (2006).
Now a fnal comment on monetary policy: there is no reason for the
central bank to only use open market operations to control the money
supply. Other policy instruments, like the legal reserve requirement,
could play an important control role without a direct impact on interest
rates.
4
A higher reserve requirement ratio may be seen in our model as a
reduction in the term t
0
in Equation (3.15). In Figure 3.2 this will show
as a downward shift of the MM schedule, indicating that there will now
be a lower level of infation at every level of the interest rate. The central
bank may thus be able to curve down price growth, but without changes
in international monetary reserves, breaking in this way the trilemma.
The alternative presented here is also in line with the arguments in Frenkel
(2004), who goes even further to point out that capital controls are feas-
ible; a similar argument on the convenience of capital controls is provided
by Ffrench Davis (2003).
Many central banks, however, would not like to use the legal reserve
requirement, or capital controls as instruments for monetary policy:
they prefer open market operations because they are, well, more market
oriented.
5
Policy makers are thus opting for more fexible exchange rate
arrangements, combined with infation targeting regimes, as will be
explained in Section 3.3.
3.2.5 Long-run Dynamics
We follow Cordero (1995) in analysing the interaction between the wage
share (A) and the real exchange rate (h), which we assumed given in the
short and medium run. But we have an additional state variable here: the
capital/labor force ratio k, so this is really a 3 3 system. We assume that,
in the long run, the variables which adjusted in the short run to clear the
goods market (u, g, r, l ), and in the medium run to clear the monetary
sector (p) remain at their equilibrium levels.
The motion of A is given by
A
^
5 W
^
1 a^ 2 p (3.16)
where the hats denote rate of growth. The coef cient a was defned as
L
Q
so we can write that y 5
Q
L
5
1
a
where y represents average labor
productivity.
Following Arrow (1962) it is argued that labor productivity depends
on learning, which is assumed to be positively related to production. We
also follow Dutt (1994) in assuming that production is associated to the
investment rate g:
y^ 5 2a^ 5 Y( g 2 g), Y
g
. 0 (3.17)
with Y
g
the partial derivative of productivity growth with respect to the
investment rate, and g is an arbitrary constant number.
The specifcation of wage growth assumes that workers try to increase
their nominal wage whenever the actual wage share falls below an exog-
enous level A
W
desired by workers as indicated in Equation (3.18):
W
^
5 q[A
W
2 A] (3.18)
Finally, the infation rate that clears the monetary market depends on
the real exchange rate and the wage share:
p
0
5 p(h, A); p
h
, 0; p
A
, 0 (3.19)
where p
h
, p
A
denote partial derivative of p with respect to h and A, respec-
tively. The dependency of p
0
on h and A comes from the presence of the
equilibrium output/capital ratio (u
0
) in Equation (3.15); and from the fact
that this u
0
was determined in the goods market (through the interaction
between g
d
and g
s
for given h and A).
We may now put Equations (3.17) through (3.19) in (3.16) to get a dif-
ferential equation in h and A:
A
^
5 q[A
W
2 A] 2 Y[g(h) 2 g] 2 p[h, A] (3.20)
The dynamic behavior of the real exchange rate is described by:
h
^
5 e^ 2 p (3.21)
According to Equation (3.14), reproduced below, the nominal exchange
rate moves up when the actual level of the real exchange rate falls below
the target chosen by the central bank:
e^ 5 W[h
T
2 h] (3.14)
Next use Equations (3.19) and (3.14) in Equation (3.21) to get:
h
^
5 W[h
T
2 h] 2 p[h, A] (3.22)
We should recall now that we had a third state variable (k), but the system
with Equations (3.20) and (3.22) may be solved independently of k, so we
may proceed as if we had a 2 2 system with A and h as the state vari-
ables.
6
The stability of this system (Equations 3.20 and 3.22) is analysed by
means of the Jacobian matrix:
Det(J) 5
0A
^
0A
0A
^
0h
0h
^
0A
0h
^
0h
5
(2 q 2 p
A
) (2Y
g
g
h
2 p
h
)
2p
A
(2 W 2 p
A
)
The determinant of the Jacobian is:

Det(J) 5 (q 1 p
A
) (W 1 p
h
) 2 p
A
(Y
g
g
h
2 p
h
)
5 q[W 1 p
h
] 1 p
A
[W 2 Y
g
g
h
]
which will be positive if W . p
h
and W . Y
g
g
h
. In other words, Det (J) will
be positive if W is large.
The trace is given by Tr(J) 5 2q 2 p
A
2 W 2 p
h
, which will be
negative if q, and are large. So the long-run equilibrium will be stable if
wages are fexible and the central bank is very committed to maintaining
the real exchange rate close to the chosen target. Finally, as the short-
run equilibrium level of growth depends on h, and h is determined in the
system formed by Equations (3.20) and (3.22), we may say that economic
growth is endogenous in this model.
3.3 THE SMALL ECONOMY UNDER AN
INFLATION TARGETING REGIME
3.3.1 The Infation Targeting Regime
In the previous section we saw the problems that may arise if, with a
fxed real exchange rate, the central bank attempts to control infation
by using open market operations. Although, as we saw, there are several
ways in which the trilemma could be faced successfully, there is a strong
tendency to move to an infation targeting regime and turn the exchange
rate system into a fexible one. This bias, according to Epstein (2005),
is part of a global change in the practice of central banking, promoted
mostly by institutions like the IMF. In this view central banking practices
must be based on independence, infation as the goal of policy (including
application of infation targeting regimes), and the use of indirect methods
of monetary policy (mostly open market operations). Arestis and Sawyer
(2003b) relate this bias to the new consensus macroeconomics in which
infation targets become the focus of monetary policy. In this section we
want to emphasize that, once this approach is adopted, everything in the
economy will be tied to the infation target; the result is great success in
bringing infation down, but a rather disappointing performance in terms
of growth and employment.
According to Galindo and Ros (2008) this monetary regime leads to
possible appreciation of the real exchange rate, and increased vulner-
ability to monetary shocks coming from the external sector. In the case of
Mexico, these authors show that infation targeting has led to an appreci-
ation of the real exchange rate, which has had a negative impact on output
growth.
Epstein (2002) reports that the utilization of this monetary system in
South Africa led to high interest rates, accompanied by low employment
and investment rates. His recommendation is that South Africa move to a
scheme in which employment is targeted, but subject to an infation goal.
He also recommends the use of capital controls in order to better control
the efect of the world economy on South Africa. Setterfeld (2006) also
argues that governments should pursue output and employment targets,
in addition to infation goals.
Other more general cross-country studies have found that there is no
reason to keep infation below the 3 to 5 percent range, especially because
in middle income economies higher single digit rates of infation could
stimulate economic growth (Pollin and Zhu, 2005). Another study con-
ducted by Ball and Sheridan (2003) concludes that infation targeting
has generated no major benefts in economic performance (other than a
decline in infation rates). In this section we develop a model to examine
the efects of infation targeting on growth and employment.
3.3.2 Revised Equations
Money market
i 5 G(p); G
p
, 0; p 5 p(i), p
i
, 0 (3.23)
M
^
s
5 t
0
2 t
1
i (3.24)
M
^
d
5 h
1
p 1 h
2
u 2 h
3
i (3.25)
p 5 p
T
(3.26)
M
^
s
5 M
^
d
(3.27)
Foreign sector
B 5 F (3.28)
B 5 B(h); B
h
, 0 (3.29)
F 5 F(i); F
i
. 0 (3.30)
Goods sector
g
d
5 g
d
(r, i) , g
d
r
. 0, g
d
i
, 0 (3.31)
G 5 g
d
(r, i) 2 B(h) (3.32)
S 5 sr (3.33)
G 5 S (3.34)
P
Q
5 P
Q
(3.4)
r 5
u
h
(1 2 A) (3.3)
l 5 auk (3.6)
In Equation (3.23) the interest rate is the instrument the central bank
uses to control infation. Once a target is defned in Equation (3.26),
open market operations allow bringing the interest rate to the level that
is exactly needed to hit the chosen infation target. As for money demand
(Equation 3.25), we use a formulation that is similar to the one in Section
3.2. Finally, and in accordance to this monetary system, we assume that
the central bank announces an infation target in the understanding that
this goal is prior to any other objective the bank might have. The priority
assigned to hitting this level will be clear as the model is explained in more
detail.
The monetary sector operates in the following manner. First, the
central bank recognizes that the interest rate and the infation target are
not chosen independently of one another: the bank knows that there is
an inverse relationship between nominal interest rates and money supply.
Thus, in order to fght infation, the bank conducts open market opera-
tions (to reduce the money supply) and pushes interest rates up. In other
words, higher interest rates, are associated to lower infation rates, and this
is exactly what Equation (3.23) tells us.
Again this is clearly diferent from the Taylor rule that appears in
the new synthesis models that we referred to in previous sections of this
chapter. Our Equation (3.23) is a behavioral relationship indicating how
infation reacts to changes in the nominal interest rate. The so-called
Taylor rule is a more mechanical expression indicating how the central
bank moves the interest rate when infation goes up. In the new synthesis
models this rule replaces the monetary sector. In this chapter, however,
we argue that economic authorities in less developed countries still believe
that their policy decisions should be based on their perception of the mon-
etary sector; hence there is no Taylor rule in this model, but there is a fully
structured monetary market.
Next we turn to the foreign sector where we start by getting rid of the
accumulation of international monetary reserves. As Equation (3.28)
shows, the current account defcit has to be fnanced by net capital infows.
This is made possible by allowing the real exchange rate to fuctuate as
required to satisfy Equation (3.28).
7
In the goods sector the specifcation looks a bit diferent from Section
3.2, but it still holds the same meaning: Equation (3.32) defnes demand
injections, while (3.33) defnes domestic savings. Equations (3.3), (3.4) and
(3.6) are borrowed from the previous section to describe prices, the rate of
proft and the employment rate.
3.3.3 Solution of the Model
Price rigidities are an important feature of this model: in the short run
prices remain fxed by Equation (3.4), and will adjust only in the medium
run (the growth of prices in the medium run is determined by the infation
target chosen by the central bank). In the short run the burden of adjust-
ment falls on the output/capital ratio and the real exchange rate.
Once the infation target p
T
is announced, Equations (3.26) and (3.23)
allow defning the nominal interest rate. In the foreign sector the real
exchange rate adjusts to equilibrate the balance of payments and we have:
h 5 h[i ], h
i
, 0 (3.35)
Next with i and h known we move to the monetary sector: Equations (3.24),
(3.25) and (3.27) provide an expression for money market equilibrium:
t
0
2 t
1
i 2 h
1
p(i) 2 h
2
u 1 h
3
i 5 0 (3.36)
This equation describes a relationship between the interest rate and the
output/capital ratio, and the slope is given by

0u
0i
5
[2 t
1
2 h
1
p
i
1 h
3
]
h
2
, 0
As before, the sign of this slope results from the conditions in Note 3
(Figure 3.3 presents the corresponding graph) and we write that:
u 5 u[i ], u
i
, 0 (3.37)
Once we know the infation target we can plug the known interest rate i
0
in
Equation (3.36) to fnd the output/capital ratio (u) that clears the money
market. This implies that, in order to derive u, it is not necessary that
savings and investment be equal: the model is overdetermined.
In the next step we go to the goods sector, and examine the behavior of
the rate of proft. From Equation (3.3), and also using Equations (3.35)
and (3.37) we may write the following expression:
r 5
[u(i) ]
[h(i) ]
(1 2 A) (3.38)
and the response of the rate of proft to changes in the interest rate is given
by:

0r
0i
5 r
i
5
(1 2 A)
[h(i) ]
2
[u
i
h 2 uh
i
] (3.39)
We do not have information to defne the sign of the term [u
i
h 2 uh
i
], so
we can just examine two possible cases:
r 5 r[i, (1 2 A) ], r
i
, 0, r
A
, 0 (3.40a)
or
r 5 r[i, (1 2 A) ], r
i
. 0, r
A
, 0 (3.40b)
Signing the derivatives of r with respect to i is very important as that will
allow to also sign the derivatives of demand injections (G) and domestic
saving (S) with respect to i.
Case 1
r
i
, 0, r
A
, 0
i
u
MM
Figure 3.3 Money market equilibrium schedule under infation targeting
From Equation (3.32) and from Equation (3.40a) we get:
G 5 g
d
{r[i, (1 2 A) ], i} 2 B[h(i) ] (3.41)
and the slope of this line is

0G
0i
5 g
d
r
r
i
1 g
d
i
2 B
h
h
i
, 0 (3.42)
Next from Equations (3.33) and (3.40a) obtain:
S 5 sr[i, (1 2 A) ] (3.43)
and the slope of the savings function is

0S
0i
5 sr
i
, 0 (3.44)
Just as in other Keynesian models, the slope of the S line has to be steeper
than that of the G line in order to secure stability of equilibrium in the
short run:
P
0G
0i
P , P
0S
0i
P (3.45)
The graphical solution appears in Figure 3.4. In the frst panel choos-
ing an infation target fxes the interest rate by means of Equation (3.23).
The foreign sector appears in the second panel, where the known interest
rate leads to the real exchange rate that clears the balance of payments.
The monetary sector in the third panel shows that, given the interest rate,
the MM schedule helps us determine the output/capital ratio. Then we
examine the behavior of demand injections (G) and domestic saving (S),
and overdetermination becomes clear. If the interest rate is below the one
that brings G and S to equality, then the economy will be at a point on the
G line. Any reduction in the infation target (and thus increase in the inter-
est rate) will lead to real appreciation, higher unemployment and a lower
rate of economic growth. A reduction of the infation target will reduce
the level of activity even if the interest rate is to the right of the one that
makes G and S equal. As the nominal exchange rate is no longer a policy
variable, the central bank cannot stimulate economic activity and reduce
the infation rate at the same time.
Case 2
r
i
. 0, r
A
, 0
Here the G function keeps the negative slope, but the S function will def-
nitely have a positive slope in the S versus i plane, as shown in Figure 3.5.
With the interest rate at the given level i
0
(which is, again, lower than the
one that brings G and S to equality), the economy must be at a point on the
l
u
l = uak
i
i
0
T
= p (i)
i
B (h) = F (i)
u
i
MM
G, S
i
i
0
G
S
Figure 3.4 An open economy under infation targeting and
overdetermination in the goods sector
S schedule (investors cannot satisfy their desires). A reduction in the infa-
tion target will require a rightward movement of the interest rate, with the
economy moving along the domestic saving line towards the right and to a
higher rate of growth. But notice that investment desires will sufer a down-
ward movement along the G schedule. So what we are observing here is that,
as infation is brought down, the investment drive also declines. Thus in this
case we see that within the infation targeting regime, overdetermination
causes a gap between actual and desired investment. When the given inter-
est rate lies to the right of the level that brings domestic savings and demand
injections to equality, the economy is positioned at a point on the G sched-
ule; in this case stabilization policies lead to a lower rate of growth.
3.3.4 Long-run Dynamics
In the long run we analyse two state variables: A and k, where k repre-
sents the capital (K) to labor force (N) ratio. The analysis here is based on
Cordero (2002).
The motion of A is described by Equation (3.16) reproduced below:
A
^
5 W
^
1 a^ 2 p (3.16)
G, S
i
i
0
G
S
Figure 3.5 Overdetermination on the goods sector when savings slopes up
and for the technical coef cient a we use Equation (3.17) also reproduced
below:
y^ 5 2a^ 5 Y(g 2 g), Y
g
. 0 (3.17)
The growth of nominal wages is again depicted by Equation (3.18)
W
^
5 q[A
W
2 A] (3.18)
but the desired wage share is now endogenized and made dependent on the
employment rate:
A
W
5 a
0
1 a
1
a
L
N
b 5 a
0
1 a
1
a
L
Q

Q
K
K
N
b 5 a
0
1 a
1
(auk) (3.46)
Finally, we recall that the local infation rate was set by a target defned by
the central bank, according to Equation (3.26):
p 5 p
T
(3.26)
The use of Equations (3.17), (3.18), (3.46), (3.37) and (3.26) in Equation
(3.16) leads to a diferential equation for the evolution of A:
A
^
5 qa
0
1 qa
1
au[i ]k 2 qA 2 Y[g 2 g] 2 p
T
(3.47)
The motion of k is described by:
k
^
5 K
^
2 N
^
where K
^
is the investment rate (g), and N
^
is assumed fxed at a level
denoted by n:
k
^
5 g 2 n (3.48)
In order to fnish writing Equations (3.47) and (3.48) in terms of A and
k, we have to replace g with its short-run equilibrium value.
8
Before we
proceed we must generate expressions for the desired investment rate and
total savings. The frst of these is provided by Equation (3.31) which, after
using Equation (3.40a), becomes:
g
d
5 g
d
{r[i, (1 2 A) ], i}, g
d
r
. 0, g
d
i
, 0, r
i
, 0, r
A
, 0 (3.49)
Total savings, on the other hand, is defned as domestic savings (sr) plus
foreign savings [B(h)]. From Equations (3.33), (3.40a) and our defnition
of foreign savings, we get
g
s
5 s{r[i, (1 2 A) ] } 1 B[h], s . 0, r
i
, 0, r
A
, 0 (3.50)
For the case depicted in Figure 3.4 the g level is determined by the corre-
sponding value of g
d
in Equation (3.49).
9
The system formed by Equations (3.47) and (3.48) may be rewritten as:
A
^
5 qa
0
1 qa
1
au[i ]k 2 qA 2 Y[g
d
{r[i, (1 2 A) ], i} 2 g] 2 p
T
(3.51)
k
^
5 g
d
{r[i, (1 2 A) ], i} 2 n (3.52)
The stability of the system is analysed by means of the Jacobian matrix:
Det(J) 5
0A
^
0A
0A
^
0k
0k
^
0A
0k
^
0k
5
(2q 2 Y
g
g
d
r
r
A
) qa
1
au[i ]
g
d
r
r
A
0
We have:
Det(J) 5 2[g
d
r
r
A
] [qa
1
au [i ] ] . 0
and
Tr(J) 5 (2q 2 Y
g
g
d
r
r
A
)
Since r
A
, 0, the long-run equilibrium will be stable when nominal wage
fexibility is important (q large) compared to learning efects (Y
g
). This
does not mean that the economy has to stay away from productivity gains;
it means instead that wage fexibility is necessary to accommodate produc-
tivity growth.
3.4 CONCLUDING REMARKS
In this chapter we compare the efects of diferent monetary regimes on
the growth and infation performance of an open economy. With a real
exchange rate target the decisions of the central bank are afected by
the trilemma of monetary policy, but the real exchange rate can still be
utilized to push employment and economic growth. Thus, although stabil-
ization based on open market operations has a tendency to lower economic
activity, the central bank can always target a level of the real exchange rate
which compensates, at least partially, the contractionary efects of stabil-
ization. The use of monetary instruments like the legal reserve requirement
may help bring infation down without the problems associated with the
trilemma. The long-run equilibrium is stable if nominal wages are fex-
ible, and the central bank maintains a competitive real exchange rate.
Under an infation targeting regime, the problems of the trilemma
disappear and prices are more easily stabilized. In this setting all eco-
nomic results depend on monetary policy: the goods sector is not neces-
sary to fnd equilibrium levels for the relevant variables. This situation
causes overdetermination as there is no level of the rate of proft which is
capable (except by accident) of bringing saving and investment to equal-
ity. Stabilization policy is harmful to growth and employment and, even
if savings increased, investment desires remain unsatisfed. The infation
targeting regime leads to a stable long-run equilibrium if nominal wages
are fexible enough to accommodate the productivity gains resulting from
learning.
The most important result in this chapter is the recognition of the criti-
cal importance that the real exchange rate has on the determination of the
rates of accumulation and employment of the small open economy. Thus
countries which have already decided to embark on infation targeting
regimes should allow themselves to tolerate higher infation goals to avoid
excessive appreciation of the real exchange rate.
For countries which still fnd themselves battling the dif culties of the
trilemma the recommendation would be to combine the use of open
market operations with the legal reserve requirement or even with some
control over the capital account of the balance of payments.
NOTES
1. The structuralist approach is described in Taylor (1991) and in Dutt (1992).
2. In other structuralist models (Dutt, 1984; Taylor, 1985, for example) income distribution
plays a critical role in the determination of economic growth. Blecker (1989) then extends
those models to examine how income distribution afects growth in an open economy.
In this chapter, however, the investment function (our Equation (2)) looks more like
the NeoKeynesian version used by Marglin (1984), and the ouput/capital ratio has been
taken out of the desired accumulation function. This modifcation prevents the wage
share from having an efect on the rate of growth.
3. The adjustment mechanism may be described by 0i/0t 5 f(M
s
2 M
d
) , f . 0; and this
will be stable only if f(2t
1
1 h
3
) , 0.
4. This point was analysed with a formal model in Cordero (2005).
5. The mainstream view argues that higher levels of the legal reserve requirement cause
higher administrative costs, and thus may lead to higher spreads and margins for
fnancial intermediation. But this is not necessarily true: in the case of Costa Rica, for
example, the reserve requirement fell from 15 percent in 1999 to 5 percent in 2002, but in
the same period the spread between active and passive interest rates declined only half a
percentage point. Then, in 2003, when the reserve requirement jumped from 5 percent to
10 percent, the spread went up by only 1.5 percent (Cordero, 2005).
6. The long-run motion of k is given by: k
^
5 g 2 N
^
. The short-run equilibrium level of g depends
on h: g
0
5 g(h), g
h
. 0. Assume that N
^
5 n(k) with n
k
. 0, and we get k
^
5 g(h) 2 n(k) .
The system formed by Equations (3.20) and (3.22) provides a solution for h, which we use to
fnd the level of k required for k
^
5 0. The equilibrium is stable if n
k
. 0.
7. With prices fxed in Equation (3.4), h adjusts as a result of variations in the nominal
exchange rate e.
8. Of course we also analysed a second case which here would require that we plug
Equation (3.40b) in Equation (3.31) for the gd. expression, and in Equation (3.33) for the
gs. equation. The long-run dynamics will be analogous to that of the frst case, but will
not be presented in the chapter.
9. If the interest rate that is consistent with the infation target falls to the right of the level
making G equal to S, then the g level is determined by the corresponding value of g
s
in
Equation (3.50). Again the dynamics of this case is analogous to the one resulting from
Figure 3.4 and will not be analysed here in detail.
REFERENCES
Arestis, P. and M. Sawyer (2003a), New consensus, new Keynesianism, and
the economics of the Third Way, Levy Economics Institute of Bard College
Arestis, P. and M. Sawyer (2003b), Infation targeting: a critical appraisal, Levy
Economics Institute of Bard College working paper no. 388.
Arestis, P. and M. Sawyer (2006), The nature and role of monetary policy when
money is endogenous, Cambridge Journal of Economics, 30 (6), 84760.
Arrow, K. (1962), The economic implications of learning-by-doing, Review of
Economic Studies, 29 (3), 15573.
Ball, R. and N. Sheridan (2003), Does infation targeting matter?, National
Bureau of Economic Research working paper no. 9577, Cambridge, MA.
Blecker, R. (1989), International competition, income distribution and economic
growth, Cambridge Journal of Economics, 13 (3), 395412.
Cordero, J. (1995), Essays on growth and distribution for open economies,
unpublished PhD dissertation, University of Notre Dame, Indiana.
Cordero, J. (2002), A model of growth and confict infation for a small open
economy, Metroeconomica, 53 (3), 26189.
Cordero, J. (2005), Infacion, politica monetaria y regimen cambiario en Costa
Rica, in J.R. Vargas and Y. Xirinachs (eds), La formacin de economistas:
Ensayos en honor de Pepita Echandi, San Jose, Costa Rica: Posgrado en
Economia, Universidad de Costa Rica.
Dutt, A. (1984), Stagnation, income distribution and monopoly power, Cambridge
Journal of Economics, 8 (1), 2540.
Dutt, A. (1992), Two issues on the state of development economics, in A. Dutt
and K. Jameson (eds), New Directions in Development Economics, Aldershot,
UK and Brookfeld, VT, USA: Edward Elgar, pp. 134.
Dutt, A. (1994), On the long run stability of capitalist economies, in A. Dutt (ed.),
New Directions in Analytical Political Economy, Aldershot, UK and Brookfeld,
VT, USA: Edward Elgar, pp. 93120.
Epstein, G. (2002), Employment-oriented central bank policy in an integrated
world economy: a reform proposal for South Africa, Political Economy
Research Institute working paper series no. 39, University of Massachusetts,
Amherst.
Epstein, G. (2005), Central banks as agents of economic development, Political
Economy Research Institute working paper series no. 104, University of
Massachusetts, Amherst.
Epstein, G. and E. Yeldan (2008), Infation targeting, employment creation
and economic development: assessing the impacts and policy alternatives,
International Review of Applied Economics, 22 (2) (March), 12930.
Ffrench Davis, R. (2003), Macroeconomic balances in emerging economies: the
confict between purely-fnancial and real-economy macrobalances, paper pre-
pared for the Task Force on Macroeconomic Policy at the Initiative for Policy
Dialogue, Columbia University, New York.
Chile and Mexico, paper prepared for the G24, Centro de Estudios de Estado
y Desarrollo, Argentina.
America, World Development, 34 (4) (April), 63146.
Galindo, L. and J. Ros (2008), Alternatives to infation targeting in Mexico,
International Review of Applied Economics, 22 (2), 20114.
Marglin, S. (1984), Growth, Distribution, and Prices, Cambridge, MA: Harvard
University Press.
Meyer, L.H. (2001), Does money matter?, Federal Reserve Bank of St Louis
Review, 83 (5), 116.
Pollin, R. and A. Zhu (2005), Infation and economic growth: a cross country
non-linear analysis, Political Economy Research Institute working paper no.
109, University of Massachusetts, Amherst.
Romer, D. (2000), Keynesian macroeconomics without the LM curve, Journal of
Setterfeld, M. (2006), Is infation targeting compatible with Post Keynesian eco-
nomics, Journal of Post Keynesian Economics, 28 (4) (Summer), 65372.
Taylor, J. (1993), Discretion versus policy rules in practice, Carnegie-Rochester
Conference Series on Public Policy, 39, 195214.
Taylor, L. (1985), A stagnationist model of economic growth, Cambridge Journal
of Economics, 9 (4), 383403.
Taylor, L. (1991), Income Distribution, Infation and Growth: Lectures on
Structuralist Macroeconomic Theory, Cambridge, MA: MIT Press.
PART II
Thematic issues: class relations and gender
impacts of infation targeting
71
4. Income, class and preferences
towards anti-infation and
anti-unemployment policies
Arjun Jayadev
1
4.1 INTRODUCTION
From one important point of view, indeed, the avoidance of infation and the
maintenance of full employment can be most usefully regarded as conficting class
interests of the bourgeoisie and the proletariat respectively, the confict being
resolvable only by the test of relative political power in the society.
Harry Johnson (Johnson, 1968, p. 986)
Among the casualties of the advent of the rational expectations revolu-
tion in the 1970s was a rich vein of political economy which analysed the
macroeconomic dynamics of unemployment and infation as deriving
from distributive struggles between capitalists and workers (Bach and
Stephenson, 1974; Boddy and Crotty, 1975; Hibbs, 1977; Rosenberg and
Weisskopf, 1981; Rowthorn, 1977). This approach often made explicit the
class confict innate in Keynesian accounts of the Phillips curve trade-of
and in considerations of the natural rate of unemployment.
2
While the
details of these studies varied, the argument, with slight modifcations,
remained the same: workers and capitalists had opposing and irreconcil-
able diferences in the trade-of between infation and unemployment.
Such confict theories not only gave political explanations for the trajec-
tory of infation and unemployment (most notably in the work of Hibbs
(1977)) but also thereby provided direct predictions for the preferences of
individuals towards anti-unemployment and anti-infation policies based
on their position within the social stratifcation. Specifcally, the working
class, broadly classifed, is more concerned about reducing unemployment
than frm owners. Unemployment reduces the lifetime income of workers
directly and also exerts a downward pressure on wages by reducing the
bargaining power of workers.
3
Infation, on the other hand, worked to the
advantage of those workers with low savings.
With the rise of new classical macroeconomics in the 1970s and its
subsequent hegemony any trade-of between infation and unemploy-
ment came to be seen as essentially short term and certainly could not
be utilized by policy makers to afect macroeconomic outcomes without
incurring severe macroeconomic costs. As a consequence, much of
the macroeconomic literature moved away from class-based, political
economy models of infation and unemployment towards what Iversen
and Soskice (2006, p. 425) elegantly call macroeconomics which [f]ocuses
attention on what democratic governments can do wrong in the short
term. The prescription that followed was to replace government with
independent central banks, and discretionary macroeconomic policy with
rules-based approaches such as infation targeting (Barro and Gordon,
1983; Cukierman, 1992).
While post-Keynesian and heterodox approaches never abandoned the
idea of demand management, the advent of New Keynesian economics
restored the space in mainstream economics for economic policy to have
benefcial macroeconomic outcomes. As opposed to rational expectations,
there is a role for demand management and other policies when imperfec-
tions arise due to wage and price rigidities whereby involuntary unem-
ployment can be reduced. The now long literature on the non-accelerating
infation rate of unemployment (NAIRU) continues to suggest that com-
bating infation and unemployment involves two independent and poten-
tially opposing targets.
4
Given that there has been a restitution of space for macroeconomic
policy, this chapter seeks to make a contribution in reconstructing the
distributional consequences of anti-infationary and anti-unemployment
policies, and attitudes towards them. Recent research on the impact of
such policies on the poor and the rich has used social survey data to
answer part of this question. Scheve (2004) and Jayadev (2006) both fnd
strong results that the relatively poor are less likely than the relatively
rich to prioritize tackling infation as opposed to tackling unemploy-
ment, presumably because they take the latter to be a more important
problem facing them. Other research (Jayadev, 2008) continues this
line of research by assessing the class content of disinfationary and
expansionary policy. This research fnds support for the contention that
those in contrary class positions respond very diferently to policies
engineered to combat infation versus those that are designed to combat
unemployment. There are substantial class-based diferences in what
may be termed relative aversion to infation and unemployment (a
preference that policy is designed to keep infation down rather than
unemployment down). Capitalists and highly skilled workers are more
likely to display relative infation aversion than less skilled and unskilled
workers. I briefy extend the analysis to look at the relationship between
Anti-infation and anti-unemployment policies 73
relative infation aversion and other attitudes towards macroeconomic
policy. Individuals who support broadly pro-labor, solidaristic and
redistributionary policies for the government are signifcantly less rela-
tively infation averse (or what is the same thing, more relatively unem-
ployment averse).
This chapter summarizes the fndings of these two lines of research
(that is, the income and class-based preferences for combating infation
rather than unemployment). As such, the chapter provides some indi-
cation for the degree and kind of political support that can be drawn
for policies that aim to combat infation versus those which combat
unemployment. The results have implications for the debate on anti-
infationary policies and infation targeting in particular as they apply to
diferent countries.
I very briefy summarize relevant research on the income and class
impacts of infation and unemployment and attitudes towards these prob-
lems in Section 4.2. Section 4.3 describes the data that I use. Section 4.4
presents the results of various logistic regressions. Section 4.5 summarizes
and concludes.
4.2 INCOME AND ATTITUDES TOWARDS
INFLATION AND UNEMPLOYMENT
The theoretical impetus for the shift in focus mentioned in Section 1 was
provided by the rational expectations revolution of the 1970s and sub-
sequent neo and new classical analyses of the state which suggested that
expansionary policies were short-term palliatives at best, and in the medium
to long run, simply infationary. That is to say, expansion, when driven
artifcially by the state using fscal or monetary authorities, may well have
a positive efect on the poor in the short run by increasing the growth and
employment rates in the economy; but in the long run will simply lead to
more infation and a fall in employment back to its natural rate. The impact
on the poor, it follows, is possibly a net deterioration in their welfare, as
unemployment remains at previous levels and goods become more costly.
Thus the standard prescription of expansionary policy in the 1960s and
1970s was replaced by the orthodoxy of infation targeting in the 1980s and
1990s, at least partly as a manner in which the poor were to be protected
from a perhaps well meaning, but misguided government.
The theoretical linkages between infation and poverty do not, however,
lead to the direct conclusion that infation is necessarily bad for the poor.
A brief account of the channels by which expansionary and potentially
infationary policy afects the poor as identifed in Romer and Romer
(1998), Easterly and Fischer (2001) and Cardoso (1992) among others is
summarized in Table 4.1.
Given these ambiguous theoretical relationships, both at the macroeco-
nomic and microeconomic levels, the link between infation and the poor
is strongly an empirical matter, and certainly highly dependent on con-
ditioning structural factors.
One of the more infuential papers linking anti-infationary policy and
the poor is that by Easterly and Fischer ( 2001). In a novel analysis the
authors turn to an unimpeachable authority on the subject of infation
Table 4.1 Some hypothesized linkages between infation and poverty
Channel Efect on welfare of poor
Short run
Growth Expansionary demand policy, in raising average
income (without a change in the distribution of
income) has a direct efect on reducing poverty.
Real wage and transfers To the extent that wages and transfers are
unindexed to infation, and that certain
commodities (for example, food) form a larger
portion of the consumption basket of the poor,
the real income of the poor falls.
Medium to long term
Growth Romer and Romer (1998) suggest a strong positive
correlation between high infation and low growth,
as well as high infation and instability of demand,
thus suggesting that short-run gains for the poor
from expansionary demand policy is eroded in the
long run.
Credit market It is assumed that unanticipated infation is
positive for debtors and negative for creditors.
Given that the poor fall overwhelmingly in the
former category, it should be the case that they
beneft to some extent. However, the size of this
efect depends on how much the poor owe and
how much the real wage declines.
Uncertainty Uncertainty caused by infation causes fnancial
market dislocations, increases capital fight and
reduces the rate of growth through reducing
physical and human capital investment; all of
which have a negative impact on the poor.
and the poor the poor themselves. Easterly and Fischer use a large social
survey conducted in 38 countries in 1996 to analyse the attitudes towards
infation of individuals at diferent income levels. Their study suggests
a robust efect: the self-identifed poor were more likely than the self-
identifed rich to mention infation
5
as among the top three concerns for
the economy. They fnd, by contrast, that there is no statistically signifcant
diference between the rich and the poor in terms of their attitude towards
recession and unemployment. From their study it is easy to draw the
conclusion that anti-infationary policy is preferred by the poor to policy
designed to combat unemployment or recession. Expansionary policy
which leads to infation can be thought of as being akin to a tax:
6
redistri-
bution from the public at large to facilitate increased public expenditure.
As such, asking about peoples preferences towards infation without
presenting the alternative is similar to asking them about their preferences
towards lowering taxes.
From the macroeconomic policy viewpoint, an important and relevant
opportunity cost of lower infation is higher unemployment and informal-
ization, especially when it is achieved by reducing government spending.
As such, a more useful question to ask would be to ask the poor and
the rich their relative preferences between infation and unemployment.
Ideally, the question that should be posed is of the form: if the govern-
ment was to reduce infation, even if it meant increasing unemployment
a little, would you support it? While the Roper Starch Survey used by
Easterly and Fischer does not unfortunately provide such an option, there
is another dataset which poses the relevant question more closely and
directly (more on which in Section 4.3). As a result, I use this survey to
analyse the attitudes of the poor towards infation and unemployment.
4.3 CLASS AND ATTITUDES TOWARDS
INFLATION AND UNEMPLOYMENT
The radical political economy approaches to macroeconomics mentioned
in the introduction saw unemployment as acting as a regulator of class
confict (Rowthorn, 1977). Unemployment was the central fulcrum of
the labor-capital confrontation. Specifcally, increases in unemployment
maintained a downward pressure on wages while tight labor markets
increased factor income going to labor by exerting upward pressure on
wages (Boddy and Crotty, 1975).
7
Confict theories of infation suggested that the rise in infation was seen
as the result of perpetual claims by workers for wage increases ahead of
productivity. The detrimental impact of infation was primarily on those
who had nominally denominated assets whose value was eroded with
increasing prices, although there were also pressures on frms which were
unable to pass on higher wages as higher prices. To that extent the nega-
tive impact of infation was more pronounced on the capitalist and rentier
class. As a result these theories would predict pronounced diferences in
attitudes towards infation and unemployment depending on the respon-
dents class position.
Drawing from this political economy research, empirical studies in the
1970s and 1980s had begun to establish the class character of individuals
preferences for infation versus unemployment aversion. Given objective
evidence that periods of relatively low unemployment and relatively high
infation coincided with an equalization of the personal distribution of
income, a larger share of national income going to labor versus capital and
a reduction in poverty as well as losses to those with savings in nominally
denominated assets (typically the rich), attitudes towards infation and
unemployment had a class character. Hibbs (1977, p.1470) summarizes the
central fndings from US and UK surveys:
Popular concern about unemployment and infation is class-related. Low and
middle income and occupational status groups are more averse to unemploy-
ment than infation, whereas, upper income and occupational status groups are
more concerned about infation than unemployment . . . it does appear that the
subjective preferences of class or status groups are at least roughly in accord-
ance with their objective economic interests . . .
Recent studies of attitudes towards infation and unemployment have
largely ignored class. While there is substantial empirical evidence from
opinion research that both infation and unemployment are seen by
respondents as disutilities (see, among others, Di Tella et al., 2001;
Easterly and Fischer, 2001; Shiller, 1997), there is less consensus on the
relative importance that individuals in diferent classes place on reducing
each of these. Part of the issue is simply that there have been few surveys
done which explicitly ask the respondent to rate their aversion to infation
versus unemployment if these were alternative outcomes. As such, the
data has limited researchers agenda. Equally, there has been little interest
in the characteristics of individuals who might support a policy designed
to combat infation versus one that tackled unemployment. To the extent
that this has been undertaken, it has been to assess infation aversion
among the rich versus the poor (Easterly and Fischer, 2001; Jayadev,
2006).
8
There has been little or no recent work which attempts to look at
politics and in particular class politics in the determination of preferences
towards anti-infation and anti-unemployment policy. It is to this exercise
that we now turn.
4.4 DATA
4.4.1 Measuring Infation Aversion
The International Social Survey Program (ISSP) conducted by the Inter-
university Consortium for Political and Social Research in 1996 focuses on
the preferences of more than 30 000 individuals in 27
9
diferent countries
regarding the role of government in society. The countries included include
OECD economies, former eastern bloc economies and unfortunately do
not include any low income countries apart from China.
10
Among the
questions asked in this survey is the following:
If the government had to choose between keeping down infation and keeping down
unemployment to which do you think it should give highest priority?
This is the key variable of analysis for the rest of the chapter. I defne a
variable relative infation aversion as taking a value of one when the
respondent prefers that the government prioritize reducing infation rather
than unemployment and zero when the opposite holds. Summary statistics
for all variables used are provided in Table 4.1.
While this provides a direct measure of the weights placed in an indi-
vidual welfare function on infation as opposed to unemployment, it is not
without some limitations. Ideally, such a question might ask how much
infation the individual might accept for reducing the level of unemploy-
ment and vice versa so as to have a more direct calibration of the marginal
rate of substitution in the social and individual welfare function. However,
the measure is certainly superior to questions which ask about infation
without reference to unemployment or any other macroeconomic policy
objective, thereby providing no implicit budget constraint.
Figure 4.1 summarizes the average preference for keeping infation
rather than unemployment down by country.
Some interesting observations suggest themselves. Nearly 42 percent of
the overall sample report being relatively infation averse. However, this
masks large diferences in the average relative infation aversion between
countries, from a low of less than 20 percent of respondents in France to a
high of above 60 percent of respondents in the Czech Republic. In only fve
countries out of 20 is the percentage of relatively infation averse respon-
dents over half (and only in two countries West Germany and the Czech
Republic is the percentage overwhelmingly above the midway mark).
In order to assess the relative infation aversion of the rich and the poor,
I use the income question of the survey which asks the respondent their
personal income in the last year. The income of the individual is grouped
according to within country quintiles and quintile dummies take a value
of 1 if the income of the respondent is in that quintile (and zero otherwise).
I also defne a relative infation aversion dummy that takes the value of
1 if the respondent prefers the government to keep down infation rather
than unemployment.
4.4.2 Measuring Class
Constructing readily comparable and objective measures of social class
are fraught with dif culties (see Leiulfsrud et al. (2005) and Wright (1997)
for an exposition on some of these). Contemporary measures of class
difer based on the elements that researchers consider important to their
theoretical approach for example, along lines of ownership, manage-
ment, career prospects, income, status, education or other such categories
(Carchedi, 1977; Erikson and Goldthorpe, 1993; Esping-Andersen, 1992;
Wright, 1985). However, not all of these are likely to bear directly upon
the question of relative infation aversion. Class has relevance in as much
as it refects the labor market position of the respondent and therefore
their preferences to policies which enhances their position. It is desirable,
1
8
.
3
2
0
.
6
2
1
.
5
2
2
.
4
3
0
.
03
4
.
0
3
4
.
9
3
9
.
5
4
2
.
4
4
4
.
0
4
6
.
3
4
7
.
0
4
7
.
1
4
9
.
5
5
0
.
2
5
3
.
4
5
4
.
8
6
1
.
8
6
4
.
9
4
1
.
5
4
7
.
6
0
10
20
30
40
P
e
r
c
e
n
t
a
g
e

(
%
)
50
60
70
F
r
a
n
c
e
I
s
r
a
e
l
-
A
r
a
b
s
S
p
a
i
n
I
t
a
l
y
L
a
t
v
i
a
I
r
e
l
a
n
d
C
h
i
n
a
C
a
n
a
d
a
A
u
s
t
r
a
l
i
a
R
u
s
s
i
a
I
s
r
a
e
l
-
J
e
w
s
U
n
i
t
e
d

S
t
a
t
e
s
P
o
l
a
n
d
N
e
w

Z
e
a
l
a
n
d
S
l
o
v
e
n
i
a
G
e
r
m
a
n
y

(
E
a
s
t
)
J
a
p
a
n
G
e
r
m
a
n
y

(
W
e
s
t
)
H
u
n
g
a
r
y
C
z
e
c
h

R
e
p
u
b
l
i
c
O
v
e
r
a
l
l

Figure 4.1 Percentage of respondents who prefer that the government
keeps infation down rather than unemployment down
therefore, to utilize a defnition which closely refects the respondents
occupation and position as employer or worker.
In this chapter, I utilize a few diferent measures of class. The ISSP
dataset provides an occupational variable based on the International
Standard Classifcations of Occupations (ISCO) 1988 classifcation which
has previously been used by researchers to construct a wide array of meas-
ures of stratifcation. A frst (crude) method is to utilize the traditional
Marxist division between frm owner and employee. If the respondent is
classifed as self-employed with employees, they are classifed as a frm
owner, and an employee otherwise. I defne a dummy variable called frm
owner which takes the value of 1 if the respondent is self-employed and has
more than one employee, and 0 otherwise.
Sociologists have expended enormous efort in providing more useful
and sophisticated categorizations of class. In order to perform a more sat-
isfactory class analysis, I replicate Wrights (1985) scheme which divides
the labor force into owners and wage laborers and wage laborers in turn
into three categories experts, skilled and low skilled. Wright uses this
to operationalize his idea of class locations and contradictions therein.
Workers may be divided according to their relative privilege in the labor
process. Based on a careful cataloging of occupations, Wright defnes
experts as those whose jobs require skills (and in particular accredited or

credentialed skills) and who are in scarce supply relative to their demand

by the market. Semi-skilled and unskilled class positions by contrast are
held by those who have uncredentialed or no skills and who are thus
in abundant supply. Using this approach has signifcant advantages.
It makes theoretical sense for the question at hand to conceive of class
measures which refect the respondents relationship to the labor market
and therefore to their bargaining power and probability of continued
employment. An expert, for example, will typically enjoy a credential rent
and be more likely to have both a higher level of bargaining power and a
lower probability of being replaced than a low-skilled worker. They may
therefore have opposing ideological and political interests based on other
workers. At the same time both an unskilled worker and a skilled worker
are more concerned about unemployment than a capitalist.
11
Appendix
4.A1 details the creation of the class variables.
Figure 4.2 shows the average relative infation aversion in each country
for each grouping of wage laborers. As is evident, in most coun-
tries experts are more relatively infation averse than semi-skilled and
unskilled workers,
12
as might be expected given the logic that more highly
skilled workers enjoy greater bargaining power and a lower probability
of unemployment than lower skilled workers, but are equally likely to see
their wages eroded by infation.
Another approach is to look at subjective evaluations of class cat-
egories. The ISSP dataset asks respondents their own evaluation of their
social class (the categories are lower middle class, upper working class,
middle class, upper middle class and upper class). Unlike more objective
measures, subjective perceptions of class probably depend on an amalgam
of factors such as the respondents education, status, income and gender
as well as factors which are of direct relevance to the question of relative
infation aversion the respondents position and prospects in the labor
market and the asset market. Nevertheless, it is useful to look at this as
a check on the robustness of the earlier measure of class. I defne three
subjective class categories the variable subjective lower class takes a value
of 1 when the respondent identifes as being in the lower class or in the
working class and 0 otherwise. Another variable subjective middle class
takes a value of 1 if the respondent is from the lower middle class or middle
class and 0 otherwise. Finally, the value of the variable subjective upper
class takes a value of 1 when the respondent is from the upper middle or
upper class.
4.5 RESULTS AND EXTENSIONS
In order to assess the income and class character of relative infation aver-
sion, I undertake a series of logistic regressions of relative infation aver-
sion on the respondents class position. Tables 4.2, 4.3, 4.4 and 4.5 provide
the detailed results of these exercises.
0
10
20
30
40
50
60
70
80
90
A
u
s
t
r
a
l
i
a
C
a
n
a
d
a
C
h
i
n
a
C
z
e
c
h

R
e
p
u
b
l
i
c
E
a
s
t

G
e
r
m
a
n
y
F
r
a
n
c
e
H
u
n
g
a
r
y
I
r
e
l
a
n
d
I
s
r
a
e
l

A
r
a
b
s
I
s
r
a
e
l

J
e
w
s
L
a
t
v
i
a
N
e
w

Z
e
a
l
a
n
d
P
o
l
a
n
d
R
u
s
s
i
a
S
l
o
v
a
k
i
a
S
p
a
i
n
U
n
i
t
e
d

S
t
a
t
e
s
W
e
s
t

G
e
r
m
a
n
y
P
e
r
c
e
n
t
a
g
e

(
%
)
Skilled workers
Low-skilled workers
Experts
Figure 4.2 Percentage of each category who are relatively infation averse
Table 4.2 Logistic regression on the likelihood that a respondent will
prefer the government to keep infation down rather than
unemployment down
Variable Coef cient z-statistic
Lowest income quintile 0.56* (10.5)
Second income quintile 0.67* (6.6)
Third income quintile 0.76* (4.8)
Fourth income quintile 0.82* (3.4)
Number of observations 14 340
Note: * 5 signifcant at the 1 percent level. Omitted variable is richest quintile. Country
intercept dummies are included but not shown.
Table 4.3 Logistic regression on the likelihood that a respondent will
prefer the government to keep infation down rather than
unemployment down, with controls
Variable Coef cient z-statistic
Lowest income quintile 0.67* (6.5)
Second income quintile 0.75* (4.5)
Third income quintile 0.83* (3.3)
Fourth income quintile 0.87** (2.3)
Some primary education 0.89** (2.1)
Some secondary education 0.86* (3.1)
Person in 20s 1.31** (2.3)
Person in 30s 1.52* (3.6)
Person in 40s 1.29* (2.2)
Person in 50s 1.25*** (1.9)
Person in 60s 1.43* (3.0)
Person in 70s 1.30** (2.0)
Person above 70s 1.26 (1.4)
Unemployed 0.82* (5.4)
Female 0.72* (3.6)
Trade union member 0.81* (4.3)
Voted for right wing party 1.39* (7.6)
Number of observations 14 340
Note: * 5 signifcant at the 1 percent level, ** 5 signifcant at the 5 percent level,
*** 5 signifcant at the 10 percent level. Omitted variable is richest quintile for the income
variables, at least some university education for the education variables and persons less
than 20 years of age for the age variables. Country intercept dummies are included but not
shown.
Table 4.2 reports the results from a logistic regression of the relative
infation aversion dummy on the income quintile dummies. The ffth,
or richest, quintile is the omitted variable, so that the coef cients on the
income quintile dummies measure the diference between the coef cient on
that income category and on the richest quintile. In order to control for
country diferences in the national averages of relative infation aversion,
the regression includes average individual country relative infation as
dummies, but these are not shown in the table.
A strong and consistent result is immediately evident. Relative infation
aversion (or the likelihood of preferring the government to keep infation
down rather than unemployment down) is decreasing in the income of the
respondent. The odds ratio falls monotonically with the fall in the income
category, and the coef cients are all highly signifcant. The odds ratios
suggest that a respondent in the lowest quintile is only about half as likely
as someone in the richest quintile to display a preference for keeping infa-
tion down rather than unemployment. Someone in the second and third
quintile is about two-thirds and three-fourths as likely as someone in the
top quintile to display relative infation aversion, respectively.
Table 4.3 repeats the analysis of Table 4.2 but controls for several other
characteristics of the respondents. The frst control is for the level of edu-
cation. The prior expectation for this variable is unclear: a lower level of
education makes an individuals welfare potentially more vulnerable to
both unemployment (lower human capital reduces employment oppor-
tunities) and infation (less education means less knowledge and ability
to protect ones income from infation). The coef cients from column 2
suggest, however, that less educated individuals are less relatively infa-
tion averse than more educated individuals. While statistically signifcant,
these efects are not large: both primary and secondary educated individ-
uals are about 90 percent as likely as university educated people to prefer
that the government keep down infation rather than unemployment if it
had to choose one of the two.
A second control used is the age of the respondent. We may expect that
the elderly are more concerned about infation versus unemployment than
the young, since the elderly are likely to be living of accumulated assets
and pensions (which can easily be eroded through infation) rather than
wage income. The regression suggests that it is certainly the case that
respondents in all age groups are more relatively infation averse than
those in the omitted group (those below age 20). However, those in their
60s and 70s do not appear to be more relatively infation averse than
those in their 20s and 30s.
The last four rows report some interesting results. Not surprisingly,
the unemployed are less likely than those who have employment to prefer
that the government pursue a policy of keeping infation down rather than
unemployment. Women, too, are less relatively infation averse than men
and are about seven-tenths as likely as men to be so. Trade union members
display statistically signifcantly less relative infation aversion than non-
union members, perhaps because union members can more easily bargain
their wages upwards than non-union members in the face of rising prices.
Finally, political af liation has an extremely strong predictive efect on the
question at hand. Those who voted for the right wing party in the country
in the last election are about 1.4 times as likely as those who did not to
display relative infation aversion.
As the coef cients on the income quintile dummies show, the result that
the poor are less likely than the rich to prefer that the government keep
infation down rather than unemployment remains true in the presence of
controls. A respondent in the poorest quintile is about two-thirds as likely
as one in the richest quintile to say so. This central result remains robust to
the inclusion of several additional controls, such as occupation, household
size and whether the individual works for the public or private sector.
Moving now to considering the results on class or job attachment,
columns I-III in Table 4.4 are the results from performing the logistic
regression on the three defnitions of class without any controls. As is
evident from column I, frm owners are signifcantly more likely than
workers to be relatively infation averse (or less likely to be relatively
Table 4.4 Logistic regression of relative infation aversion on class
variables (without controls)
Variable I II III
Odds
ratio
Z-statistic Odds
ratio
Z-statistic Odds
ratio
Z-statistic
Firm owner 1.47*** (5.53)
Unskilled 0.69*** (6.66)
Semi-skilled 0.70*** (6.44)
Subjective
lower class 0.69*** (6.37)
Subjective
middle class 0.84*** (3.15)
Number of
observations 23 824 13 955 20 437
Note: *** 5 signifcant at the 1 percent level, ** 5 signifcant at the 5 percent level, * 5
signifcant at the 10 percent level. Omitted variable is experts (high skilled) in column II, and
subjective upper class in column III. Country average infation dummies are included but
not shown.
unemployment averse). Column II uses the classifcation for wage laborers
developed by Wright. The omitted dummy is expert workers and hence the
results suggest that as compared to experts, both low-skilled and skilled
workers display less relative infation aversion suggests that in comparison
with the omitted group. Similarly, in column III in comparison to those
who consider themselves upper class, those who consider themselves
middle and lower class are much less relatively infation averse.
Columns I-III in Table 4.5 show that these results persist in the pres-
ence of a variety of plausible controls, including dummies for income,
gender, age, employment status and union membership. Column I shows
Table 4.5 Logistic regression of relative infation aversion on class
variables (with controls)
Variable I II III
Odds
ratio
Z-statistic Odds
ratio
Z-statistic Odds
ratio
Z-statistic
Firm owner 1.20** (2.11)
Unskilled 0.87** (1.98)
Semi-skilled 0.82*** (2.86)
Subjective
lower class 0.75*** (3.72)
Subjective
middle class 0.86** (2.08)
Lowest income
quintile 0.63*** (7.95) 0.66*** (5.51) 0.68*** (6.28)
Second income
quintile 0.73*** (5.17) 0.75*** (3.65) 0.76*** (4.24)
Third income
quintile 0.80*** (3.97) 0.81*** (3.18) 0.85*** (2.67)
Fourth income
quintile 0.86** (2.50) 0.92 (1.17) 0.89** (1.86)
Unemployed 0.70*** (3.80) 0.74** (1.86) 0.71*** (3.58)
Female 0.82*** (5.43) 0.80*** (4.75) 0.80*** (5.76)
Union member 0.79*** (5.10) 0.81*** (3.90) 0.80*** (4.76)
Age 1.00 (0.97) 1.00 (0.53) 1.00 (1.24)
Number of
observations 14 245 9 273 13 469
Note: *** 5 signifcant at the 1 percent level, ** 5 signifcant at the 5 percent level,
* 5 signifcant at the 10 percent level. Omitted variable are experts (high skilled) and highest
income quintile in column II, and subjective upper class and highest income quintile in
column III. Country average infation dummies are included but not shown.
that a frm owner is about a fourth more likely to prefer anti-infation to
anti-unemployment policies as a worker. Column II shows that relative
to experts, semi-skilled are about eight-tenths and unskilled workers
are about nine-tenths as likely to report supporting anti-infation to anti-
unemployment policies. The results in column III show that as compared
to the subjective upper class, subjective lower classes display signifcantly
less infation aversion. A respondent who considers themself as being
part of the lower or lower middle class is about three-fourths as likely
as someone who is in the upper class to prefer that the government keep
infation down. A respondent in the middle or upper middle class is about
eight-tenths as likely as someone in the upper middle or upper class to
prefer that infation be kept down rather than unemployment.
What kinds of other policy preferences are associated with relative infa-
tion aversion? If the class content narrative is correct and not simply an
artifact of the occupational categorization used in this chapter to defne
class, we should fnd that there are other attitudes in common with respect
to class politics. To test this, I use questions from the ISSP dataset which
speak to the respondents stance towards pro-business and pro-labor poli-
cies and attitudes, on the one hand, and towards redistributionary policies
on the other. The dataset provides a host of questions which could be used
for the purpose and while those used here are not exhaustive, using other
indicators provides very similar results.
In order to gauge attitudes towards labor and towards business I con-
struct several dummy variables which could be said to represent anti-labor
and pro-business sentiment. I defne a dummy variable pro wage control
when the respondent replies being strongly in favor or in favor of wage
control by law. Another dummy variable anti jobs for all refers to the situ-
ation when the respondent believes that it should not be the governments
responsibility to provide a job for everyone who wants one. If the respon-
dent is against the idea of the government fnancing of new jobs then they
are said to be anti government jobs. The variable pro deregulation refers to
the situation the respondent is for government deregulation of business.
Anti protect jobs takes a value of 1 if the respondent is against the protec-
tion of declining industries in order to protect jobs. Finally the dummy
variables power of labor and power of business refer to situations where the
respondent believes the political power of labor and the political power
of business to be too strong respectively. Column I in Table 4.6 shows the
results of a logistic regression of the infation aversion variable on these
dummies. In concordance with the narrative of class-based diferences, all
the odds ratios are statistically signifcant and in the expected direction.
There is a strong correspondence between pro-business and anti-labor pref-
erences, on the one hand, and relative infation aversion, on the other.
Attitudes towards redistribution are approximated by three dummies.
A respondent is anti redistribution if they respond that the government
should not redistribute wealth, and is anti equalization if they respond
that it is not the governments responsibility to reduce income diferences
between the rich and the poor. If the respondent prefers reducing taxes
even if this means lower social services then the dummy variable lower
social services takes a value of 1. Column II in Table 4.6 shows the results
of a logistic regression of the infation aversion variable on these dummies.
The odds ratios are all statistically signifcant and show that relative infa-
tion aversion is stronger among respondents who are also against gov-
ernment action to redistribute income and wealth and to provide social
services at the potential cost of higher taxation.
4.6 CONCLUSION
In the last three decades the hopeful message of Keynesian demand
management has fallen out of favor with policy makers; as more faith
has been placed in market-based solutions, in independent central banks
and in microeconomic interventions to handle the problems of infation
and unemployment. The theoretical impetus for this shift was provided
by the rational expectations revolution of the 1970s and subsequent new
Table 4.6 Logistic regression of relative infation aversion on other
attitudes
Variable I II
Odds ratio Z-statistic Odds ratio Z-statistic
Pro wage control 1.10** (2.55)
Anti jobs for all 1.51*** (10.03)
Anti government jobs 1.49*** (5.39)
Pro deregulation 1.37*** (2.70)
Anti protect jobs 1.56*** (9.96)
Power of labor 1.12*** (2.97)
Power of business 0.86*** (3.22)
Anti redistribution 1.32*** (6.28)
Anti equalization 1.45*** (8.98)
Lower social services 1.40*** (10.26)
Number of observations 14 533 17 743
Note: *** 5 signifcant at the 1 percent level, ** 5 signifcant at the 5 percent level,
* 5 signifcant at the 10 percent level. Country average infation dummies are included but
not shown.
classical analyses of the state. These in turn argued that democratic gov-
ernments were often bad for macroeconomic ef ciency as they would tend
to increase defcits and be unable to credibly tackle infationary pressures.
Worse still, any attempt to artifcially reduce the unemployment rate
would lead inevitably to higher infation with no efect beyond the very
short term on the unemployment rate. Thus, the standard prescription of
earlier times was replaced by the orthodoxy of central bank independence
and increasingly a narrow focus on infation targeting (Bernanke et al.,
1999). For proponents of this view, delegating responsibility to an author-
ity which can credibly commit to a single target is benefcial in increasing
macroeconomic ef ciency and protecting the public from a perhaps well
meaning, but misguided government.
A movement away from commitment to full employment and towards
low targeted infation has potentially profound distributional conse-
quences. Despite the claims made by some that anti-infationary policy is,
for example, pro poor, it is an empirical question as to whether infation
or unemployment is seen as a bigger problem by diferent individuals if
there is a trade-of between these. Both infation and unemployment are
more likely, for example, to afect the poor, but this study as well as recent
research on infation versus unemployment aversion (Jayadev, 2006;
Scheve, 2004) found that the rich are more relatively infation averse than
the poor. Other results presented in this chapter further strengthen this
idea by fnding that relative infation aversion is more pronounced among
the privileged or elite broadly defned in class terms. Since class is a critical
variable in determining an individuals labor market opportunities as well
as the source and variability of their income, the results make sense. The
fndings are in concordance with confict-based models of unemployment
and infation which argue that macroeconomic policies may have system-
atically diferential efects on the welfare of workers and owners (as well
as on diferent segments of the working class) and that the preferences of
individuals in separate class positions refect these diferences. These fnd-
ings have important consequences for research on the implications of anti-
infation versus anti-unemployment policies in general and on the more
current debate around infation targeting in particular.
NOTES
1. The author is grateful to Gerald Epstein, Christian Weller, Sam Bowles, Kade
Finnof, Ozgur Orhangazi, James Heintz, Suresh Naidu and Erinc Yeldan for useful
comments.
2. Indeed, as Pollin (1998) points out, class confict is the implicit mechanism that drives
the natural rate of unemployment even in orthodox neoclassical accounts. As he puts
it Marx and Kalecki also share a common conclusion with natural rate proponents, in
that they would all agree that positive unemployment rates are the outgrowth of class
struggle over distribution of income and political power. . . . Of course, Friedman and
the New Classicals reach this conclusion via analytic and political perspectives that
are diametrically opposite to those of Marx and Kalecki. To put it in a nutshell, mass
unemployment results in the Friedmanite/New Classical view when workers demand
more than they deserve, while for Marx and Kalecki, capitalists use the weapon of
unemployment to prevent workers from getting their just due (Pollin, 1998).
3. The idea of unemployment as a labor disciplining device has, of course, a provenance
from Marx. Marxist theories of the labor market maintain this as a key fact of the labor
market (for example, Bowles, 1985).
4. For a recent review see Ball and Mankiw (2002).
5. At the outset, it is important to note a potentially serious problem with the data. The
choice for infation as worded in the survey is for infation and high prices. As Easterly
and Fischer themselves note, this raises the problem that in complaining about high
prices, the poor are only naturally complaining about low incomes. They suggest that
the low correlation between this variable and another option available to the respon-
dent money not enough to live by is proof that the poor are indeed protesting infation
and not low income. This is not, however, truly satisfactory. It is not entirely clear that
a poor respondent would list both infation and high prices and money not enough to
live by as being within the top three problems facing the economy if they interpret the
frst option as being about their low income.
6. Indeed it is often spoken about as being a cruel tax, whereby the publics real earnings
are eroded and transferred to the state.
7. Recent empirical studies, especially about the US experience of the 1990s boom, support
these ideas indirectly. Bernstein and Baker (2003) fnd that the low unemployment
period in the US economy of the late 1990s aided in improving the welfare of workers
according to several metrics. Real wages increased after a generation of decline, and the
infation adjusted income of low income families grew by twice the amount that they
did in the 1980s expansion (when average unemployment was higher). Abraham and
Haltiwanger (1995) who review mainly US evidence suggest that real wages are more
likely to be pro-cyclical than counter-cyclical.
8. Scheve (2003, 2004) remains an exception in providing more detailed evidence for the
characteristics of individuals supporting each policy.
9. In the survey respondents from Israel and Germany are divided in two separate cat-
egories each. The former is divided between Israeli Arabs and Israeli Jews and the latter
is split between East and West Germans.
10. Survey was conducted during 2006.
11. This is an example of what Wright terms a contradictory class location.
12. The major exception is Israel perhaps because of the impact that the period of hyper-
infation in the 1970s had upon even skilled workers.
REFERENCES
Abraham K.G. and J.C. Haltiwanger (1995), Real wages and the business cycle,
Journal of Economic Literature, 33, 121564.
Bach, G. and J. Stephenson (1974), Infation and the redistribution of wealth,
Review of Economics and Statistics, 61, 113.
Ball, L. and G.N. Mankiw (2002), The NAIRU in theory and practice, Johns
Hopkins University working papers no. 475.
Barro, R. and D. Gordon (1983), A positive theory of monetary policy in a
natural rate model, Journal of Political Economy, 31, 589610.
Bernstein, J. and D. Baker (2003), The Benefts of Full Employment: When Markets
Work for People, Washington, DC: Economic Policy Institute.
Bernanke, B., T. Laubach, F. Mishkin and A.S. Posen (1999), Infation Targeting:
Lessons from the International Experience, Princeton, NJ: Princeton University
Press.
Boddy, R. and J. Crotty (1975), Class confict and macro-policy: the political busi-
ness cycle, Review of Radical Political Economics, 7, 119.
Bowles, S. (1985), The production process in a competitive economy: Walrasian,
non-Hobbesian, and Marxian models, American Economic Review, 75 (1),
1636.
Carchedi, G. (1977), The Economic Identifcation of Social Class, London:
Routledge and Kegan Paul.
Cardoso, Eliana (1992), Infation and Poverty, National Bureau for Economic
Research working paper no. 6793, March, accessed at http://ssrn.com/
abstract5293237.
Cukierman, A. (1992), Central Bank Strategy, Credibility, and Independence,
Cambridge, MA: MIT Press.
Di Tella, R., R. MacCulloch and A. Oswald (2001), Preferences over infation
and unemployment: evidence from surveys of happiness, American Economic
Review, 91, 33541.
Easterly, W. and S. Fischer (2001), Infation and the poor, Journal of Money,
Credit, and Banking, 33 (2), 16078.
Erikson, R. and J.H. Goldthorpe (1993), The Constant Flux, Oxford: Oxford
University Press.
Esping Andersen, G. (ed.) (1992), Changing Classes: Stratifcation and Mobility in
Post-Industrial Societies, London: Sage.
Hibbs, D. (1977), Political parties and macroeconomic policy, American Political
Science Review, 71, 146787.
Iversen, T. and D. Soskice (2001), An asset theory of social preferences, American
Political Science Review, 95 (4) (December), accessed at www.people.fas.
harvard.edu/~iversen/data/ISCO_conversion_tables.htm.
Iversen, T. and D. Soskice (2006), New macroeconomics and political science,
Annual Review of Political Science, 9, 42553.
Jayadev, A. (2006), Difering preferences between anti-infation and anti-
unemployment policy among the rich and the poor, Economics Letters, 91,
6771.
Jayadev, A. (2008), The class content of preferences towards anti-infation and
anti-unemployment policies, International Review of Applied Economics, 22 (2),
16172.
Johnson, H. (1968), Problems of ef ciency in monetary management, Journal of
Political Economy, 76, 986.
Leiulfsrud, H., Bison, I. and H. Jensberg (2005), Social class in Europe, European
Social Survey 2002/3, NTNU Social Research Ltd., unpublished, accessed
http://ess.nsd.uib.no/fles/2003/ESS1SocialClassReport.pdf.
Pollin, R. (1998), The natural rate of unemployment: its all about class confict,
Dollars and Cents Magazine, (September/October).
Romer, Christina D. and David H. Romer (1998), Monetary policy and the well-
being of the poor, National Bureau for Economic Research working paper no.
6793.
Rosenberg, S. and T. Weisskopf (1981), A confict theory approach to infation in
the postwar US economy, American Economic Review Papers and Proceedings,
71 (2), 427.
Rowthorn, R.E. (1977), Confict, infation and money, Cambridge Journal of
Economics, 1 (3), 21539.
Scheve, K. (2003), Public demand for low infation, Bank of England working
paper no. 172, London.
Scheve, K. (2004), Public infation aversion and the political economy of macro-
economic policymaking, International Organization, 58 (1), 134.
Shiller, R. (1997), Why do people dislike infation? in C. Romer and D. Romer
(eds), Reducing Infation: Motivation and Strategy, Chicago, IL: University of
Chicago Press, pp. 1371.
Wright, E.O. (1985), Classes, London: New Left Books.
Wright, E.O. (1997), Class Counts: Comparative Studies in Class Analyses,
Cambridge: Cambridge University Press.
APPENDIX 4.A1 CLASS MEASURES
The ISSP dataset provides ISCO-88 classifcations for all but four of the
countries. For each of these countries we have ISCO-68 classifcations.
Iversen and Soskice (2001) provide a bridge between these coding mecha-
nisms based on previous work by Ganzebloom. Using this code (available
from Iversens webpage at the Harvard School of Government), I recode
all respondents as per ISCO-88 codes. I drop the 300 or so observations
for which there is no bridge available.
Wright (1997) provides a mechanism by which to classify ISCO-88
classifcations into three skill categories, experts, skilled and low skilled
workers. Using the codes provided by Leilsfrud et al. (2005), I replicate
these categorizations for the ISSP dataset.
The code is available upon request.
Table 4.X Summary statistics
Variable Source Obs Mean Std
dev.
Min. Max.
Infation down V63: ISSP 23 824 0.41 0.49 0 1
Expert Constructed from
V202: ISSP
35 313 0.07 0.25 0 1
Semi-skilled Constructed from
V202: ISSP
35 313 0.23 0.42 0 1
Unskilled Constructed from
V202: ISSP
35 313 0.24 0.43 0 1
Subjective
lower class V221: ISSP 35 313 0.31 0.46 0 1
Subjective
middle class V221: ISSP 35 313 0.49 0.50 0 1
Subjective
upper class V221: ISSP 35 313 0.07 0.26 0 1
Firm owner Constructed from
V213 and V214:
ISSP dataset
35 313 0.04 0.19 0 1
First income
quintile
Constructed from
V217: ISSP
35 313 0.15 0.36 0 1
Second income
quintile
Constructed from
V217: ISSP
35 313 0.11 0.31 0 1
Third income
quintile
Constructed from
V217: ISSP
35 313 0.13 0.34 0 1
Fourth income
quintile
Constructed from
V217: ISSP
35 313 0.11 0.31 0 1
Table 4.X (continued)
Variable Source Obs Mean Std
dev.
Min. Max.
Fifth income
quintile
Constructed from
V217: ISSP
35 313 0.10 0.30 0 1
Union member V222: ISSP 35 313 0.18 0.38 0 1
Female V200: ISSP 35 228 0.52 0.50 0 1
Unemployed V206: ISSP 35 313 0.06 0.23 0 1
Age V201: ISSP 35 109 44.82 16.65 15 97
Pro wage control Constructed from
V17: ISSP
33 798 0.33 0.47 0 1
Anti jobs for all Constructed from
V36: ISSP
33 820 0.27 0.44 0 1
Anti government
jobs
Constructed from
V20: ISSP
34 188 0.06 0.24 0 1
Pro deregulation Constructed from
V21: ISSP
28 350 0.92 0.28 0 1
Anti protect jobs Constructed from
V23: ISSP
34 035 0.21 0.40 0 1
Power of labor Constructed from
V33: ISSP
30 614 0.65 0.48 0 1
Power of business Constructed from
V34: ISSP
29 484 0.84 0.37 0 1
Anti
redistribution
Constructed from
V16: ISSP
33 542 0.24 0.43 0 1
Anti equalization Constructed from
V42: ISSP
33 096 0.30 0.46 0 1
Lower social
services
Constructed from
V56: ISSP
26 652 0.52 0.50 0 1
93
5. The gendered political economy
of infation targeting: assessing its
impacts on employment
1
Elissa Braunstein and James Heintz
5.1 INTRODUCTION
Central banks in developing countries are increasingly maintaining low
infation rates as the central goal of monetary policy, without much con-
sideration of how these policies impact real outcomes like employment,
investment and growth (Epstein, 2003). Although targeting very low
infation rates seems to have done little to raise economic growth, these
policies remain a key feature of neoliberal approaches to monetary policy
(Epstein, 2000). Gerald Epstein and Juliet Schor argue that anti-infation
policy and neoliberal approaches to central banking emerged from the
contested terrain of central banks the class and intra-class conficts
over the distribution of income and power in the macroeconomy (Epstein,
2000; Epstein and Schor, 1990). Their work underscores the importance of
understanding monetary policy from a political economy perspective, as
the distribution of the gains and costs of economic policy profers insight
into both a policys genesis and its longer term consequences.
In this chapter we build on their analysis by considering the employ-
ment costs of infation reduction in developing countries from a gender
perspective. We explore two empirical questions: (1) what is the impact,
if any, of infation reduction on employment, and is the impact diferent
for women and men, and (2) how are monetary policy indicators con-
nected to defationary episodes and gender-specifc employment efects?
We fnd a common pattern among countries undergoing what we term
contractionary infation reduction, or periods of declining infation that
are accom panied by a loss of employment. After controlling for long-term
employment trends, we fnd that the ratio of womens to mens employ-
ment tends to decline during these periods in the majority of countries
examined. During periods of expansionary infation reduction, however,
there are no clear patterns to the relative changes in womens and mens
employment. In terms of monetary policy, we fnd that countries that
respond to infation by raising real interest rates or tightening the real
money supply (both relative to their long-run trends) are also more likely
to experience employment contractions, with concomitantly higher costs
for womens employment. Conversely, those that maintain competitive
real exchange rates are likely to reverse the negative impact of contraction-
ary infation reduction on womens relative employment.
That the costs of infation reduction, at least in terms of employment,
are inequitably distributed means that the contested terrain of monetary
policy is also gender specifc, with the result that the costs of implementing
these sorts of policies are actually quite diferent and potentially higher
than is generally presumed. After a discussion of the literature on the
diferences in womens and mens unemployment in developing countries,
and presenting our empirical results, we will develop this last point in more
detail in the closing section.
5.2 GENDER DIFFERENCES IN EMPLOYMENT
AND UNEMPLOYMENT IN SEMI-
INDUSTRIALIZED COUNTRIES
In thinking about the diferences between womens and mens unemploy-
ment, it is helpful to think in terms of supply-side factors and demand-side
factors (Seguino, 2003). On the supply side, diferences in human capital
are probably the most commonly considered. However, gender-based
diferences in education, skill and experience are themselves rooted in
workers productive roles outside the factory door and the institutional,
social and material contexts in which they live. A useful way of consider-
ing these diferences is through what Folbre (1994) terms structures of
constraint the preferences, norms, assets and rules that shape individual
choice.
Women and men make decisions about whether or not to look for wage
work. However, self-perception, what individuals value, and what choices
they perceive as possible are constituted by the social world, and are dif-
ferent for women and men (Sen, 1990), and so individual preferences
must be understood in this light. Norms are the traditional structures of
gender and kinship that constitute the social expectations of women and
men in the household. For example, women are primarily associated with
the care of the family, and much of their work time is spent outside of the
market, whereas mens work is typically viewed as directly productive
and more fully incorporated into the market sphere. Assets also engender
labor supply. Systematic diferences by gender in ownership of assets are
The gendered political economy of infation targeting 95
common, and partly determine how much wage employment women seek.
Similarly, laws that confer patriarchal property rights, where eldest men
have the right to claim and apportion the fruits of household members
labor time, can create incentives for high fertility and lower female labor
force participation (Braunstein and Folbre, 2001). Conversely, not having
a legal claim on a spouses income in the event of separation means that a
paying job can be an insurance policy against loss of that support (Folbre,
1997).
On the demand side, gender discrimination results in diferential access
to employment opportunities. The presumption that men should bear the
primary fnancial responsibility for provisioning families has been linked
with higher unemployment for women relative to men in OECD countries
(Algan and Cahuc, 2004). Women are laid of frst because employers
presume that it is more important for men to be able to fulfll their tra-
ditional breadwinning responsibilities (Azmat et al., 2004). Diferential
hiring practices also contribute to gender segregation in employment. For
example, labor-intensive exporters prefer to hire women both because
womens wages are typically lower than mens, and because employ-
ers consider women to be more productive in these types of jobs due to
employer perceptions that women possess nimble fngers, are less prone
to worker unrest, are better suited to tedious work and are more reliable
workers than men (Anker and Hein, 1985; Elson, 1996; Elson and Pearson,
1981; Fernndez-Kelly, 1983).
2
By extension, women may lose their com-
parative advantage when industries upgrade, leading to a defeminization
of employment as has happened in Mexico, India, Ireland and Singapore
(Elson, 1996; Fussell, 2000; Ghosh, 2001; Joekes, 1999).
As a result of these gender diferences in labor demand and supply,
changes in macroeconomic structure and policy have diferential efects
on mens and womens work (Seguino, 2003). For example, feminists have
long argued that the interaction between gender relations and structural
adjustment programs (SAPs) have implications both for the distribution
of costs and benefts between diferent groups of women and men, and
for the achievement of the economic objectives of the SAPs themselves
(Bakker, 1994; Benera and Feldman, 1992; Benera and Roldan, 1987;
Cagatay et al., 1995; Elson, 1991, 1995).
Macroeconomic policies and variables have been shown to infuence
womens labor supply decisions. Cagatay and Ozler (1995) pool cross
country data for 1985 and 1990 to show that SAPs led to increased femin-
ization of the labor force via worsening income distribution and openness.
Others have also argued that, in the case of Latin America, macroeco-
nomic crises place pressures on household resources which increase labor
force participation and hours worked, particularly for women (Arriagada,
1994; Cerrutti, 2000). Similarly, research into the determinants of womens
labor supply in post-apartheid South Africa has shown that womens
labor force participation responds positively to growing unemployment,
thereby further increasing the countrys average unemployment rate
(Casale, 2003).
Trade liberalization and structural adjustment policies can cause a
reallocation of paid employment from traditional activities, which are
adversely afected, to export-oriented production, which is encouraged
(Cagatay and Ozler, 1995; Standing, 1999). In addition, female-intensive
export-oriented industries are often more cyclically volatile than mens
industries, resulting in higher overall rates of unemployment (Howes and
Singh, 1995). Emphasis on export-oriented industrialization has also been
associated with increases in informalization of certain types of employ-
ment due to increases in competitive pressures (Carr et al., 2000; Standing,
1999). As female labor force participation rises in the context of crisis and
structural adjustment, the increasing dominance of informal work has
become a key new reality for women (Arriagada, 1994; Benera, 2001;
Patnaik, 2003).
Similar work has been done on the gendered employment efects of the
Asian fnancial crisis in 199798 (Aslanbeigui and Summerfeld, 2000;
Lim, 2000; United Nations, 1999). Women were typically the frst to be
laid of both because they worked in more cyclically volatile frms, such
as small export-oriented enterprises, and because of eforts to protect the
jobs of male breadwinners. The lack of formal employment opportuni-
ties and pressures for greater labor force participation led to an increase in
womens informal employment in some cases. A slightly diferent pattern
was found by Lim (2000) in the Philippines, where the post-crisis decline
increased male unemployment more than female unemployment despite a
rapid displacement of women from the manufacturing sector (especially
in traded goods). The reason was the relative resilience of the service and
trade sectors, which are more female intensive. Women did, however,
increase their labor force participation to deal with male unemployment,
and their total work hours relative to men increased as well.
These gender dynamics have implications for macroeconomic outcomes.
For example, increases in womens employment that are associated with
lower unit labor costs have important implications for infation dynamics.
For the non-tradable and import-competing sectors in which some degree
of competition exists, lower unit labor costs associated with growth in
womens employment reduce infationary pressures in the countries where
the women work.
Clearly, there are signifcant structural diferences between womens
and mens labor markets on both the supply and demand sides that are
diferently afected by macroeconomic structure and policy. The literature
reviewed above on semi-industrialized countries suggests that economic
contractions have a larger negative efect on womens formal employment
than mens, though women tend to increase their labor force participation
at the same time to protect household income. In the next section we turn
to a more direct test of this employment hypothesis, looking at whether
the process of infation reduction in particular can be associated with
gender diferentials in employment.
5.3 INFLATION REDUCTION AND WOMENS
EMPLOYMENT
The empirical exercise we present here explores the efects of infation
reduction on womens and mens formal employment. We compiled data
for 51 infation reduction episodes in 17 low- and middle-income coun-
tries.
3
To assess the employment efects of infation reduction periods,
we looked at actual employment trends during each infation reduc-
tion episode, disaggregated by gender, and compared these to long-run
employment trends, estimated by applying a Hodrick-Prescott flter to the
employment series. We also examine indicators that suggest how mon-
etary policy responded during infation reduction episodes using a similar
approach. We compare average short-term real interest rates, growth
rates of the real money supply, and indicators of the real exchange rate to
their long-run trends to see if these variables deviated from trend during
infation reduction episodes. We drew from three diferent data sources in
conducting this exercise: employment data came from the International
Labor Organizations (ILO) LABORSTA online database, and the macro-
economic data was compiled from World Development Indicators 2005
(Washington, DC: World Bank) and International Financial Statistics
(Washington, DC: IMF, October 2005).
4
The methodology used is drawn from the literature on measuring sac-
rifce ratios that is, the loss of output or employment associated with a
given reduction in infation. Sacrifce ratios can be measured simply as the
slope of the Phillips Curve showing an employment-infation trade-of.
However, this approach assumes that the sacrifce ratio remains constant
over long periods of time. Other approaches examine changes in employ-
ment or output over specifc defationary periods. This technique allows
the sacrifce ratio to change over time and makes it possible to analyse the
behavior of economic variables during specifc defationary periods. This
is the approach followed here.
Ball (1993) suggests one method for identifying defationary periods.
Infation fgures are notoriously volatile, making the identifcation of
turning points dif cult. A moving average of infation in our case, a three-
year moving average, encompassing one previous year and one subsequent
year can be used to smooth the series. Peaks and troughs in the smoothed
infation series are identifed. Peaks occur when the value in a particular
year exceeds the values of immediately adjacent years. Troughs occur
when the value in a given year falls below the values of the adjacent years.
A defationary period runs from a peak year to the next trough year.
For the purposes of this exercise, we use the term infation reduction
episode to refer to these defationary periods. The reason for this is that,
during some of the periods identifed, employment actually expands more
rapidly than its long-run trend. It seems confusing to refer to these periods
as defationary. Therefore, we use the terms expansionary infation
reduction episode and contractionary infation reduction episode.
The infation reduction episodes identifed for the 17 countries exam-
ined are remarkably similar in terms of the timing of the episodes. That
is, infation reduction episodes occur in many of these countries simulta-
neously. Figure 5.1 illustrates this pattern. The fgure shows, for each year
0
10
20
30
40
50
60
70
80
90
100
1
9
7
1
1
9
7
3
1
9
7
5
1
9
7
7
1
9
7
9
1
9
8
1
1
9
8
3
1
9
8
5
1
9
8
7
1
9
8
9
1
9
9
1
1
9
9
3
1
9
9
5
1
9
9
7
1
9
9
9
2
0
0
1
P
e
r
c
e
n
t

o
f

s
a
m
p
l
e

c
o
u
n
t
r
i
e
s

u
n
d
e
r
g
o
i
n
g

d
i
s
i
n
f
l
a
t
i
o
n
0
2
4
6
8
10
12
14
I
n
t
e
r
e
s
t

r
a
t
e

(
p
e
r
c
e
n
t
)
Interest
rate
Figure 5.1 Infation reduction episode frequencies and the real US interest
rate, 19712003
19702003, the percent of all countries in our study in which an infation
reduction episode occurred. Infation reduction episodes are concentrated
over certain sub-periods: 197477, 198086, 199194 and 19972000. Also
Figure 5.1 clearly demonstrates that infation reduction episodes were
much more common throughout the 1990s than throughout the prior two
decades, the 1970s and 1980s.
This pattern of infation reduction suggests that one or more common
factors determine infation rates and monetary variables in the diferent
countries examined. One obvious possibility is the world interest rate.
Figure 5.1 shows the real yield on US Treasury bills over the entire period
as a proxy for short-term, infation-adjusted world interest rates. An
increase in the real yield of US Treasury bills preceded three out of the four
common infation reduction episodes. Only in the most recent episode,
from approximately 1997 to 2000, did interest rates fail to increase before-
hand.
5
Because the data for the diferent countries show similar trends in
the variables analysed, we emphasize deviations from long-run trends,
rather than absolute levels or changes, in order to investigate common
patterns and divergent trends.
We restricted the countries used as follows:
Only low- or middle-income countries were examined.
6
Countries must have at least 20 years of gender-disaggregated
employment data (it would be hard to estimate a meaningful long-
run trend with a shorter series).
Time series with missing values were used. However, time series with
two consecutive missing observations were rejected. Missing values
were estimated for the purposes of computing a long-run trend by
extrapolating between the previous and subsequent values in the
series.
Changes in employment across infation reduction episodes were calcu-
lated as the annualized value of the overall rate of change in employment
across the entire peak-to-trough period. As mentioned earlier, values for
the long-run employment trends were computed by applying a Hodrick-
Prescott flter to the actual employment time series for each country
(men, women and total). The employment time series used most likely
underestimates the magnitude of informal employment in these countries.
Therefore, the results of this analysis should be interpreted bearing this in
mind.
Table 5.1 summarizes the results for all the infation reduction episodes
studied. The table shows the country name, the dates of each infation
reduction episode, and the deviation from the long-run trend for womens
Table 5.1 Infation reduction episodes and deviations from long-run
employment trends, disaggregated by gender
A. Contractionary infation reduction episodes
period Defation
as percent
of avg.
infation
Deviations from long-run
employment trends (percent)
W M Ratio
Barbados 198086 143 1.8 0.8 1.0
199094 113 2.5 2.6 0.1****
199699 117 1.0 0.3 1.4
Brazil 199399 305 0.7 0.1 0.6
Colombia 198085 29 3.2 2.5 0.7
Costa Rica 198285 123 1.5 0.1 1.4
India 197377 175 0.2 0.4 0.2****
198286 37 0.1 0.2 0.1****
199194 26 0.1 1.2 0.2****
199702 93 1.0 0.9 0.2
Jamaica 197476 45 0.5 0.2 0.3
19922000 190 0.5 0.1 0.6
Kenya 197580 36 2.2 0.1 2.1
198187 66 0.8 0.3 1.1****
Malaysia 198186 181 0.4 0.8 0.4****
Mauritius 198086 182 0.6 1.6 0.9****
198993 46 1.3 0.3 0.9
199496 21 1.8 0.9 0.8
Philippines 197376 69 1.6 0.4 1.2
198082 55 0.2 0.3 0.5****
198487 130 2.4 0.0 2.3
Singapore 197476 171 6.7 0.7 5.9
198186 267 1.8 2.0 0.1****
South Korea 198085 196 1.4 0.9 0.5
199194 45 0.4 0.0 0.4
19972000 82 1.2 1.1 0.1
Sri Lanka 198186 106 0.7 0.1 0.8
199799 56 0.7 2.6 1.9****
Taiwan 197476 106 4.9 0.3 5.1
198085 243 0.6 0.5 1.1****
199102 197 0.4 0.2 0.2****
Thailand 197476 88 1.3 0.8 0.5
198085 180 2.6 0.7 1.8
199093 31 0.8 0.1 0.9
199700 136 0.8 0.7 0.1
Trinidad and
Tobago 198087 56 1.0 0.6 0.4
employment, mens employment, and the female to male employment
ratio. Negative values indicate that the series grew more slowly than the
long-run trend (a negative value could also indicate a more rapid decrease
in the actual value compared to the long-run trend). Table 5.1 is divided
into contractionary and expansionary infation reduction episodes. During
contractionary infation reduction episodes the rate of increase of total
employment fell below its long-run trend. During expansionary infation
reduction episodes, the rate of increase of total employment was equal to
or greater than its long-run trend.
In 67 percent of contractionary infation reduction episodes the rate
of change of the female to male employment ratio fell below its long-
run trend, indicating that womens employment was disproportionately
afected by the slowdown. If India were excluded from the sample, this
proportion would rise to over 72 percent.
7
However, in expansionary
Brazil 198992 39 1.9 0.8 2.8****
Chile 198488 39 0.8 2.3 1.5
Costa Rica 199193 42 0.2 1.6 1.4
Jamaica 197982 101 0.1 0.5 0.4
198588 90 2.9 0.5 2.3****
Kenya 199396 134 0.9 0.3 1.3****
Malaysia 199296 27 0.6 1.4 0.8
Mauritius 197477 59 3.9 1.6 2.1****
Philippines 199094 60 0.2 0.3 0.1
Singapore 199099 139 0.1 0.1 0.0
Sri Lanka 197476 100 1.9 3.1 1.1
198994 50 8.9 3.6 4.9****
Trinidad and
Tobago
197477 48 0.2 1.0 0.8
198992 37 3.8 1.6 2.1****
199396 69 0.7 0.6 0.2****
Note: **** Infation reduction episodes in which the ratio of womens to mens employment
increased more rapidly than the long-run trend.
Summary:
Contractionary infation reduction episodes with declining female:male
employment ratios relative to the trend: 67 percent.
Contractionary infation reduction episodes with declining female:male employ
ratios relative to the trend (excluding India): 72 percent.
Expansionary infation reduction episodes with increasing female:male
employment ratios relative to the trend: 53 percent.
infation reduction episodes there was no clear distinction. The female to
male employment ratio increased faster than trend in 53 percent of cases
and at or below trend in 47 percent of cases nearly an even split.
The contractionary infation reduction episodes were associated with
a larger decline in infation from peak to trough relative to the average
infation rate for the entire episode. Averaging across all contractionary
episodes, the decline in infation was 115 percent of the average infation
rate during the episode in question. For expansionary episodes, the decline
was 69 percent.
The diference in employment experiences across countries during
infation reduction episodes for example, expansion or contraction of
employment might be explained, in part, by policy choices. For example,
if real interest rates rose above the long-run trend in reaction to an accel-
eration of infation, this could trigger a contractionary infation reduction
episode. However, if real interest rates were not raised above the long-run
trend (for example, they were kept in line with the long-run trends of
global interest rates), a contraction of employment might be avoided.
To examine this possibility, we looked at average real short-term inter-
est rates across the infation reduction episodes.
8
In most cases short-term
rates linked directly to monetary policy choices were used (for example,
a discount rate or bank rate). If these rates were unavailable, yields on
short-term (three-month) Treasury bills were calculated instead.
9
If actual
average real interest rates were negative, these infation reduction episodes
were tabulated separately.
Table 5.2 shows patterns in short-term interest rates over expansion-
ary infation reduction episodes. Average actual real interest rates are
compared to the average interest rates associated with the long-run trend,
calculated by applying a Hodrick-Prescott flter to the real interest rate
series. The diference in average real interest rates (actual rates minus the
rates associated with the long-run trend) is expressed as a percentage of
the average long-run trend in real interest rates over the infation reduction
episode. Only infation reduction episodes with positive actual average
real interest rates over the episode in question are included in Table 5.2.
In almost all of the infation reduction episodes, actual real interest rates
were, on average, kept below the long-run trend. This is consistent with
the argument that countries that do not raise interest rates in the face of
growing infationary pressures are less likely to experience employment
losses, or a slow-down in the growth rate of employment, during infation
reduction episodes. There were only two exceptions: Sri Lanka 197476
and Trinidad and Tobago 198992.
One problem with looking at interest rates over the entire infation
reduction episode is that policy makers might increase interest rates in
anticipation of infation or in response to domestic economic conditions
prior to the year in which infation peaks. Therefore, we also examined
average interest rates over an alternative periodization in order to see if
the timing of the interest rate changes infuenced our interpretation of
the nature of the infation reduction event. We looked at the period one
year before the beginning of an infation reduction episode and extending
half-way into the episode. For example, if the infation reduction episode
spanned the years 1980 to 1984, then the alternative period over which
interest rates were compared would be 1979 to 1982.
This alternative periodization for the real interest rate analysis shows
that real interest rates were negative on average in Sri Lanka from 1973
to 1975. Therefore, the increase in real interest rates documented in Table
5.2 might represent an efort by policy makers to move from negative real
interest rates to positive (yet still relatively low) rates. Raising negative
interest rates (that is, making them more positive) could have a diferent
impact on employment than raising positive real interest rates a point
to which we will return shortly. However, the alternative periodization
does not shed much light on the case of Trinidad and Tobago from 1989
to 1992.
Table 5.3 shows a similar set of calculations for another sub-set of
Table 5.2 Patterns in average real short-term interest rates over infation
reduction episodes in which total employment expanded (only
episodes with positive average real interest rates included).
Period Diference in actual
and long-run average
real interest rates as
a percent of the
long-run average
Notes
Chile 198488 31.3
Costa Rica 199193 9.8
Kenya 199396 2.1
Malaysia 199296 9.7
Philippines 199094 42.8
Singapore 199099 6.9
Sri Lanka 197476 124.0 197375: avg. neg.
interest
198994 5.6
Trinidad and
Tobago
198992 16.3 198891: 15.4 per
cent
199396 3.0
infation reduction episodes. This time, contractionary infation reduction
episodes are featured. As with Table 5.2, only infation reduction episodes
with positive actual average real interest rates over the episode in question
are included.
In most cases the opposite pattern observed previously in Table 5.2 is
reduction episodes in which total employment contracted (only
episodes with positive average real interest rates included)
Period Diference in actual
and long-run average
real interest rates as
a percent of the
long-run average
Notes
Barbados 198086 136.7
199094 14.4
199699 3.3 199598: 11.0
percent
Colombia 198085 1.1 197983: 134.6
percent
India 198286 114.2
199194 9.3 199093: 19.3
percent
199702 18.4 19962000: 12.8
percent
Jamaica 19922000 8.4 Interest rates
are t-bill rates.
199196: avg. neg.
interest
Malaysia 198186 151.9
Mauritius 198993 79.2 198891: 76.3
percent
199496 116.4
Singapore 198186 117.9
Sri Lanka 198186 1398
199799 12.9
Taiwan 198085 124.3
199102 6.5 199097: 11.9
percent
Thailand 198085 16.6
199093 15.2
19972000 114.2
evident. During contractionary infation reduction episodes actual inter-
est rates were kept above the long run trend on average. In two cases,
Barbados 199699 and Colombia 198085, actual real interest rates fell
below the long-run trend during the entire infation reduction episode.
However, if we apply the alternative periodization we can see that real
interest rates increased before infation peaked and during the earlier
stages of the infation reduction period. For example, in Colombia from
1979 to 1983 average actual real interest rates were nearly 35 percent
higher than the long-run average over this same period.
There were fve contractionary infation reduction episodes in which
real interest rates behaved diferently: India 199194, India 19972002,
Jamaica 19922000, Mauritius 198993 and Taiwan 19912002. In one
case Jamaica the alternative periodization shows that interest rates
were negative on average going into the infation reduction episode. This
alters the impact on employment of keeping rates below their long-run
trend. However, in the other four cases interest rates and employment
exhibited a diferent pattern when compared to the other episodes in Table
5.3.
This discussion suggests that negative real interest rates are important
in analysing the employment trends across infation reduction episodes.
Table 5.4 summarizes trends in real interest rates for those infation reduc-
tion episodes in which average actual real interest rates were negative.
In the majority of the cases in which average real interest rates were
negative, interest rates were kept below the long-run average and employ-
ment grew slower than the trend rate of growth. This suggests that
keeping interest rates negative and below the long-run trend will not help
to increase employment. In two cases employment expanded despite low,
negative real interest rates: Jamaica 197982 and Trinidad and Tobago
197477. However, these appear to be the exception rather than the rule.
One reason why negative real interest rates tend to be associated with a
contraction in employment relative to its long-run trend is that many of
these infation reduction episodes are stagfationary. That is, supply-
side shocks produce a situation in which infation accelerates, economic
growth slows down, resulting in negative real interest rates.
Real interest rates are only one set of variables that potentially link
monetary policy to infation reduction and employment dynamics. Two
other possibilities are the real exchange rate and the growth rate of the
real money supply. Table 5.5 examines trends in these two variables
across the various infation reduction episodes. Specifcally, it shows dif-
ferences between the actual average annual growth rate and the growth
rate associated with the long-run trend for the real exchange rate and
the real money supply across various infation reduction episodes. The
infation reduction episodes are separated into contractionary and expan-
sionary periods.
In terms of the real exchange rate, for the purposes of this study, we
measured the rate as the nominal US dollar exchange rate adjusted for
changes in the US GDP defator relative to the specifc countrys GDP
defator. A decrease in the value of our real exchange rate measurement,
therefore, represents an appreciation in the real exchange rate. Likewise,
an increase in value represents a depreciation.
There do not appear to be any systematic patterns with respect to changes
in the real exchange rate across infation reduction episodes and whether
the episode was contractionary or expansionary. In 34 percent of infation
reduction episodes the average annual percent change in the real exchange
rate was below that of the long-run exchange rate (that is, the exchange
rate appreciated relative to its long-run trend); in 60 percent of the episodes
the diference in average growth rate was positive (that is, the actual real
exchange rate depreciated relative to the long-run trend); and in 6 percent
reduction episodes in which average real interest rates were
negative
Period Actual interest
rates above or
below
long-run trend
Employment:
expansionary or
contractionary
Costa Rica 198285 below contract
India 197377 below contract
Jamaica 197476 below contract
197982 below expand
198588 above expand
Kenya 197580 below contract
Mauritius 198086 below contract
Philippines 197376 below contract
198082 below contract
Singapore 197476 below contract
South Korea 198085 below contract
19972000 above contract
Taiwan 197476 below contract
Trinidad and Tobago 197477 below expand
Table 5.5 Diferences in the average actual annual growth rate and the
average annual growth rate associated with the long-run trend
for the real exchange rate (RER) and the real money supply
across infation reduction episodes
Country Period RER (percent) Money supply
(percent)
Total employment contracts on average
Barbados 198086 0.3 1.1
199094 11.1 1.5
199699 1.2 13.0
Brazil 199399 11.1 10.3
Colombia 198085 12.2 10.2
Costa Rica 198285 6.7 9.4
India 197377 11.6 1.6
198286 10.8 10.6
199194 13.6 0.4
19972002 10.4 11.7
Jamaica 197476 9.0 0.7
19922000 0.6 10.5
Kenya 197580 3.5 11.9
198187 11.7 11.8
Malaysia 198186 14.5 1.4
Mauritius 198086 0.0
*
2.2
198993 11.8 1.9
199496 4.0 1.4
Philippines 197376 3.9 1.6
198082 10.8 0.4
198487 0.0 10.2
Singapore 197476 13.9 0.6
198186 12.8 2.4
South Korea 198085 17.0 1.9
199194 0.4 2.7
19972000 16.7 16.2
Sri Lanka 198186 0.7 1.1
199799 11.7 0.0
Thailand 197476 0.2 0.3
198085 13.7 2.5
199093 1.9 12.8
19972000 17.0 0.4
Trinidad and Tobago 198087 12.7 1.0
Total employment expands on average
Brazil 198992 0.0 14.9
Chile 198488 12.2 12.7
of the episodes there was no diference between the growth rate of the actual
and long-run real exchange rates. These ratios were approximately the same
for contractionary and expansionary infation reduction episodes.
10
However, real exchange rates appear to have an impact on the gender
bias observed in contractionary infation reduction episodes. Recall that,
in the majority of cases, womens formal employment was disproportion-
ately afected by the slow-down in employment growth. However, about
a third of the time the ratio of womens to mens employment actually
improved when compared to its long-run trajectory. In each of these cases
the real exchange rate either depreciated or showed no deviation relative
to its long-run trend. In other words, maintaining a competitive exchange
rate may ofset some of the gender bias observed during contractionary
infation reduction. Why would this be the case? As previously noted, in
many countries the growth of womens employment particularly formal
employment and wage employment has tended to be concentrated in
tradable sectors, either export-oriented or import-competing (Benera,
2003; Elson, 1996; Elson and Pearson, 1981; Kabeer and Mahmud, 2004).
A real depreciation of the exchange rate favors tradable sectors and could
help protect womens employment in certain cases.
Turning to the real money supply,
11
there is some indication that money
Country Period RER (percent) Money supply
(percent)
Costa Rica 199193 4.8 1.1
Jamaica 197982 2.2 11.9
198588 7.8 14.9
Kenya 199396 15.7 111.4
Malaysia 199296 4.6 15.1
Mauritius 197477 n.a. 15.4
Philippines 199094 1.8 0.5
Singapore 199099 10.4 11.4
Sri Lanka 197476 1.0 4.8
198994 0.3 0.7
Trinidad and Tobago 197477 1.0 1.0
198992 3.8 2.7
199396 11.4 15.4
Note:
*
Based on 198186 average, due to data limitations.
Source: Authors calculations based on data from the IMF publication International
Financial Statistics. Real US interest rate is T-bill rate less the infation rate.
supplies grew more slowly during contractionary episodes. In 67 percent
of all contractionary episodes for which data is available actual average
annual growth rates in the real money supply fell below the long-run
trend. In 60 percent of all expansionary episodes actual annual growth
rates in the real money supply were greater than the average for the long-
run trend. Relative to the gendered efects of infation reduction then,
tightening the real money supply may have a tendency to be associated
with greater sacrifces in womens employment.
This analysis suggests a number of preliminary fndings. However, it is
important to recognize that a myriad of factors afect the variables exam-
ined here in particular, employment, infation, interest rates, exchange
rates and the money supply and therefore any generalizations must be
tentative. Although some commonalities do arise, the diversity of country
experiences prevents us from drawing defnitive conclusions without ad-
ditional in-depth analysis and country-specifc case studies.
Nevertheless, we can make some general observations from the analysis
presented.
Infation reduction episodes occur simultaneously across a large
number of countries. This suggests that there are common, external
factors that infuence infation dynamics in low- and middle-income
countries.
If employment contracts during an infation reduction episode, it
is likely that women will experience a larger loss of employment, in
percentage terms, than men. However, during infation reduction
episodes in which employment expands, the gender-specifc impact
is ambiguous.
Countries that respond to infationary pressures by raising real
interest rates above the long-run trend are more likely to experience
a slow-down in the growth of employment relative to those coun-
tries that keep interest rates in line with or below the long-run trend,
with concomitantly higher losses for relative female employment.
However, countries with negative real interest rates do not appear
to be able to increase employment growth by lowering real interest
rates still further.
We did not fnd a link between changes in the real exchange rate
and the impact of infation reduction on employment in general.
However, we did fnd that real exchange rates seem to impact the
gender bias of contractionary infation reduction episodes. In all
cases where women experienced relative employment gains during
employment contractions, exchange rates either depreciated or
showed no deviation relative to long-run trends.
Tightening the real money supply also seems to be negatively asso-
ciated with employment in general and womens employment in
particular.
These results suggest that contractionary monetary policy aimed at
reducing infation often has a disproportionately negative impact on
womens employment, an efect that may be eased by maintaining a com-
petitive exchange rate. Conversely, non-contractionary infation reduc-
tion is not necessarily favorable to womens formal employment in all
circumstances.
5.4 DISCUSSION AND CONCLUSIONS
The empirical analysis presented here concerns the short-run, gender-
specifc impacts of policy responses during infation reduction episodes.
The results say little about the long-run impact of diferent policy
responses. Supporters of infation targeting frequently acknowledge that
short-run trade-ofs might exist, but the long-run benefts of low infa-
tion for growth and development are more signifcant. This argument is
problematic when transitory policy shocks have long-run consequences
for real economic variables (Fontana and Palacio-Vera, 2004). Similarly,
short-term gender-specifc shocks can have long-run efects for a countrys
human and economic development.
A number of empirical studies suggest that gender-based inequities in
employment and unemployment have implications for long-term develop-
ment. For example, this body of research shows that a positive relationship
exists between gender equality (measured most commonly as educational
equity) and economic growth in developing countries (Dollar and Gatti,
1999; Hill and King, 1995; Klasen, 1999). Some of the efects are quite
large: Klasen (1999), in a panel data study between 1960 and 1992, fnds
that had South Asia and sub-Saharan Africa had more gender equity in
education, growth would have been 0.9 percent per year faster. Investing
in girls makes for a higher productivity workforce, but higher rates of
unemployment and cyclical volatility in womens jobs will discourage
these types of investments at both the individual and community levels.
In a related sense, lower incomes and higher income volatility for
women could lead to lower investments in human capital overall, thereby
lowering long-term growth. Theory and evidence have aptly demonstrated
a higher co-incidence between a mothers income and the familys basic
needs than a fathers income (Benera and Roldan, 1987; Blumberg, 1991;
Chant, 1991), a fnding underlying what has been termed the good mother
hypothesis. Income that is controlled by women is more likely to be spent
on childrens health and nutrition (Dwyer and Bruce, 1988; Hoddinott
et al., 1998). In many countries a large proportion of fathers provide
little or no economic support for their children (Folbre, 1994). But faced
with cyclically higher rates of unemployment during disinfation, good
mothers will have fewer opportunities to invest in their children.
In the nearer term systematic discrimination in hiring and fring results
in a misallocation of workers as the most able are not matched with the
most appropriate jobs. Such misallocations can result in ef ciency losses
with efects on output and growth. Tzannatos (1999) evaluates the efects
of eliminating gender-based occupational segregation in a number of
developing countries, and fnds on average a one-time gain in output of
6 percent. His results are signifcant but incomplete. He only considers
occupational segregation, not industrial segregation, which can be a much
more extensive part of gender segregation in some countries, as in China.
In addition, Tzannatos only considers the static picture, and not the
dynamic ef ciency losses from being on a particular growth path charac-
terized by a high degree of gender segregation (Elson, 1999).
Less instrumental and ultimately more central than the economic impli-
cations of gender-biased infation reduction discussed above is the issue
of equity, as we fnd that women as a group shoulder a disproportionate
share of the costs of contractionary infation reduction. From a social
justice perspective, then, it is important to better understand and redress
these gender-biased outcomes. One way of doing so would be to focus
research and policy on better understanding the link between infation
targeting and overall social welfare. To get at these connections, central
banks should incorporate gender-specifc indicators in the creation of
targets, such as gender disaggregated employment fgures or gender aware
infation rates (that account for gendered consumption and employment
patterns). These market-based targets should also be supplemented by
longer term gender aware human development targets to ultimately get at
the links among infation targeting, equity and well-being.
Treating gender equity as secondary to other, more general economic
concerns may also be instrumental in the political sustainability of infa-
tion targeting in monetary policy. As our analysis implies, incorporating
concerns over gender equity into monetary policy formulation would
involve a move away from infation targeting as it is currently practiced
and could harm the interests of those invested in a low infation, high
interest rate environment. Moreover, if womens labor force participa-
tion keeps unit labor costs and infation lower than it would otherwise be,
then a focus on gender equity within the context of sustainable levels of
infation could require other mechanisms for price control that are more
consistent with long-run development. Such a move might be resisted by
those that beneft (perhaps only in the short run) from womens more pre-
carious employment for example, their employers and employed men.
From this perspective, gender-biased central bank policy may help solve
the political problems introduced by infation targeting in that gender
bias concentrates the costs of these policies on a less powerful segment of
society women. As such, infation targeting policies should be considered
in terms of their social content (for example, what are the social structures
that underlie this policy) as well as its social impact (Elson and Cagatay,
2000). Such an approach underscores the gendered nature of the contested
terrain in macroeconomic policy making.
NOTES
1. This chapter is part of the PERI/Bilkent project on Alternatives to Infation Targeting.
We thank the Ford Foundation, the UN-DESA, the Rockefeller Brothers Fund and the
Political Economy Research Institute at the University of Massachusetts for fnancial
support. We would also like to thank Gerald Epstein, Erinc Yeldan and Diane Elson
for their help and comments, as well as the participants (and our hosts) of the AIT
meeting at CEDES in Buenos Aires in June 2005. All mistakes are our own, of course.
2. Similar reasoning can be applied to the pink collar aspects of the international pro-
duction of services, a sector where women predominate in the administrative support
aspect of the trade.
3. Our choice of countries was limited to those for which reasonably reliable, gender-
disaggregated formal employment data were available.
4. In addition, the GDP defator for the USA used to construct the real exchange rate
measure was taken from the online database of the US Bureau of Economic Analysis.
5. Other real US interest rates did increase around the time of the last infation reduction
episode, for example, the yield on three-year government bonds.
6. The sample of countries includes Singapore which could arguably be classifed as
a high-income country today. However, for much of the period considered in this
chapter, 19702003, Singapore can be considered a middle-income country.
7. India is notable in terms of its frequent deviation from the behavior exhibited by other
countries, both with respect to the gender dynamics of slower employment growth and
the observed trends in real interest rates during infation reduction episodes.
8. Due to its extreme volatility, Brazil was excluded from this analysis.
9. Only in the case of Jamaica were the treasury-bills used to determine actual interest
rates and to estimate long-run trends.
10. In contractionary episodes 36 percent of the episodes showed an appreciation, 58
percent a depreciation and 6 percent no diference in the average growth of the real
exchange rate relative to its long-run trend. In expansionary episodes the percentages
were 34 percent, 60 percent and 7 percent, respectively.
11. The defnition of the money supply used is M2, money plus quasi-money, which
includes cash; checking accounts; and relatively liquid deposits like savings and money
market deposit accounts.
REFERENCES
Algan, Y. and P. Cahuc (2004), Job protection: the macho hypothesis, Society for
Economic Dynamics 2004 meeting papers no. 332, September.
Anker, R. and C. Hein (1985), Why Third World urban employers usually prefer
men, International Labour Review, 124 (1), 7390.
Arriagada, I. (1994), Changes in the urban female labour market, CEPAL
Review, 53, 92110.
Aslanbeigui, N. and G. Summerfeld (2000), The Asian crisis, gender, and the
international fnancial architecture, Feminist Economics, 6 (3), 81103.
Azmat, Ghazala Maia Gvell and Alan Mangging (2004), Gender gaps in unem-
ployment rates in OECD countries, London School of Economics Centre for
Economic Policy discussion paper no. 604.
Bakker, I. (ed.) (1994), The Strategic Silence: Gender and Economic Policy,
London. Zed Books with the North-South Institute.
Ball, L. (1993), What determines the sacrifce ratio?, National Bureau for
Economic Research working paper no. 4306.
Benera, L. (2001), Shifting the risk: new employment patterns, informalization,
and womens work, International Journal of Politics, Culture and Society, 15
(1), 2753.
Benera, L. (2003), Gender, Development, and Globalization: Economics as if All
People Mattered, London: Routledge.
Benera, L. and S. Feldman (eds) (1992), Unequal Burden: Economic Crises,
Persistent Poverty, and Womens Work, Boulder, CO: Westview Press.
Benera, L. and M. Roldn (1987), The Crossroads of Class and Gender: Industrial
Homework, Subcontracting, and Household Dynamics in Mexico City, Chicago,
IL and London: University of Chicago Press.
Blumberg, R. (1991), Income under female versus male control, in R. Blumberg
(ed.), Gender, Family and Economy: The Triple Overlap, Newbury Park, CA:
Sage Press, pp. 97127.
Braunstein, E. and N. Folbre (2001), To honor and obey: ef ciency, inequality
and patriarchal property rights, Feminist Economics, 7 (1), 2544.
Cagatay, N. and S. Ozler (1995), Feminization of the labor force: the efects of
long-term development and structural adjustment, World Development, 23 (11),
188394.
Cagatay, N., D. Elson and C. Grown (eds) (1995), Gender, adjustment and
macro economics, World Development, 23 (11) (special issue), 145390.
Carr, M., M. Chen and J. Tate (2000), Globalization and home-based workers,
Feminist Economics, 6 (3), 12342.
Casale, D. (2003), The rise in female labour force participation in South Africa:
an analysis of household survey data, 19952001, PhD dissertation, Division of
Economics, University of KwaZulu-Natal.
Cerrutti, M. (2000), Economic reform, structural adjustment and female labor
force participation in Buenos Aires, Argentina, World Development, 28 (5),
87991.
Chant, S. (1991), Women and Survival in Mexican Cities: Perspectives on Gender,
Labour Markets and Low-Income Households, New York: Manchester University
Press.
Dollar, D. and R. Gatti (1999), Gender inequality, income and growth: are good
times good for women?, World Bank policy research report on gender and
development working paper series no. 1, Washington, DC.
Dwyer, D. and J. Bruce (1988), A Home Divided: Women and Income in the Third
World, Palo Alto, CA: Stanford University Press.
Elson, D. (ed.) (1991), Male Bias in the Development Process, Manchester:
Manchester University Press.
Elson, D. (1995), Gender awareness in modeling structural adjustment, World
Development, 23 (11), 185168.
Elson, D. (1996), Appraising recent developments in the world market for nimble
fngers, in A. Chhachhi and R. Pittin (eds), Confronting State, Capital and
Patriarchy: Women Organizing in the Process of Industrialization, New York: St
Martins Press, pp. 3555.
Elson, D. (1999), Labor markets as gendered institutions: equality, ef ciency and
empowerment issues, World Development, 27 (3), 61127.
Elson, D. and N. Cagatay (2000), The social content of macroeconomic policies,
World Development, 28 (7), 134764.
Elson, D. and R. Pearson (1981), Nimble fngers make cheap workers: an analy-
sis of womens employment in Third World export manufacturing, Feminist
Review, 7, 87107.
Epstein, G. (2000), Myth, mendacity and mischief in the theory and practice of
central banking, mimeo, University of Massachusetts Amherst.
Epstein, G. (2003), Alternative to infation targeting monetary policy for stable
and egalitarian growth: a brief research summary, Political Economy Research
Institute working paper no. 62.
Epstein, G. and J. Schor (1990), Macropolicy in the rise and fall of the golden age,
in S. Marglin and J. Schor (eds), The Golden Age of Capitalism: Reinterpreting
the Postwar Experience, Oxford: Clarendon Press, pp. 12652.
Fernndez-Kelly, M. (1983), For We are Sold, I and My People: Women and
Industrialization in Mexicos Frontier, Albany, NY: SUNY Press.
Folbre, N. (1994), Who Pays for the Kids? Gender and the Structures of Constraint,
London and New York: Routledge.
Folbre, N. (1997), Gender coalitions: extrafamily infuences on intrafamily
inequality, in L. Haddad, J. Hoddinott and H. Alderman (eds), Intrahousehold
Resource Allocation in Developing Countries. Models, Methods, and Policy
Baltimore, MD: Johns Hopkins University Press, pp. 26374.
Fontana, G. and A. Palacio-Vera (2007), Are long-run price stability and short-
run output stabilization all that monetary policy can aim for?, Metroeconomica,
58 (2), 26998.
Fussell, E. (2000), Making labor fexible: the recomposition of Tijuanas maqui-
ladora female labor force, Feminist Economics, 6 (3), 5979.
Ghosh, J. (2001), Globalisation, export-oriented employment for women and
social policy: a case study of India, paper prepared for United Nations
Research Institute for Social Development project on Globalization, Export-
Oriented Employment for Women and Social Policy, Geneva.
Hill, M. and E. King (1995), Womens education and economic well-being,
Feminist Economics, 1 (2), 2146.
Hoddinott, J., H. Alderman and L. Haddad (eds) (1998), Intrahousehold Resource
Allocation in Developing Countries: Methods, Models and Policy, Baltimore,
MD: Johns Hopkins University Press.
Joekes, S. (1999), A gender-analytical perspective on trade and sustainable
development, in Trade, Sustainable Development and Gender, New York and
Geneva: United Nations Conference on Trade and Development, pp. 3359.
Kabeer, N. and S. Mahmud (2004), Globalization, gender, and poverty:
Bangladeshi women workers in export and local markets, Journal of International
Development, 16, 93109.
Klasen, S. (1999), Does gender inequality reduce growth and development?
Evidence from cross-country regressions, World Bank policy research report on
gender and development working paper series no. 7, Washington, DC.
Lim, Joseph (2000), The efects of the East Asian crisis on the employment of
women and men: the Philippine case, World Development, 28 (7), 1285306.
Patnaik, U. (2003), Global capitalism, defation and agrarian crisis in developing
countries, United Nations Research Institute for Social Development, social
policy and development programme paper no. 15.
Seguino, S. (2003), Why are women in the Caribbean so much more likely than
men to be unemployed?, Social and Economic Studies, 52 (4), 83120.
Sen, A. (1990), Gender and cooperative conficts, in I. Tinker (ed.), Persistent
Inequalities: Women and World Development New York and Oxford: Oxford
University Press, pp 12349.
Standing, G. (1999), Global feminization through fexible labor: a theme re-
visited, World Development, 27 (3), 583602.
Tzannatos, Z. (1999), Women and labor market changes in the global economy:
growth helps, inequalities hurt and public policy matters, World Development,
27 (3), 55169.
United Nations (1999), World Survey on the Role of Women in Development.
Globalization, Gender and Work, New York: United Nations.
116
6. Infation and economic growth:
a cross-country non-linear analysis
Robert Pollin and Andong Zhu
1
6.1 INTRODUCTION
This chapter presents new cross-country evidence between 1961 and 2000
on the relationship between infation and economic growth. Despite the
central importance of this infation-growth relationship for macroeco-
nomic theory and policy, there is nothing close to a professional consensus
as to what the empirical evidence tells us about this relationship.
The results we present here have direct relevance to the debate on infa-
tion targeting as an appropriate framework for conducting monetary
policy. Over the past decade governments throughout the world have
embraced infation targeting as a dominant policy framework. For the
most part, this specifcally means that they have set a low band of accept-
able infation rates as a primary target in the conduct of economic policy.
This band is usually between a 35 percent annual infation rate. They
have then maintained suf ciently high short-term interest rates as the
intermediate policy instrument for preventing infation from exceeding
that target band. Higher interest rates are aimed, in turn, at reducing eco-
nomic growth. Slower economic growth should then dampen infationary
pressures. At least in the short run, the costs in terms of slower growth
of containing infation within this 35 percent band are evident. But pro-
ponents of infation targeting hold that, over a longer term framework,
maintaining low infation will itself yield benefts for growth that exceed
these short-term costs.
2
Some limitations of infation targeting have been widely recognized by
mainstream economists and even US central bankers Ben Bernanke and
Alan Blinder (see Bernanke et al., 1999; Blinder, 1998). The Bernanke/
Blinder view is that infation targeting does not provide an approach to
maintaining low infation that is clearly superior to other approaches. This
is true as such, but this concern about infation targeting as an operating
procedure alone begs a more important question. This crucial question is
whether maintaining infation within a band of 35 percent itself is, as a
Infation and economic growth: a cross-country analysis 117
generalization, supportive of economic growth, regardless of the technique
being used to maintain infation within that low band. It is this broader
question that we address in this chapter. That is, are countries making sac-
rifces in terms of their economic growth path by focusing macroeconomic
policy on maintaining infation at no more than 35 percent?
In Section 6.2 we briefy review the overarching and longstanding
analytic debates on the relationship between infation and economic
growth, then focus specifcally on the recent econometric research that has
explored that relationship. In Section 6.3 we present basic descriptive data
from our data sample, then examine the main results from our various
economic exercises. In the concluding Section 6.4 we consider the broader
implications of our fndings, especially as they relate to policy debates
around infation targeting and possible alternative approaches to infation
control.
6.2 LITERATURE OVERVIEW
6.2.1 Analytic Perspectives
We begin by separating out the phenomenon of hyperinfation, which
we broadly defne as being annual infation rates in excess of 40 percent
per year. Hyperinfations occur through a variety of specifc factors. But
regardless of their specifc origins, hyperinfations represent a breakdown
of economic functionings. We will assume that hyperinfations corre-
spond with, and are detrimental to, a positive economic growth path. We
are therefore leaving aside here the possibility that there may be some
positive correspondence between infation above 40 percent and economic
growth.
Hyperinfations aside, the relationship between infation and growth has
been at the very center of macroeconomic theory debates since the mon-
etarist counterrevolution against Keynesianism beginning in the 1960s.
3

The main progeny of that counterrevolution the natural rate of unem-
ployment, the vertical Phillips Curve, and New Classical Economics more
generally have been focused largely around demonstrating that there can
be no positive benefts for economic growth or employment of operating
an economy at anything above a minimal infation rate in the range of 23
percent. From this perspective, infation impedes ef cient resource alloca-
tion by obscuring the signaling role of relative price changes, which, in
turn, is the most important guide to ef cient economic decision making.
This position contrasts sharply with the Keynesian perspective and
the early Phillips Curve models, which held that infation and economic
growth can be positively associated when infationary pressures emerge as
a byproduct of rising aggregate demand. In this Keynesian framework it
is not the case that infation is itself a positive engine of growth, certainly
not a primary growth-inducing force. The point is rather that if rising
aggregate demand is leading to increased growth, then some infation-
ary pressures are likely to emerge in this scenario as a relatively benign
byproduct. Within this Keynesian framework, there could also be reasons
for infation and growth to be negatively correlated. This would occur
when infation results from monopolistic pricing practices, exchange rate
volatility or supply shocks. These problems can also be compounded
when adequate policy interventions do not occur to dampen the infation-
ary impulses induced by monopolistic pricing, exchange rate volatility or
supply shocks.
6.2.2 Recent Empirical Studies
Probably the most infuential recent contribution to the econometric lit-
erature on infation and growth is that of Bruno and Easterly (with results
presented in both Bruno (1995) and Bruno and Easterly (1998)). Bruno
and Easterly examined the relationship between infation and economic
growth for 127 countries between 1960 and 1992. Their examination of
this data set is historical and descriptive. They do not present a formal
econometric model.
Their key conclusion was that there is no robust evidence from this data
sample demonstrating a trade-of between output growth and infation.
More specifcally, only on the basis of two conditions could one observe
a negative growth-infation relationship at all in their data sample. These
were: (1) the inclusion in the data sample of very high infation experi-
ences, that is, rates of infation of 40 percent and higher and (2) increasing
the frequency of the data observations. As they write, The results get
stronger as one goes from the cross-section to ten year averages to fve year
averages to annual data (1998, p. 4).
Once one controls for these two factors, Bruno and Easterly found that
average growth rates fell only slightly as infation rates moved up to 2025
percent. For infation rates below 20 percent, Bruno concluded that there
is no obvious empirical evidence for signifcant long-run growth costs
(Bruno, 1995, p. 38). Moreover, of particular importance for our concerns
with aggregate demand efects on infation and growth, Bruno found that
during 196072, economic growth on average increased as infation rose,
from negative or low rates to the 1520 percent range. This is because,
as Bruno explained, in the 1950s and 1960s, low-to-moderate infation
went hand in hand with very rapid growth because of investment demand
pressures in an expanding economy (ibid., p. 35). Thus, infation that
results directly from economic expansion does not, according to Brunos
fndings, create any signifcant barriers to expansion.
Despite these fndings, Bruno still makes clear in his single-authored
paper that he does not advocate complacency with respect to infation rates
in the 20 percent region. According to Bruno, once infation moves into
the 20 percent region, it is dif cult to contain at this level. This is because,
within the 20 percent infation region, the systems of indexing wages and
fnancial assets, as well as exchange rate adjustments, become more fre-
quent. This then creates a momentum toward accelerating infation.
Neither Bruno alone nor Bruno and Easterly provide systematic evi-
dence on behalf of Brunos concerns about infation within the 20 percent
region. Nevertheless, Bruno is clear in his conclusion that getting infation
down to single digits is important even for longer-term growth reasons
(ibid., p. 38). But even within this less systematic discussion on the dangers
of infation in the 20 percent range, it is still notable that Bruno never sug-
gests that infation needs to be pushed below a single-digit threshold and
specifcally down into the 35 percent range advocated by proponents of
infation targeting.
Since the Bruno and Easterly study, various researchers have examined
the output growth-infation relationship through more formal techniques
than those employed by Bruno and Easterly while still searching out, as
with Bruno and Easterly, potential non-linearities. For example, in a 1998
paper, International Monetary Fund (IMF) economists Atish Ghosh and
Steven Phillips combine panel regression techniques with non-linear treat-
ment of the infation-growth relationship. They also utilize a decision-tree
technique that, in their view, is more robust to outliers and non-linearities
than is standard regression analysis. Their model draws from a data
sample of IMF member countries over 196096. According to this model,
they fnd evidence of a negative infation threshold at 2.5 percent. But they
also acknowledge that thresholds of 5 or 10 percent generate statistical
results very similar to the 2.5 percent threshold.
A 2001 publication by two separate IMF economists, Moshin Khan
and Abdelhak Senhadji, ofers two innovations relative to Ghosh and
Phillips (Khan and Senhadji, 2001). The frst is their use of conditional
least squares, a new non-linear estimation technique. The second, and
more straightforward, innovation was to divide their data sample into
industrial and developing countries. Based on this approach, they fnd
that the threshold level above which infation signifcantly slows growth
is 13 percent for industrial countries and 1112 percent for developing
countries.
More recently still, a 2004 paper by Burdekin et al. followed Khan and
Senhadji in allowing for diferent threshold efects among the industrial
and developing countries (Burdekin et al., 2004). They also allow for
non-linearities in the growth-infation relationship through utilizing spline
estimation techniques. The results from this research diverge sharply
from Khan and Senhadji. In terms of point estimates, they found that the
turning point for industrial countries was 8 percent while that for develop-
ing countries was 3 percent.
In short, all of these studies are in broad concurrence with Bruno and
Easterly as to the presence of non-linearities in the growth-infation rela-
tionship. They also broadly concur with Brunos conclusion that the nega-
tive efects of infation will occur somewhere below a 20 percent threshold,
most likely in the single-digit range. However, they diverge sharply as to
where the turning point occurs within a range of roughly 12 percent infa-
tion or less. Moreover, the two studies that adopted the simple innova-
tion of dividing the sample between industrial and developing countries
reached opposite conclusions as to which set of countries had a higher
infation threshold. Thus, despite the deployment of sophisticated tech-
niques for capturing the impact of non-linearities in the growth-infation
relationship, major questions remain unresolved. In particular, there
remains no robust evidence in support of a policy goal of maintaining an
infation target in the range of 35 percent.
6.3 DESCRIPTIVE DATA AND ECONOMETRIC
EVIDENCE
Our own model is a straightforward panel data model, in which we aim
to isolate the efects of infation on economic growth through including a
series of control variables as well as allowing for a non-linear component
to the growth-infation relationship. Our data sample runs from 1961 to
2000, including data from a total of 80 countries. We have excluded from
the model countries whose population is less than 2 million people. We
do this to focus our empirical exercises on countries whose economies are
minimally large enough so that the countries patterns of economic activ-
ity can be understood as having features that are distinct to that country.
Appendix 6.1A provides a full list of the countries in our data sample.
6.3.1 Descriptive Statistics
We frst provide some basic descriptive statistics from our data sample in
both Table 6.1 and Figure 6.1. Table 6.1 shows both means and standard
deviations for infation and growth, for the full sample, and broken out
according to our three income-level groupings. For all countries in the
sample, as we see, the average rate of GDP growth is 1.9 percent and the
average infation rate is 10.2 percent. However, from the standard deviations
2.7 percent for GDP growth and 7.2 percent for infation we also see that
there are wide disparities among the observations in the sample.
The disparities do diminish as we break out the full sample of countries
into income-level groupings. Not surprisingly, the OECD countries expe-
rience the highest average rate of economic growth (virtually by defnition;
see Note 4) and the lowest average infation rates. Average growth is sig-
nifcantly faster in the middle-income countries relative to the low-income
countries, but average infation is somewhat lower in the low-income
countries.
The four scatter plots in Figure 6.1 show the range of values for our
data sample more fully. No strong patterns at all emerge from these fgures
in terms of the infation/GDP growth relationship. Of course, these data
plots do not control for factors other than infation that could be afecting
economic growth.
We label in the four diagrams in Figure 6.1 the data points that emerge
as outliers through simple observation. This provides some useful perspec-
tive. For example, with the full set of countries, the most rapid growth
spurt was experienced by Haiti from 1996 to 2000. Haiti grew on average
by 15.2 percent in this period, even while infation was rising at an average
Table 6.1 Descriptive statistics on GDP growth and infation 80 country
sample, 19612000
All
countries
(80 countries)
(%)
OECD
countries
(21 countries)
(%)
Middle-income
countries
(32 countries)
(%)
Low-income
countries
(25 countries)
(%)
GDP growth
Mean 1.9 2.6 1.8 0.9
Standard
deviation 2.7 1.8 2.7 3.4
Infation
Mean 10.2 7.0 12.8 11.3
Standard
deviation 7.2 4.9 8.3 6.6
Note: Israel and Singapore are not included in the country groupings because they are non-
OECD high-income countries.
Source: See Appendix 6.1A.
of 14.9 percent. In terms of other outliers in the all-country diagram, we
see that the very high infation and/or very low growth outliers are all low-
or middle-income countries, with Rwanda, Nicaragua and Zimbabwe all
experiencing severe political conficts during their low growth/high infa-
tion years.
With the OECD and middle-income country diagrams, we see that
the countries able to experience the most rapid economic growth rates
were Japan, Ireland, South Korea and China. In all cases the rapid GDP
growth was tied to reaching new levels of export success. Infation in these
8
4
0
4
8
12
16
10 0 10 20 30 40 50
Inflation rate (%)
G
D
P

g
r
o
w
t
h

r
a
t
e

(
%
)
All countries
Haiti
19962000
Venezuela
198690
Ecuador
199195
Rwanda
199195
Nicaragua
197681
Costa Rica
198185
Zimbabwe
19962000
2
0
2
4
6
8
10
12
0 4 8 12 16 20 24
Inflation rate (%)
G
D
P

g
r
o
w
t
h

r
a
t
e

(
%
)
OECD countries
Japan
196670
S. Korea
198690
Ireland
19962000
Greece
198690
Puerto Rico
197681
Puerto Rico
198185
S. Korea
196670
New Zealand
197680
Finland 199195
Switzerland 199195
8
4
0
4
8
12
10 0 10 20 30 40 50
Inflation rate (%)
G
D
P

g
r
o
w
t
h

r
a
t
e

(
%
)
Middle-income countries
China
199195
Venezuela
198690
Ecuador
197175
Costa Rica
198185
Ecuador 199195
Venezuela
197175
Jordan 198690
8
4
0
4
8
12
16
0 10 20 30 40
Inflation rate (%)
G
D
P

g
r
o
w
t
h

r
a
t
e

(
%
)
Low-income countries
Haiti
19962000
Rwanda
199195
Nicaragua 197680
Ghana
198690
Zimbabwe
19962000
Note: Annual data are grouped into fve-year averages.
Source: See Appendix 6.1A.
Figure 6.1 Infation and economic growth, 19612000
countries over the relevant years ranged widely, between 2.6 and 13.2
percent. Clearly, it is dif cult to ofer generalizations from these fgures
as to the interrelationship between infation and economic growth. It is
evident that we need to examine this relationship more systematically, the
task to which we now turn.
6.3.2 Econometric Model
Our approach has been to build a formal model that is still consistent with
the main strength of the Bruno-Easterly framework, which is its simplicity.
To do this, we work with a panel model that incorporates non-linearities
through two relatively simple procedures.
The frst feature of our non-linear model is to simply exclude from our
data set all observations in which infation exceeded 40 percent. As we
mentioned above, we accept the fnding of Bruno and Easterly that infa-
tion in that high range will produce negative efects on growth. We are
therefore efectively asking with our model whether an annual infation
rate below 40 percent exerts a negative efect on economic growth, and if
so, at what point are such negative efects likely to emerge?
The second way that we introduce non-linearity in our model is to
include the squared term on infation as an explanatory variable, which
means we are estimating the regression equations as a second-degree
polynomial. This is a straightforward, widely used technique for estimat-
ing non-linear relationships, through allowing for changes in slopes as a
function of changes in the independent variable. In this case the slope of
the estimating equation can vary with changes in the infation rate. This
enables us to observe turning points in the relationship between infation-
growth and infation-equality. We can observe such possible turning
points through this calculation:
Turning point 5 2((inflation coefficient) /
(2 * (inflation 2 squared coefficient)).
Within this framework, we then also pursue robustness tests through
three sets of straightforward procedures:
1. We utilize four diferent panel data techniques: pooled ordinary least
squares (OLS), between efects fxed efects, and random efects. In
principle, researchers are supposed to establish through diagnostic
exercises which of the four techniques is appropriate with a given
data sample. In practice, however, it is frequently dif cult to know
which technique is the most reliable. Each of the techniques has both
strengths and weaknesses. A pooled OLS model implicitly assumes
there are no problems of omitted variables in a model, which is not
likely to be true, even through frequently the problems may not
be serious enough to substantially distort ones results. A between
efects model averages the data for each country into one observa-
tion. It is therefore testing more narrowly for variation on a country-
by-country basis, as opposed to considering variation between time
periods as well as countries. With the fxed efects model, we are
allowing for intercept shifts to occur for each country, based on the
range of possible omitted variables in evaluating country-by-country
determinants of economic growth. But the fxed efects model
efectively creates dummy variables for each country in the sample,
which reduces degrees of freedom. Finally, the random efects model
also allows for a diferent intercept for each country in the sample.
But the random efects model isolates these individual country efects
in the error term, and therefore does not reduce degrees of freedom
in the manner of the fxed efects estimator. But, at the same time, to
be an unbiased estimator, the random efects model requires that the
omitted variable efects will be uncorrelated with the explanatory
variables.
Given this range of concerns with the various techniques, we report
results utilizing all four techniques. By examining results generated by
all four techniques, we are able to assess the robustness of our fndings
across the range of panel data estimators.
2. We run regressions both on the full set of countries as one sample, then
through dividing the countries into three groupings, OECD countries,
middle-income countries and low-income countries. We are therefore
able to observe the extent to which diferences in the results are due
to broad diferences in the various countries level of development, as
distinct from the individual country diferences that we control for
through the fxed and random efects models.
4
3. We decompose the full time period into four decade-long sub-periods.
This enables us to examine how the relationships may have changed
over time. We are especially interested in following up on Brunos
observation that infation and growth were positively correlated from
the 1960s up until the 1973 oil shock. This was a period in which, as
Bruno said, infation emerged out of explicit eforts to stimulate aggre-
gate demand.
Beyond these distinct features of our model, we also incorporate
a set of control variables in each specifcation of the model. These
control variables are standard in cross-country estimates of the
determinants of economic growth. They include (1) the initial level
of GDP; (2) the share of investment spending in GDP; (3) the share
of government spending in GDP; (4) the fscal defcit; (5) educational
levels; (6) the level of overall health, as measured by life expectancy;
(7) the change in terms of trade; (8) the efects of natural disasters; and
(9) the efects of wars. In addition, we include dummy variables for
each year in the pooled OLS, fxed efects and random efects models
to control for the time efects within each set of country observations.
Full descriptions of each of the control variables is reported in the
Appendix 6.1B. We do not report here the full set of results on the
control variables, but these results are available on request.
5
We report the key fndings of our econometric models in Tables 6.2
and 6.3. Both tables report the coef cients and t-statistics for the infa-
tion and infation-squared variables only for each of the regressions.
We also report the turning points estimated by each equation when
infation switches from becoming a positive to negative, or negative to
positive, infuence on growth.
Results for the full time period
The results for the full time period are presented in Table 6.2. Considering
frst the data for all countries in the sample, we see that the sign of the
infation coef cient is consistently positive across specifcations, and is
statistically signifcant in all but the fxed efects specifcation. Moveover,
the coef cient values across specifcations are similar, ranging between
0.090.15. The coef cients on the infation-squared terms are also similar
and statistically signifcant in all cases. With the coef cients on the infa-
tion and infation-squared terms, we can then calculate the turning points
as being between a 15.2 and 18.6 infation rate. Overall, this frst set of tests
with the full data sample suggest that the rate of economic growth rises by
between about 0.1 and 0.15 percent for every percentage point increase in
the infation rate up to a 1518 percent threshold. Infation then becomes
a damper on growth beyond this threshold.
The clear fndings we obtain with the full data set is, however, not main-
tained when we consider OECD, middle-income and low-income coun-
tries separately. With the data grouped by income levels, we expect that
the signifcance levels will go down due to the smaller sample sizes. And
we do indeed observe generally lower signifcance levels with the results
grouped by income levels.
More specifcally, in the case of the OECD countries, none of the coef-
fcients for infation or infation-squared are statistically signifcant in any
of the specifcations. Moreover, the signs on the infation variable shift to
negative in the pooled OLS, between efects and random efects models. In
126
T
a
b
l
e

6
.
2

I
n
f

a
t
i
o
n

a
n
d

e
c
o
n
o
m
i
c

g
r
o
w
t
h
,

1
9
6
1
2
0
0
0
M
o
d
e
l
A
l
l

c
o
u
n
t
r
i
e
s
O
E
C
D

c
o
u
n
t
r
i
e
s

P
o
o
l
e
d
O
L
S
F
i
x
e
d

e
f

e
c
t
s
R
a
n
d
o
m

e
f

e
c
t
s
B
e
t
w
e
e
n

e
f

e
c
t
s
P
o
o
l
e
d
O
L
S
F
i
x
e
d

e
f

e
c
t
s
R
a
n
d
o
m

e
f

e
c
t
s
B
e
t
w
e
e
n

e
f

e
c
t
s
N
o
.

o
f

o
b
s
e
r
v
a
t
i
o
n
s
:
3
5
6
3
5
6
3
5
6
8
0
1
3
5
1
3
5
1
3
5
2
1
I
n
f

a
t
i
o
n
0
.
1
1
*
0
.
0
9
1
0
.
1
1
*
0
.
1
4
9
*
0
.
0
5
5
0
.
0
2
5
0
.
0
3
4
0
.
1
3
0
(
2
.
4
9
)
(
1
.
6
1
)
(
2
.
5
5
)
(
1
.
9
9
)
(
0
.
6
6
)
(
0
.
2
3
)
(
0
.
3
6
)
(
0
.
3
0
)
(
I
n
f

a
t
i
o
n
)
0
.
0
0
3
*
*
0
.
0
0
3
*
0
.
0
0
3
*
*
0
.
0
0
4
*
0
.
0
0
5
0
.
0
0
7
0
.
0
0
5
0
.
0
0
1
(
3
.
7
3
)
(
2
.
4
1
)
(
3
.
4
2
)
(
2
.
3
5
)
(
1
.
4
3
)
(
1
.
7
9
)
(
1
.
3
7
)
(
0
.
0
6
)
T
u
r
n
i
n
g

P
o
i
n
t
1
8
.
3
1
5
.
2
1
8
.
3
1
8
.
6
5
.
5
1
.
8
3
.
4
6
5
.
0
M
o
d
e
l
M
i
d
d
l
e
-
i
n
c
o
m
e

c
o
u
n
t
r
i
e
s
L
o
w
-
i
n
c
o
m
e

c
o
u
n
t
r
i
e
s
P
o
o
l
e
d
O
L
S
F
i
x
e
d

e
f

e
c
t
s
R
a
n
d
o
m

e
f

e
c
t
s
B
e
t
w
e
e
n

e
f

e
c
t
s
P
o
o
l
e
d
O
L
S
F
i
x
e
d

e
f

e
c
t
s
R
a
n
d
o
m

e
f

e
c
t
s
B
e
t
w
e
e
n

e
f

e
c
t
s
N
o
.

o
f

o
b
s
e
r
v
a
t
i
o
n
s
1
2
7
1
2
7
1
2
7
3
2
8
6
8
6
8
6
2
5
I
n
f

a
t
i
o
n
0
.
0
6
0
.
0
2
8
0
.
0
5
7
0
.
1
2
9
0
.
3
5
9
0
.
5
5
9
*
0
.
3
8
6
*
0
.
2
3
8
(
1
.
1
2
)
(
0
.
3
1
)
(
0
.
8
4
)
(
1
.
1
9
)
(
1
.
3
8
)
(
2
.
3
8
)
(
2
.
2
9
)
(
0
.
7
3
)
(
I
n
f

a
t
i
o
n
)
0
.
0
0
2
*
0
.
0
0
1
0
.
0
0
2
0
.
0
0
4
0
.
0
1
0
.
0
1
2
*
0
.
0
1
*
0
.
0
0
8
(
2
.
4
5
)
(
0
.
8
4
)
(
1
.
6
7
)
(
1
.
7
1
)
(
1
.
5
5
)
(
2
.
2
0
)
(
2
.
3
6
)
(
0
.
7
2
)
T
u
r
n
i
n
g

P
o
i
n
t
1
5
.
0
1
4
.
0
1
4
.
3
1
6
.
1
1
8
.
0
2
3
.
3
1
9
.
3
1
4
.
9
N
o
t
e
s
:
1
.

D
a
t
a

g
r
o
u
p
e
d

a
s

f

v
e
-
y
e
a
r

a
v
e
r
a
g
e
s
;

d
e
p
e
n
d
e
n
t

v
a
r
i
a
b
l
e

i
s

G
D
P

g
r
o
w
t
h
;

t
-
s
t
a
t
i
s
t
i
c
s

i
n

p
a
r
e
n
t
h
e
s
e
s
;

p

,

0
.
0
5

5

*
,

p

,

0
.
0
1

5

*
*
.
2
.

A
l
l

c
o
u
n
t
r
i
e
s

w
i
t
h

l
e
s
s

t
h
a
n

2

m
i
l
l
i
o
n

p
e
o
p
l
e

w
e
r
e

e
x
c
l
u
d
e
d

f
r
o
m

t
h
e

s
a
m
p
l
e
.

A
l
l

o
b
s
e
r
v
a
t
i
o
n
s

w
i
t
h

i
n
f

a
t
i
o
n

a
b
o
v
e

4
0

p
e
r
c
e
n
t

w
e
r
e

e
x
c
l
u
d
e
d
.

F
i
v
e
-
y
e
a
r

p
e
r
i
o
d

a
v
e
r
a
g
e

d
a
t
a

w
e
r
e

u
s
e
d

i
n
s
t
e
a
d

o
f

y
e
a
r
l
y

d
a
t
a

i
n

t
h
e

r
e
g
r
e
s
s
i
o
n
.

A
l
l

c
o
u
n
t
r
i
e
s

r
e
f
e
r
s

t
o

a
l
l

t
h
e

c
o
u
n
t
r
i
e
s

i
n

t
h
i
s

t
a
b
l
e
.
short, we do not obtain any reliable results on the infation-growth rela-
tionship for the OECD countries.
With the middle-income countries, the signs on the infation coef cient
are all positive. However, the coef cients are insignifcant in all cases, the
coef cient values correspondingly jump from 0.06 to 0.129. However,
the estimated turning points in these equations are within a tight band of
between 1416 percent.
Finally, with the low-income countries, we do again obtain consistently
positive coef cient values from the infation variable. These coef cients
are also signifcant in the fxed efects and random efects models. The
coef cient values in these regressions are substantially higher than with
the other country groupings, ranging between 0.24 and 0.56. The infation-
squared terms are also strongly signifcant in the fxed and random efects
models. The turning point estimates range between 1523 percent.
Results by decade
In Table 6.3 we report results from regressions run separately for each of
the four decades in our data sample. These regressions are run with annual
data rather than fve-year averages in order to generate a larger number
of observations. In doing this, we recognize the point, emphasized by
Bruno and Easterly, that we should expect more of a negative correspond-
ence between infation and growth as we move to higher frequency data
samples. This is because of the likelihood that negative efects on growth
will occur through short bursts of high infation rates that approach our
cut-of fgure of 40 percent. Such short bursts of high infation and slow
or negative growth will be smoothed out when data are grouped at lower
frequencies.
One key point emerges from the results in Table 6.3: that the evidence
for a positive association between growth and infation is far stronger in
the 196170 decade than in subsequent decades. For 196170, the coef-
fcient values on infation are all positive, though statistically signifcant
only in the between efects model. The fxed efects model stands apart
with a low infation coef cient value of 0.065, but otherwise the coef c-
ients for the other specifcations are high, at 0.11 for the pooled OLS and
random efects models and a very high 0.61 for the between efects model.
With the 197180 sample, the infation coef cients remain positive, but
the coef cient values and levels of signifcance fall of, especially with the
random efects and between efects models. For 198190, the infation
coef cients all turn negative, and there is a statistically signifcant negative
value in the fxed efects model. Finally, for 19912000, we obtain negative
infation coef cients with two tests and close to zero coef cients for the
other two.
128
T
a
b
l
e

6
.
3

I
n
f

a
t
i
o
n

a
n
d

e
c
o
n
o
m
i
c

g
r
o
w
t
h

a
l
l

c
o
u
n
t
r
i
e
s

b
y

d
e
c
a
d
e
s
Y
e
a
r
s
1
9
6
1
7
0
1
9
7
1
8
0
M
o
d
e
l
P
o
o
l
e
d
O
L
S
F
i
x
e
d

e
f

e
c
t
s
R
a
n
d
o
m

e
f

e
c
t
s
B
e
t
w
e
e
n

e
f

e
c
t
s
P
o
o
l
e
d
O
L
S
F
i
x
e
d

e
f

e
c
t
s
R
a
n
d
o
m

e
f

e
c
t
s
B
e
t
w
e
e
n

e
f

e
c
t
s
N
o
.

o
f

o
b
s
e
r
v
a
t
i
o
n
s
4
8
0
4
8
0
4
8
0
5
9
6
2
0
6
2
0
6
2
0
7
1
I
n
f

a
t
i
o
n
0
.
1
1
2
0
.
0
6
5
0
.
1
0
9
0
.
6
1
*
0
.
0
8
4
0
.
0
6
0
.
0
8
4
0
.
0
4
7
(
1
.
0
4
)
(
.
0
7
0
)
(
1
.
2
4
)
(
2
.
0
8
)
(
0
.
9
9
)
(
0
.
6
1
)
(
0
.
9
2
)
(
0
.
2
0
)
(
I
n
f

a
t
i
o
n
)
.
0
0
0
5
0
.
0
0
7
*
0
.
0
0
6
0
.
0
2
4
*
0
.
0
0
5
*
0
.
0
0
6
*
0
.
0
0
5
*
0
.
0
0
1
(
1
.
4
7
)
(
2
.
1
0
)
(
1
.
7
4
)
(
2
.
0
8
)
(
2
.
0
0
)
(
2
.
2
1
)
(
2
.
0
3
)
(
0
.
1
3
)
T
u
r
n
i
n
g

P
o
i
n
t
1
1
.
2
4
.
6
9
.
1
1
2
.
7
8
.
4
5
.
0
8
.
4
2
3
.
5
Y
e
a
r
s
1
9
8
1
9
0
1
9
9
1
2
0
0
0
M
o
d
e
l

P
o
o
l
e
d
O
L
S
F
i
x
e
d

e
f

e
c
t
s
R
a
n
d
o
m

e
f

e
c
t
s
B
e
t
w
e
e
n

e
f

e
c
t
s
P
o
o
l
e
d
O
L
S
F
i
x
e
d

e
f

e
c
t
s
R
a
n
d
o
m

e
f

e
c
t
s
B
e
t
w
e
e
n

e
f

e
c
t
s
N
o
.

o
f

o
b
s
e
r
v
a
t
i
o
n
s
7
1
8
7
1
8
7
1
8
8
5
6
9
8
6
9
8
6
9
8
1
0
6
I
n
f

a
t
i
o
n
0
.
0
1
6
0
.
1
1
8
*
0
.
0
1
6
0
.
0
1
6
0
.
0
2
5
0
.
1
3
0
.
0
1
0
.
1
8
9
(
0
.
3
1
)
(
1
.
9
8
)
(
0
.
3
3
)
(
0
.
1
7
)
(
0
.
3
1
)
(
1
.
6
9
)
(
0
.
1
5
)
(
1
.
3
2
)
(
I
n
f

a
t
i
o
n
)
0
.
0
0
0
.
0
0
1
0
.
0
0
0
.
0
0
3
0
.
0
0
3
0
.
0
0
1
0
.
0
0
3
0
.
0
0
4
(
0
.
0
0
8
)
(
0
.
6
1
)
(
0
.
0
8
)
(
0
.
9
1
)
(
1
.
3
0
)
(
0
.
4
3
)
(
1
.
7
0
)
(
0
.
8
7
)
T
u
r
n
i
n
g

P
o
i
n
t
5
9
.
0
2
.
7
4
.
2
6
5
.
0
1
.
7
2
3
.
6
N
o
t
e
s
:
1
.

A
n
n
u
a
l

d
a
t
a
;

d
e
p
e
n
d
e
n
t

v
a
r
i
a
b
l
e

i
s

G
D
P

g
r
o
w
t
h
;

t
-
s
t
a
t
i
s
t
i
c
s

i
n

p
a
r
e
n
t
h
e
s
e
s
;

p

,

0
.
0
5

5

*
,

p

,

0
.
0
1

5

*
*
)
.
2
.

A
l
l

c
o
u
n
t
r
i
e
s

w
i
t
h

l
e
s
s

t
h
a
n

2

m
i
l
l
i
o
n

p
e
o
p
l
e

w
e
r
e

e
x
c
l
u
d
e
d

f
r
o
m

t
h
e

s
a
m
p
l
e
.

A
l
l

o
b
s
e
r
v
a
t
i
o
n
s

w
i
t
h

i
n
f

a
t
i
o
n

a
b
o
v
e

4
0

p
e
r
c
e
n
t

w
e
r
e

e
x
c
l
u
d
e
d
.

F
i
v
e
-
y
e
a
r

p
e
r
i
o
d

a
v
e
r
a
g
e

d
a
t
a

w
e
r
e

u
s
e
d

i
n
s
t
e
a
d

o
f

y
e
a
r
l
y

d
a
t
a

i
n

t
h
e

r
e
g
r
e
s
s
i
o
n
.

A
l
l

c
o
u
n
t
r
i
e
s

r
e
f
e
r
s

t
o

a
l
l

t
h
e

c
o
u
n
t
r
i
e
s

i
n

t
h
i
s

t
a
b
l
e
.
These results provide broad support for Brunos observation cited
above, about infation and growth moving positively together during the
1960s in correspondence with, as he put it, very rapid growth because of
investment demand pressures in an expanding economy (Bruno, 1995, p.
35). During the 1970s, demand management policies were still in favor to
support growth. But the positive associations between growth and infa-
tion as a byproduct of growth in this period were undermined by the two
oil price shocks in 1973 and 1979. The overall infation experience of this
decade therefore is a combination of demand-pull efects from growth and
supply side shocks. It is therefore not surprising that the infation coef c-
ients in the 1970s fall in value and lose signifcance.
The 1980s marked the beginning of what Angus Maddison (2001),
among others, has term the neoliberal era. Probably the single defning
feature of this era is the virtual abandonment by governments throughout
the world of Keynesian demand management policies as a tool for stimu-
lating growth and employment. Thus, as a broad generalization, the infa-
tion that is experienced in the 1980s and 1990s emerges almost entirely as
a result of supply shocks and inertia, as opposed to demand-pull pressures.
Within this context it is also not surprising that the infation coef cients
become consistently negative, albeit generally not to a statistically signif-
cant extent.
6.4 CONCLUSIONS
Considering frst our full data set of 80 countries between 1961 to 2000, we
have consistently found that higher infation is associated with moderate
gains in GDP growth up to a roughly 1518 percent infation threshold.
However, the fndings diverge when we divide our full data set accord-
ing to income levels. With the OECD countries, no clear pattern emerges
at all with either the infation coef cient or our estimated turning point.
Both the signs on the infation coef cients as well as the turning points
are highly sensitive to specifcations. With the middle-income countries,
by contrast, we return to a consistently positive pattern of infation coef-
fcients, though none are statistically signifcant. However, the turning
points range within a narrow band in this sample, between 14.0 and 16.3
percent. With the low-income countries, we obtain positive and higher
coef cient values on the infation coef cient than with the middle-income
countries. These coef cients are also statistically signifcant with the fxed
and random efects models.
Finally, with the groupings by decade, the results broadly indicate that
infation and growth will be more highly correlated to the degree that
macroeconomic policy is focused on demand management as a stimulus to
growth as opposed to macroeconomic austerity and infation targeting.
Overall, there is no evidence from this research supportive of a policy
of maintaining infation within a low band of about 35 percent, to the
degree that government policy makers are interested in promoting eco-
nomic growth and employment, rather than merely low infation as an
end in itself. At the same time, there is also no evidence that governments
should allow infation to rise above a 1520 percent range in an efort to
spur growth.
This suggests that there is still a wide range of infation rates that are
very likely to be associated positively with economic growth. Certainly for
the middle- and low-income countries, our results strongly suggest that
allowing infation to be maintained in the range of 10 percent or somewhat
higher is very likely to be consistent with higher rates of economic growth.
This is most especially the case when infation is resulting from, as Bruno
put it, investment demand pressures in an expanding economy.
For the OECD countries, the primary conclusion that we can reach
from our results is a negative one: that no generalization about the
infation-growth relationship is likely to fnd robust support from the
available evidence. What appears likely for the OECD countries is that
the wide range of relationships that emerge from the data refect the dif-
ferences in the sources of infation that is, whether infation has resulted
primarily from Keynesian type demand-pull forces as opposed to supply
shocks and inertia.
Some broad policy implications fow from these results. The frst is that
there is no justifcation for infation targeting policies as they are currently
being practiced throughout the world, that is, to maintain infation with
a 35 percent band and to adjust short-term interest rates as needed to
dampen infationary pressures beyond that targeted band. As a corollary,
there is very likely to be positive growth benefts in middle- and low-income
countries from allowing infation to rise to a high single-digit range or even
in some cases up to about 15 percent rather than dampening infationary
pressures through raising short-term interest rates. This is especially true
to the extent that infation within this range is resulting from demand-pull
forces as opposed to supply shocks and inertia.
6
A second implication is that researchers are likely to make productive
contributions through giving increased attention on the infation-growth
relationship to some relatively underexplored aspects of the issue. The frst
is to be able to sort out with increased specifcity the sources of infationary
pressures, given the likely wide disparities in the infation-growth relation-
ship depending on what is fueling infation. A second is to focus more on
policy measures for dampening infation not at very low levels, but rather
at levels approaching the upper limit of the positive growth-infation
association. This would be in the range of 1015 percent for middle- and
low-income countries. With the OECD countries, the acceptable range is
likely to depend entirely on what are the primary sources of infationary
pressures.
One well-known policy tool for maintaining infationary pressures
within a positive threshold range is some variation of incomes policies.
Incomes policies have been widely used as an infation control tool in a
variety of contexts. One common situation has been in bringing down
infation after it has risen to a range above 40 percent. For example, in
their paper Moderate infation Dornbusch and Fischer (1991) describe
how Mexico in the 1980s drew upon experiences in Argentina, Brazil, Peru
and Israel in developing a strategy to bring infation down from the 100
percent range to something closer to 20 percent. Dornbusch and Fischer
reported that the Mexicans learned two lessons from these experiences,
(1) that disinfation without fscal discipline was unsustainable; but that
(2) disinfation without incomes policy, relying solely on tight money and
tight budgets, would be unnecessarily expensive (p. 31). In analysing the
Israeli experience with disinfation over the 1980s, Bruno documents in
detail the major contributions of incomes policies to the success of the
efort (Bruno 1993, Chapter 5).
A more directly relevant set of experiences with respect to infations
at more moderate levels have been the Nordic countries. This is because,
in these countries, incomes policies have been used successfully as a tool
for maintaining relatively low infation over long periods of time rather
than as primarily an instrument of disinfation after infation exceeded 40
percent, as was true with Mexico and Israel. Sweden, for example, suc-
ceeded in maintaining unemployment at an average rate below 2 percent
between 1951 and 2000 while still holding infation at a 4.4 percent average
rate. The application of incomes policies in Sweden, moreover, primarily
took the form of centralized bargaining between unions and business,
through which the aim of infation control was recognized in the bargain-
ing process. As such, the government did not have to rely on setting man-
dates for acceptable wage and price increases. The government did also
utilize fscal and monetary policies as tools for controlling infation. But
they did not have to apply these tools stringently, precisely because they
were able to rely on their well-developed system of incomes policies as a
complement to monetary and fscal policies.
7
The most basic critique of incomes policies is that, in order for the
approach to have any chance of success, it is necessary that a country
operate with a high level of organization among workers, and that there
be some reasonable degree of common ground between workers and
business. Otherwise, there will be no realistic prospect for economy-wide
bargaining to yield results that will be honored widely. By its very nature,
the relationship between unions and business in capitalist economies is
likely to be highly contentious. But this could possibly diminish to the
extent that both sides see the benefts accelerated economic growth and
employment expansion as opposed to maintaining tight monetary policy
for the purpose of holding infation within a 35 percent band.
This point brings us to a fnal issue for further research that includes
both purely analytic as well as policy oriented implications. This is to
examine the relationship between infation and inequality in addition to
the infation-growth relationship. To the extent that infation is associ-
ated with faster economic growth, it is likely to also be correlated with
faster employment growth and thereby increased equality. At the same
time, to the extent that wage agreements and social benefts do not include
adequate cost-of-living adjustments, even a growth-generated infation
could yield greater inequality. In terms of policy implications, the issues
that are central in the exploration of the infation-inequality relationship
will also be closely linked to the question of infation control policies. For
example, are incomes policies or infation targeting a more efective means
of promoting greater equality as well as economic growth? These are
crucial questions that deserve substantial additional research in an efort
to design more efective analytic foundations for the conduct of macro-
economic policy.
NOTES
1. This chapter was written as part of a broader project on pro-poor macroeconomic poli-
cies in sub-Saharan Africa, under the sponsorship of the United Nations Development
Programme (UNDP). We are most grateful for the support of the UNDP. We are also
grateful for comments on this specifc aspect of the broader project from our co-workers
Jerry Epstein, James Heintz and Leonce Ndikumana as well as from Michael Ash.
2. An excellent survey of infation targeting and related issues in global monetary macro-
economics is Saad Filho (2005).
3. The literature on this issue is of course vast. Three references ofering diferent perspec-
tives are Cross (1995), Krueger and Solow (2001), and Saad Filho (2005).
4. Grouping the countries in the sample by average GDP levels does raise the potential for
signifcant bias in the regression. This is because the dependent variable in the model is
GDP growth. Strictly speaking, we are not dividing the sample based on the dependent
variable, but there is obviously a close correspondence between the growth of GDP,
our dependent variable and GDP levels, the variable on which we truncate the sample.
To test for bias here, we have also divided the full sample based on pre-1960 GDP level
groupings that is, on the basis of data points that precede in time our sampling period.
In this case the division is between current OECD countries and the non-OECD coun-
tries that is, the demarcation between middle- and low-income countries was not so
evident in the pre-1960 data, and only becomes evident over the 40 years that constitute
our data sample. However, the results of this exercise do not vary substantially from
those reported with the three GDP-level groupings reported here. This suggests that any
potential bias from the GDP-level groupings is not a serious problem for our substantive
understanding of the fndings.
5. By exploring the infation-growth relationship within this framework of a standard
cross-country growth regression model, we are building in an assumption that causality
in the relationship is running from infation to growth. We do not explore the issue of
simultaneity or reverse causality in this exercise, while we recognize it as an important
issue for further research.
6. Even some of the most recent work by IMF economists has recognized that, at least for
the low-income countries, infation in the range of 510 percent is likely to be supportive
of economic growth. See IMF (2005).
7. Diferent perspectives on the Nordic experiences are presented in Calmfors (1993),
Pekkarinen et al. (1992), Flanigan (1999), Marshall (1994) and Iversen et al. (2000).
REFERENCES
Bernanke, B., T. Laubach, A.S. Posen and F.S. Mishkin (1999), Infation Targeting:
Lessons from the International Experience, Princeton, NJ: Princeton University
Press.
Blinder, A.S. (1998), Central Banking in Theory and Practice, Cambridge, MA:
MIT Press.
Bruno, M. (1993), Crisis, Stabilization, and Economic Reform: Therapy by
Consensus, Oxford: Clarendon Press.
Bruno, M. (1995), Does infation really lower growth?, Finance and Development,
32 (3) (September), 358.
Bruno, M. and W. Easterly (1998), Infation crises and long-run growth, Journal
of Monetary Economics, 41, 326.
Burdekin, R.C.K., A.T. Denzau, M.W. Keil, T. Sitthiyot and T.D. Willett (2004),
When does infation hurt economic growth? Diferent nonlinearities for difer-
ent economies, Journal of Macroeconomics, 26, 51932.
Calmfors, L. (1993), Centralization of wage bargaining and macroeconomic per-
formance, OECD Economic Studies, 21 (Winter), 16191.
Cross, R. (ed.) (1995), The Natural Rate of Unemployment: Refections on 25 Years
of the Hypothesis, Cambridge: Cambridge University Press.
Dornbusch, R. and S. Fischer (1991), Moderate infation, National Bureau of
Economic Research working paper no. 3896.
Flanigan, R.J. (1999), Macroeconomic performance and collective bargaining: an
international perspective, Journal of Economic Literature, 37 (3) (September),
115075.
Ghosh, A. and S. Phillips (1998), Warning: infation may be harmful to your
growth, IMF Staf Papers, 45 (4), 67286.
International Monetary Fund (IMF) (2005), Monetary and fscal policy design
issues in low-income countries, 8 August draft, Policy Development and Review
Department and Fiscal Afairs Department, IMF, Washington, DC.
Iversen, T., J. Pontusson and D. Soskice (eds) (2000), Unions, Employers, and
Central Banks, Cambridge: Cambridge University Press.
Khan, M.S. and A.S. Senhadji (2001), Threshold efects in the relationship
between infation and growth, IMF Staf Papers, 48 (1), 121.
Krueger, A. and R.M. Solow (eds) (2001), The Roaring Nineties: Can Full
Employment Be Sustained?, New York: The Russell Sage Foundation and the
Century Foundation.
Maddison, A. (2001), The World Economy: A Millennial Perspective, Paris:
Development Centre of the Organisation for Economic Co-operation and
Development.
Marshall, M. (1994), Lessons from the experience of the Swedish model, in P.
Arestis and M. Marshall (eds), The Political Economy of Full Employment,
Aldershot, UK and Brookfeld, US: Edward Elgar, Ch. 10.
Pekkarinen, P., M. Pohjola and B. Rowthorn (1992), Social Corporatism: A
Superior Economic System?, New York: Oxford University Press.
Saad Filho, A. (2005), Pro-poor monetary and anti-infation policies: developing
alternatives to the new monetary policy consensus, manuscript, Department
of Development Studies, School of Oriental and African Studies, University of
London.
APPENDIX 6.1A COUNTRIES INCLUDED IN DATA
POOL FOR ANALYSING THE
INFLATION-ECONOMIC GROWTH
RELATIONSHIP
Data Sample is 1960 to 2001
OECD
countries
Middle-income countries Non-OECD
high-income
countries
Low-income
countries
Australia Algeria Paraguay Israel Bangladesh
Austria Argentina Peru Singapore Burundi
Belgium Bolivia Philippines Cameroon
Canada Brazil Poland Central Afr. R.
Denmark Chile South Africa Congo
Finland China Sri Lanka Ghana
France Colombia Syria Haiti
Greece Costa Rica Thailand India
Ireland Dominican
Rep.
Tunisia Indonesia
Italy Ecuador Uruguay Kenya
Japan Egypt Venezuela Lesotho
Korea El Salvador Malawi
Netherlands Guatemala Mali
New Zealand Honduras Nepal
Norway Hungary Nicaragua
Portugal Iran, I.R. of Niger
Spain Jamaica Pakistan
Sweden Jordan Papua New
Guinea
Switzerland Malaysia Rwanda
United
Kingdom
Mexico Senegal
United States Panama Sierra Leone
Togo
Uganda
Zaire
Zimbabwe
APPENDIX 6.2B SPECIFICATIONS OF VARIABLES
IN THE FULL INFLATION/
ECONOMIC GROWTH MODEL
Economic growth. Real GDP per capita (Constant price: Laspeyres)
growth rate. The nth years growth rate is calculated as the log value of the
ratio of the nth years per capita GDP to the (n-1)th years per capita GDP
(Penn World Table (PWT) 6.1, http://pwt.econ.upenn.edu/).
Infation. The increase of consumer price index (World Bank World
Development Indicators (WDI), 2003).
Initial output level. The log value of per capita GDP (Constant price:
Laspeyres) at the beginning year of each period (PWT 6.1, http://pwt.econ.
upenn.edu/).
Investment. The share of gross investment in GDP (current prices)
(PWT 6.1, http://pwt.econ.upenn.edu/).
Fiscal policy. (1) The share of government consumption in GDP
(current prices) (PWT6.1, http://pwt.econ.upenn.edu/). (2) Government
budget defcit as percentage of GDP (WDI, 2003).
Life expectancy. Life expectancy at birth (WDI CD-ROM, 2003, World
Bank).
Education level. Average years of secondary schooling of the total popu-
lation aged 25 and over (R. Barro and J.W. Lee, 2000, http://www2.cid.
harvard.edu/ciddata/barrolee/panel_data.xls).
Terms of trade. The change of terms of trade weighted by foreign trade
dependence ratio (the sum of exports and imports divided by GDP)
(Easterly, William and Mirvat Sewadeh (2001), Global Development
Network Growth database, www.worldbank.org/research/growth/
GDNdata.htm).
Natural disaster. The share of population afected by the natural disas-
ters happened in the year weighted by the share of agricultural output in
GDP. Unreported natural disasters, if any, are treated as 0 (The natural
disaster data come from the Centre for Research on the Epidemiology
of Disasters (CRED), the OFDA/CRED International Disaster data-
base, http://www.cred.be/emdat/intro.htm. The agricultural data are from
WDI, 2003).
War. A war is defned as an armed confict with more than 25 deaths.
Value 1 is given to those countries that experienced war within its border;
1 is given to those countries involved in war in other countries. Other
situations are given value 0 (Gleditsch et al., 2002, Armed Confict 1946
2002 database, http://www.prio.no/cwp/ArmedConfict/).
PART III
Infation targeting: critiques and
country-specifc alternatives
139
7. Infation targeting in Brazil:
19992006
Nelson H. Barbosa-Filho
Brazil adopted infation targeting after a brief period of exchange rate
targeting that ended up in a major currency crisis. More specifcally,
in 199498 the Brazilian government used high domestic interest rates
and privatization to attract foreign capital and sustain an appreciated
exchange rate peg. The main objective of the Brazilian economic policy at
that time was to reduce infation and the main side efect of exchange rate
appreciation was a substantial increase in the countrys current account
defcit and net public debt. Similar to what happened in Mexico and
Argentina during the 1990s, the Brazilian macroeconomic stabilization
strategy was heavily dependent on the continuous infow of foreign capital
and, as a result, the international fnancial position of Brazil became
increasingly fragile after the contagion efects from the East Asian cur-
rency crises of 1997 and the Russian currency crisis of 1998. In fact, by the
end of 1998 Brazils current account defcit reached 4.5 percent of GDP
and the low stock of foreign reserves of the Brazilian Central Bank (BCB)
did not allow a defense of the Brazilian currency, the real, in case another
speculative attack hit the country.
The inevitable currency crisis came in the beginning of 1999 and
resulted in a maxi-devaluation of the real. In numbers, the Brazilian
real/US dollar exchange rate rose 57 percent in just two months, that
is, from 1.21 in December 1998 to 1.90 in February 1999. After that the
exchange rate dropped a little and then remained around 1.84 during
the rest of 1999, that is, the exchange rate stabilized at a level 52 percent
higher than before the crisis. The initial response of the BCB to such
an abrupt devaluation of the real was a substantial increase in its base
interest rate to stop the capital fight from Brazil and to reduce the
pass-through of exchange rate depreciation to infation.
1
Then, after
the exchange rate stabilized at a higher level in mid 1999, the govern-
ment announced that it would start to target infation. At that time the
basic justifcation for such a move was that the government needed to
substitute a price target for the former exchange rate target in order
to coordinate market expectations and control infation in a context of
foating exchange rates.
In terms of the classic policy trilemma of open economies, the option
for infation targeting meant that Brazil chose to have an indepen dent
monetary policy, free capital fows and a foating exchange rate. In
practice the situation was obviously diferent because the exchange rate
was an important determinant of the Brazilian infation rate during the
period under analysis. On the one hand, the change in the domestic price
of tradable goods in Brazil was (and it continues to be) basically deter-
mined by foreign infation and exchange rate variations. On the other
hand, the prices of some key non-tradable goods in Brazil were also tied
to the exchange rate because, during the privatizations of the 1990s, the
government allowed the price of some public utilities (basically in the
telecommunication and energy sectors) to follow a price index that is
heavily infuenced by the exchange rate. The inevitable result was that a
major part of the Brazilian infation rate was linked to the exchange rate
in 19992006.
2
Given the centrality of the exchange rate for Brazilian infation, it is no
surprise that infation targeting resulted in a disguised and loose exchange
rate targeting by the BCB in 19992006. In theory the exchange rate was
free to foat, but in practice the country ended up with an asymmetric dirty
foating regime, that is, a regime in which the BCB had to fght devalu-
ations but to tolerate revaluations in order to meet the infation targets
set by the Brazilian government. As a result, the interaction between the
international fnancial conditions, on the one side, and the Brazilian inter-
est rates, on the other side, explains most of the behavior of the exchange
rate in Brazil in 19992006, which in its turn explains most of the successes
and failures of infation targeting during that period. The objective of this
chapter is to describe and analyse this experience.
3
7.1 THE BRAZILIAN INFLATION TARGETING
REGIME
The institutional basis of the Brazilian infation targeting system can be
described as follows:
The National Monetary Council (Conselho Monetrio Nacional or
CMN), formed by the Minister of Finance, the Minister of Planning
and the President of the BCB, establishes the infation targets based
on the recommendations of the Finance Minister. All three members
are appointed by the president and do not have fxed mandates.
Infation targeting in Brazil: 19992006 141
In June of every year the CMN establishes the infation targets, and
their corresponding intervals of tolerance, for the next two years.
The target consists of the desired variation of a consumer price index
(the IPCA index) estimated by the governments statistical branch
(the Instituto Brasileiro de Geografa e Estatstica or IBGE).
The BCB is responsible for achieving the target, but no specifc
instrument or strategy is specifed.
4
The Monetary Policy Committee (Comit de Poltica Monetria
or Copom), formed by the president and the directors of the BCB,
decides the level of the BCBs base interest rate, the so-known
SELIC (Sistema Especial de Liquidao e Custodia) rate, needed to
achieve the infation target. Occasionally, additional actions such as
changes in the banks reserve requirements can be taken.
5
The target is considered met whenever the observed accumulated
infation during each calendar year falls within the interval of toler-
ance specifed by the CMN.
If the targets are not met, the president of the BCB has to issue an
open letter to the Minister of Finance explaining the causes of the
failure, the measures to be adopted to ensure that infation returns
to the target level, and the period of time needed for this to happen.
Since the BCB is not independent and its penalty for not meeting the
infation target is just to explain why that happened, we can conclude that
the Brazilian infation targeting regime is basically a loose way for the
federal government to assure society, especially fnancial markets, that it
will not let infation run out of control.
7.2 MACROECONOMIC PERFORMANCE
The macroeconomic performance of Brazil since the adoption of infa-
tion targeting has been mixed. First, considering infation itself, the gov-
ernments targets were met when the international fnancial conditions
allowed it, that is, infation targeting was successful when the exchange
rate dynamics helped the BCB eforts to control infation. Second, when
compared to the period of exchange rate targeting, infation target-
ing brought a reduction in the base domestic real interest rate, but the
Brazilian rate remained well above the international standards because the
country needed to appreciate the exchange rate in order to meet its infa-
tion targets. Third, the high real interest rate put an expansionary pressure
on the net public debt and this had to be compensated by an increase in
the governments primary surplus, that is, the government budget surplus
excluding net interest payments. Fourth, the growth performance of the
economy under infation targeting did not improve much when compared
to the previous period of exchange rate targeting, and this happened even
though the international environment was much more favorable for eco-
nomic growth in 19992006 than in 199498.
6
To facilitate the analysis, we
look at each one of these issues separately below.
Starting with infation itself, Table 7.1 presents the annual infation
targets set by the CMN and the efective rates of infation and exchange
rate variation in 19992006. If we consider the whole period the infation
targets were met in fve out of eight years, which on a frst look seems to be
a very good starting performance of infation targeting in Brazil. However,
when we look closer the real story cannot be considered that successful
because the infation targets were frequently changed according to the
shocks that hit the Brazilian economy and, most importantly, disinfation
was obtained mostly through exchange rate appreciation. In fact, if we
exclude 1999 from the analysis, the infation targets were met only when
the exchange rate appreciated in nominal terms. To illustrate this point,
Figure 7.1 shows that in 2000, 2004, 2005 and 2006 the target was met and
the exchange rate appreciated, in 2001 and 2002 the target was not met
and the exchange rate depreciated, and in 2003 the target was not met and
the exchange rate appreciated.
Now, before we move to the real interest rate, let us make a brief pause
to present some of the history behind the numbers shown in Table 7.1.
First, since the Brazilian infation targeting regime started immediately
after a currency crisis, there was a 7.2 percentage point (pp) increase in
infation in 1999 when compared to 1998.
7
Despite such acceleration,
the government infation target for 1999 was met for two idiosyncratic
reasons: frst, the 1999 target was set in June of that year, when the gov-
ernment authorities already knew half of the annual infation rate of 1999
and, second, the target for 1999 was not ambitiously low in order not to
undermine the credibility of the new policy regime.
8
Continuing with our infation story, in 2000 the Brazilian exchange
rate stabilized at a new and higher level, both in real and nominal terms,
and this helped the BCB to reduce infation and meet the infation target
set for that year. Then, in 2001, Brazil was hit by two major shocks that
made the 6.0 percent infation target unfeasible. First, because of an
unexpected drought, there was a shortage in the supply of electricity from
hydro- electric plants during 2001 and, since most of Brazilian electrical
power comes from such a source, the adverse energy shock pushed infa-
tion upwards. Second, and most important, the 2001 currency crisis in
Argentina resulted in capital fight from and exchange rate depreciation
in Brazil, pushing the domestic price of tradable goods up. Altogether
143
T
a
b
l
e

7
.
1

I
n
f

a
t
i
o
n
,

e
x
c
h
a
n
g
e

r
a
t
e
s
,

G
D
P

g
r
o
w
t
h

a
n
d

i
n
t
e
r
e
s
t

r
a
t
e
s

i
n

B
r
a
z
i
l
1
9
9
8

%
1
9
9
9

%
2
0
0
0

%
2
0
0
1

%
2
0
0
2

%
2
0
0
3

%
2
0
0
4

%
2
0
0
5

%
2
0
0
6

%
E
f

e
c
t
i
v
e

i
n
f

a
t
i
o
n

r
a
t
e

E
n
d

o
f

t
h
e

p
e
r
i
o
d

1
.
7
8
.
9
6
.
0
7
.
7
1
2
.
5
9
.
3
7
.
6
5
.
7
3
.
1

A
v
e
r
a
g
e

o
f

t
h
e

p
e
r
i
o
d

3
.
2
4
.
9
7
.
0
6
.
8
8
.
4
1
4
.
7
6
.
6
6
.
9
4
.
2
T
a
r
g
e
t

i
n
f

a
t
i
o
n

r
a
t
e

(
e
n
d

o
f

t
h
e

p
e
r
i
o
d
)

I
n
i
t
i
a
l

t
a
r
g
e
t
n
.
a
.
8
.
0
0
6
.
0
0
4
.
0
0
3
.
5
0
4
.
0
0
5
.
5
0
4
.
5
0
4
.
5
0

C
e
i
l
i
n
g
n
.
a
.
1
0
.
0
0
8
.
0
0
6
.
0
0
5
.
5
0
6
.
5
0
8
.
0
0
7
.
0
0
6
.
5
0

F
l
o
o
r
n
.
a
.
6
.
0
0
4
.
0
0
2
.
0
0
1
.
5
0
1
.
5
0
3
.
0
0
2
.
0
0
2
.
5
0

W
a
s

t
h
e

t
a
r
g
e
t

r
e
v
i
s
e
d
?
n
.
a
.
N
O
N
O
N
O
N
O
Y
E
S
N
O
Y
E
S
N
O

R
e
v
i
s
e
d

t
a
r
g
e
t
8
.
5
5
.
1

W
a
s

t
h
e

t
a
r
g
e
t

m
e
t
?
n
.
a
.
Y
E
S
Y
E
S
N
O
N
O
N
O
Y
E
S
Y
E
S
Y
E
S
N
o
m
i
n
a
l

e
x
c
h
a
n
g
e

r
a
t
e

v
a
r
i
a
t
i
o
n

E
n
d

o
f

t
h
e

p
e
r
i
o
d

8
.
3
3
6
.
7
2
.
2
1
6
.
0
3
5
.
1
1
5
.
3
4
.
8
2
0
.
5
1
.
3

A
v
e
r
a
g
e

o
f

t
h
e

p
e
r
i
o
d

7
.
7
5
6
.
4
0
.
9
2
8
.
4
2
4
.
3
5
.
4
4
.
9
1
6
.
8
1
0
.
6
G
D
P

g
r
o
w
t
h

r
a
t
e

0
.
0
0
.
3
4
.
3
1
.
3
2
.
7
1
.
1
5
.
7
2
.
9
3
.
7
B
a
s
e

i
n
t
e
r
e
s
t

r
a
t
e

N
o
m
i
n
a
l

b
a
s
e

i
n
t
e
r
e
s
t

r
a
t
e
2
8
.
8
2
5
.
6
1
7
.
4
1
7
.
3
1
9
.
2
2
3
.
3
1
6
.
2
1
9
.
1
1
5
.
1

R
e
a
l

b
a
s
e

i
n
t
e
r
e
s
t

r
a
t
e
2
6
.
6
1
5
.
3
1
0
.
8
8
.
9
6
.
0
1
2
.
8
8
.
0
1
2
.
7
1
1
.
6
N
o
t
e
:

T
h
e

i
n
f

a
t
i
o
n

t
a
r
g
e
t
s

a
r
e

t
h
e

t
a
r
g
e
t
s

s
e
t

a
t

l
e
a
s
t

o
n
e

y
e
a
r

i
n

a
d
v
a
n
c
e
.

T
h
e

e
x
c
e
p
t
i
o
n

i
s

1
9
9
9
,

w
h
e
n

t
h
e

t
a
r
g
e
t

w
a
s

s
e
t

i
n

J
u
n
e

o
f

t
h
a
t

y
e
a
r
.

T
h
e

r
e
v
i
s
e
d

t
a
r
g
e
t

i
s

t
h
e

t
a
r
g
e
t

s
e
t

i
n

t
h
e

c
o
r
r
e
s
p
o
n
d
i
n
g

y
e
a
r
.

T
h
e

e
x
c
h
a
n
g
e

r
a
t
e

v
a
r
i
a
t
i
o
n

r
e
f
e
r
s

t
o

t
h
e

B
r
a
z
i
l
i
a
n

r
e
a
l
/
U
S

d
o
l
l
a
r

e
x
c
h
a
n
g
e

r
a
t
e
,

a
n
d

t
h
e

b
a
s
e

i
n
t
e
r
e
s
t

r
a
t
e

i
s

t
h
e

S
E
L
I
C

i
n
t
e
r
e
s
t

r
a
t
e
,

s
e
t

b
y

t
h
e

B
C
B
,

a
n
d

c
u
m
u
l
a
t
e
d

o
v
e
r

t
h
e

y
e
a
r
.

T
h
e

r
e
a
l

i
n
t
e
r
e
s
t

r
a
t
e

w
a
s

o
b
t
a
i
n
e
d

b
y

d
e
f

a
t
i
n
g

t
h
e

n
o
m
i
n
a
l

i
n
t
e
r
e
s
t

r
a
t
e

b
y

t
h
e

i
n
f

a
t
i
o
n

r
a
t
e

c
u
m
u
l
a
t
e
d

o
v
e
r

t
h
e

y
e
a
r
,

a
c
c
o
r
d
i
n
g

t
o

t
h
e

I
P
C
A

i
n
d
e
x
.
S
o
u
r
c
e
:

B
r
a
z
i
l
i
a
n

C
e
n
t
r
a
l

B
a
n
k

(
h
t
t
p
:
/
/
w
w
w
.
b
c
b
.
g
o
v
.
b
r
)

a
n
d

a
u
t
h
o
r
s

c
a
l
c
u
l
a
t
i
o
n
.
the fnal result of these two adverse shocks was a 1.7 pp increase in the
Brazilian infation rate in 2001, when compared to 2000.
9
The macroeconomic turmoil of 2001 was followed by more instability in
2002, this time because of the Brazilian political cycle and the speculative
international capital fows associated with it. More specifcally, in 2002
Brazil had its presidential election and the looming victory of a leftist can-
didate, Lula, resulted in massive capital outfows and an unprecedented
cut in Brazils access to foreign lines of credit. The exchange rate shot up
as a result of such a move, reaching its highest level in real terms since
the debt crisis of the early 1980s. The Brazilian infation rate followed
soon after and reached double-digit levels at the end of 2002. In fact, the
exchange rate depreciation and infation acceleration were so sharp and
fast that they even made the Brazilian monthly base interest rate tempor-
arily negative at the end of 2002. Altogether the infation rate increased in
another 4.8 pp in 2002, when compared to 2001.
Lula was elected president in 2002 and his frst term in of ce started
with a sharp monetary and fscal crunch in order to reduce infation
and restore the countrys access to foreign fnance. Lulas move and the
very own excesses of the speculative attack to the Brazilian currency in
2002 quickly resulted in a return to normality, that is, a return to capital
25
20
15
10
5
0
5
10
15
20
25
30
35
40
1998 1999 2000 2001* 2002* 2003* 2004 2005 2006
P
e
r
c
e
n
t

(
%
)
Exchange rate variation
Inflation
Note: * Target not met. The infation rate is the variation in the IPCA index, December to
December, and the exchange rate variation the rate of change in the Brazilian real/US dollar
exchange rate, also December to December.
Source: Brazilian Central Bank (http://www.bcb.gov.br).
Figure 7.1 Consumer infation rate and exchange rate variation in Brazil
infows and exchange rate appreciation. Infation decelerated to one-digit
levels in 2003, with a reduction of 3.2 pp in relation to 2002. Despite this
disinfation, the infation target for 2003 was not met, even after the Lula
administration revised it upwards because of the exchange rate deprecia-
tion of 2002.
10
After the failure in 200103, the Brazilian infation target regime started
to perform well again in 200406. Most of this success can be credited to
the exchange rate appreciation during these three years, which in its turn
resulted from both the high domestic interest rates of Brazil and the favor-
able international trade and fnancial conditions in the rest of the world.
More specifcally, on the one hand, the BCB continued to practice high
real interest rates in 200406 and this resulted in a gradual appreciation
of the Brazilian exchange rate, in nominal and in real terms, as shown
in Figures 7.2 and 7.3.
11
On the other hand, the exchange rate apprecia-
tion of 200406 was also driven by the boom in Brazilian exports, pulled
by the rise in the world demand for commodities, which in their turn
increased the countrys trade and current account surpluses. It should be
noted that the improvement in the Brazilian balance of payments hap-
pened despite the exchange rate appreciation for two reasons: frst, the
increase in international prices ofset the reduction in the exchange rate
for many sectors and, second, the depreciation in 2001 and 2002 was so
1.0
1.5
2.0
2.5
3.0
3.5
4.0
J
a
n
/
9
8
J
u
l
/
9
8
J
a
n
/
9
9
J
u
l
/
9
9
J
a
n
/
0
0
J
u
l
/
0
0
J
a
n
/
0
1
J
u
l
/
0
1
J
a
n
/
0
2
J
u
l
/
0
2
J
a
n
/
0
3
J
u
l
/
0
3
J
a
n
/
0
4
J
u
l
/
0
4
J
a
n
/
0
5
J
u
l
/
0
5
J
a
n
/
0
6
J
u
l
/
0
6
J
a
n
/
0
7
J
u
l
/
0
7
Source: IPEADATA (http://www.ipeadata.gov.br).
Figure 7.2 Nominal exchange rate in Brazil Brazilian real per US dollar
intense that only at the end of 2005 did the Brazilian real exchange rate
return to the level verifed in 1999.
Altogether the infation targeting performance of Brazil in 19992006
can be divided into three phases: (1) implementation, in 19992000, when
the favorable international conditions and the modest infation targets
proved to be very successful; (2) crisis, in 200103, when a combination of
adverse supply shocks and fnancial crises made the ambitiously low infa-
tion targets unfeasible; and (3) consolidation, in 200406, when the very
favorable international conditions and the high real domestic interest rates
resulted in a quick reduction in infation. Let us now turn to the evolution
of the Brazilian base real interest rate throughout this process.
Figure 7.4 shows the real base interest rate of Brazil since 1994 in order
to allow a comparison between the exchange rate targeting and the infa-
tion targeting regimes. The frst thing that draws ones attention is the very
high real interest rate of Brazil during its exchange rate targeting regime
and the substantial reduction brought by infation targeting. In numbers,
the average annual real base interest rate of Brazil was 21.9 percent in
199498, and 10.7 percent in 19992006.
12
Despite such a substantial
reduction in the real interest rate, it should also be noted that during the
infation targeting regime the Brazilian rate continued to be extremely high
60
80
100
120
140
160
180
J
a
n
/
9
8
J
u
l
/
9
8
J
a
n
/
9
9
J
u
l
/
9
9
J
a
n
/
0
0
J
u
l
/
0
0
J
a
n
/
0
1
J
u
l
/
0
1
J
a
n
/
0
2
J
u
l
/
0
2
J
a
n
/
0
3
J
u
l
/
0
3
J
a
n
/
0
4
J
u
l
/
0
4
J
a
n
/
0
5
J
u
l
/
0
5
J
a
n
/
0
6
J
u
l
/
0
6
J
a
n
/
0
7
J
u
l
/
0
7
Source: Central Bank of Brazil (http://www.bcb.gov.br).
Figure 7.3 Index of the real efective exchange rate of Brazil (1992 5
100)
by international standards and, more important, the Brazilian real interest
rate was extremely high when compared to the average GDP growth rate
of the economy in 19992006, that is, 2.7 percent.
Focusing our analysis on 19992006, Figure 7.4 shows that infation
targeting started with very high real interest rates in Brazil. As we men-
tioned earlier, the reason for this was the Brazilian 1999 currency crisis,
which led the BCB to increase interest rates abruptly and substantially to
stop the capital outfows from the country. Then, after infation targeting
was formally introduced, in mid 1999, the real base interest rate started to
fall until it reached 9.5 percent at the end of 2000. After that the real inter-
est rate fuctuated between 8 percent and 10 percent until the end of 2002,
when the devaluation of the real associated with Lulas election and the
subsequent increase in infation made BCBs base interest rate temporarily
negative. Normality was quickly restored at the beginning of 2003, when
the Lula administration raised the real interest rate of the economy to
14.3 percent in order to stop devaluation and decelerate infation. When it
became clear that such an efort was successful, in mid 2003 the real base
interest rate started to fall gradually. The interest rate cuts persisted until
the end of 2004, when the acceleration of GDP growth made the BCB
fear that the markets expectation for infation would accelerate above the
0
5
10
15
20
P
e
r
c
e
n
t

(
%
)
25
30
1
9
9
4
-
1
1
9
9
5
-
1
1
9
9
6
-
1
1
9
9
7
-
1
1
9
9
8
-
1
1
9
9
9
-
1
2
0
0
0
-
1
2
0
0
1
-
1
2
0
0
2
-
1
2
0
0
3
-
1
2
0
0
4
-
1
2
0
0
5
-
1
2
0
0
6
-
1
2
0
0
7
-
1
EXCHANGE RATE
TARGETING
INFLATION
TARGETING
Note: The real interest rate is the SELIC interest rate, set by the BCB, cumulated in the
past 12 months, defated by the consumer infation rate also cumulated in the past 12 months,
measured by the IPCA index.
Source: Central Bank of Brazil (http://www.bcb.gov.br) and authors calculation.
Figure 7.4 Real annual base interest rate in Brazil
target set for 2005.
13
The BCBs response to such an expectational threat
was another interest rate hike, which made the Brazilian real interest rate
reach 13.4 percent in mid 2006.
The increase in the interest rate in 2005 resulted in a sharp appreciation
of the real and a deceleration of GDP growth. By the end of 2005 it was
clear that the BCB had exaggerated in its response to the increase in the
level of economic activity at the end of 2004, as well as that the huge dis-
crepancy between Brazilian and international interest rates could compro-
mise the countrys fscal stability. The BCBs response was another round
of gradual interest rate cuts, which made the Brazilian real interest fall to
just 11.6 percent at the end of 2006, that is, a still abnormally high real
interest rate by international standards and well above the GDP growth
rate of the economy in that year, that is, 3.7 percent.
Moving to fscal policy, the main side efect of the Brazilian high real
interest rates was the expansionary impact of net interest payments on
public expenditures and debt. Table 7.2 presents the main fscal numbers
for the period, from which we can identify two distinct phases in the evolu-
tion of the net public debt of Brazil. First, in the frst four years of infation
targeting most of the government net debt was directly or indirectly tied
to the exchange rate because the Brazilian government was a net debtor
in foreign currency and most of its domestic debt was formally indexed to
either the exchange rate or the interest rate.
14
So, in the wake of the 1999
currency crisis and the interest rate hike that followed it, the net interest
payments by the Brazilian public sector shot up to 12.5 percent of GDP,
and the ratio of its net public debt to GDP increased in 5.6 pp in just one
year. Then, as the exchange rate stabilized at a new and higher level, the net
interest payments by the public sector fell to less than 10 percent of GDP
in 200001, but the net domestic debt of the government continued to rise
gradually in relation to GDP because of the huge diference between the
real interest rate, on the one side, and the growth rate of the economy, on
the other side. The frst phase of debt dynamics ended in 2002, when Brazil
experienced another sharp and abrupt exchange rate depreciation, the net
interest payments by the public sector shot up to 13 percent of GDP, and
the net public debt of the country rose to 50.5 percent of its GDP.
The second phase in the dynamics of public debt began in 2003, when
the exchange rate appreciation started to reduce the government net inter-
est payments. The ratio of net public debt to GDP followed the same path
in 200406, when the Brazilian government was able to repay most of
its foreign debt and accumulate a huge amount of foreign reserves. This
movement was obviously facilitated by the favorable international trade
and fnancial conditions of the period and, at the end of 2006, the Brazilian
government became a net creditor in foreign currency. In contrast, on
149
T
a
b
l
e

7
.
2

F
i
s
c
a
l

p
o
l
i
c
y

i
n

B
r
a
z
i
l
1
9
9
8
1
9
9
9
2
0
0
0
2
0
0
1
2
0
0
2
2
0
0
3
2
0
0
4
2
0
0
5
2
0
0
6
P
u
b
l
i
c

d
e
b
t

i
n

%

o
f

G
D
P

N
e
t

d
o
m
e
s
t
i
c

d
e
b
t
3
3
.
2
3
5
.
2
3
6
.
5
3
8
.
9
3
7
.
5
4
1
.
7
4
0
.
2
4
4
.
1
4
7
.
6

M
o
n
e
t
a
r
y

b
a
s
e
4
.
0
4
.
2
3
.
9
3
.
9
4
.
2
4
.
2
4
.
4
4
.
7
5
.
1

N
o
n
-
m
o
n
e
t
a
r
y

d
e
b
t
2
9
.
2
3
1
.
0
3
2
.
7
3
4
.
9
3
3
.
3
3
7
.
5
3
5
.
8
3
9
.
4
4
2
.
5

N
e
t

f
o
r
e
i
g
n

d
e
b
t
5
.
8
9
.
4
9
.
0
9
.
6
1
3
.
0
1
0
.
7
6
.
8
2
.
3
2
.
7

N
e
t

d
o
m
e
s
t
i
c

a
n
d

f
o
r
e
i
g
n

d
e
b
t
3
8
.
9
4
4
.
5
4
5
.
5
4
8
.
4
5
0
.
5
5
2
.
4
4
7
.
0
4
6
.
5
4
4
.
9
B
u
d
g
e
t

b
a
l
a
n
c
e

o
f

t
h
e

p
u
b
l
i
c

s
e
c
t
o
r

i
n

%

o
f

G
D
P

P
r
i
m
a
r
y

d
e
f

c
i
t
0
.
0
3
.
0
3
.
3
3
.
4
3
.
7
3
.
9
4
.
2
4
.
4
3
.
9

N
e
t

i
n
t
e
r
e
s
t

p
a
y
m
e
n
t
s
7
.
4
1
2
.
5
7
.
4
8
.
2
1
3
.
0
7
.
2
6
.
5
7
.
1
6
.
8

O
n

f
o
r
e
i
g
n

d
e
b
t
0
.
3
0
.
9
0
.
8
1
.
1
1
.
2
1
.
1
0
.
9
0
.
7
0
.
3

O
n

d
o
m
e
s
t
i
c

d
e
b
t
7
.
1
1
1
.
5
6
.
6
7
.
1
1
1
.
8
6
.
1
5
.
6
6
.
5
6
.
5

N
o
m
i
n
a
l

d
e
f

c
i
t
7
.
4
9
.
5
4
.
1
4
.
8
9
.
3
3
.
3
2
.
3
2
.
8
2
.
9
P
r
i
m
a
r
y

b
u
d
g
e
t

o
f

t
h
e

f
e
d
e
r
a
l

g
o
v
e
r
n
m
e
n
t

i
n

%

o
f

G
D
P

R
e
v
e
n
u
e
s
1
5
.
8
1
6
.
4
1
6
.
5
1
7
.
2
1
7
.
9
1
7
.
4
1
8
.
1
1
8
.
8
1
9
.
4

E
x
p
e
n
d
i
t
u
r
e
s
1
5
.
1
1
4
.
5
1
4
.
7
1
5
.
6
1
5
.
7
1
5
.
1
1
5
.
6
1
6
.
4
1
7
.
2

N
e
t

e
r
r
o
r

a
n
d

o
m
i
s
s
i
o
n
s
0
.
3
0
.
2
0
.
0
0
.
0
0
.
0
0
.
0
0
.
2
0
.
1
0
.
1

B
a
l
a
n
c
e
0
.
5
2
.
1
1
.
7
1
.
7
2
.
2
2
.
3
2
.
7
2
.
6
2
.
2
N
o
t
e
:

T
h
e

p
r
i
m
a
r
y

d
e
f

c
i
t

i
s

t
h
e

g
o
v
e
r
n
m
e
n
t

b
u
d
g
e
t

d
e
f

c
i
t

e
x
c
l
u
d
i
n
g

n
e
t

i
n
t
e
r
e
s
t

p
a
y
m
e
n
t
s
.

T
h
e

n
o
m
i
n
a
l

d
e
f

c
i
t

i
s

t
h
e

t
o
t
a
l

b
u
d
g
e
t

d
e
f

c
i
t
,

t
h
a
t

i
s
,

t
h
e

p
r
i
m
a
r
y

d
e
f

c
i
t

p
l
u
s

n
e
t

i
n
t
e
r
e
s
t

p
a
y
m
e
n
t
s

b
y

t
h
e

p
u
b
l
i
c

s
e
c
t
o
r
.

T
h
e

n
u
m
b
e
r
s

f
o
r

t
h
e

w
h
o
l
e

B
r
a
z
i
l
i
a
n

p
u
b
l
i
c

s
e
c
t
o
r

r
e
f
e
r

t
o

t
h
e

f
e
d
e
r
a
l
,

s
t
a
t
e

a
n
d

m
u
n
i
c
i
p
a
l

g
o
v
e
r
n
m
e
n
t
s
,

a
n
d

t
h
e

s
t
a
t
e
-
o
w
n
e
d

e
n
t
e
r
p
r
i
s
e
s
.
S
o
u
r
c
e
:

B
r
a
z
i
l
i
a
n

C
e
n
t
r
a
l

B
a
n
k

(
h
t
t
p
:
/
/
w
w
w
.
b
c
b
.
g
o
v
.
b
r
)

a
n
d

a
u
t
h
o
r
s

c
a
l
c
u
l
a
t
i
o
n
.
the domestic front, the high real interest rate continued to increase the
Brazilian net public domestic debt, which reached the unprecedented high
level of 47.6 percent of GDP at the end of the period. Altogether, we can
say that in this second phase of debt dynamics infation targeting was char-
acterized by an increase in the ratio of net public domestic debt to GDP
and a substitution of domestic debt for foreign debt.
To complete our fscal outlook, Table 7.2 also shows the evolution of
the federal government primary surplus and nominal budget defcit in the
infation targeting period.
15
Compared to the last year of exchange rate
targeting, most of the infation targeting period was characterized by a
gradual increase in the federal primary surplus.
16
Most of this increase
happened in 1999, when the Brazilian federal government cut its spend-
ing and increased its revenues in proportion to GDP in order to avoid an
explosive increase in its public debt. After that the federal governments
revenues and expenditures tended to grow together in relation to GDP,
which in its turn had a small positive impact on economic growth through
the balanced-budget multiplier. The second main fscal crunch happened
in 2003, when the federal government once more cut its spending in pro-
portion to GDP in order to avoid an explosive increase in the countrys
net public debt. However, diferently from what happened in 1999, the
federal government revenues also fell in proportion to GDP in 2003,
which ended up stabilizing the primary surplus in terms of GDP.
17
The
federal government revenues and expenditures resumed growing faster
than GDP in 200406, and fscal policy became expansionary at the end
of the period.
18
Finally, considering GDP growth, the infation targeting period had a
slower rate of expansion than the exchange rate period, but a smaller vola-
tility as well. To illustrate this point, Figure 7.5 shows the moving-average
annual GDP growth rate of Brazil in 19942006. In numbers, on the one
hand, the average GDP growth rate was 3.8 percent during the exchange
rate targeting period, against 2.7 percent during the infation targeting
period. On the other hand, the maximum and minimum growth rates
during exchange rate targeting were 8.5 percent and zero, respectively,
whereas during infation targeting the maximum and minimum rates were
5.7 percent and minus 0.8 percent, respectively. Finally, it should also be
noted that the trend growth rate was stationary or declining during the
exchange rate targeting period, but rising during the infation targeting
period.
Figure 7.5 also reveals clearly the sequence of booms and busts expe-
rienced by the Brazilian economy since 1994. The downturns are usually
associated with adverse shocks, and the upturns with expansionary
macro economic policy. More specifcally, the period starts with the boom
brought by infation reduction in 1994. This boom ended in 1995, when
the Brazilian economy sufered the contagion efect from the Mexican
crisis and the Brazilian government adopted restrictive macroeconomic
measures to avoid exchange rate depreciation and infation acceleration.
The second boom started after the Mexican crisis was absorbed by Brazil
and lasted until 1997. Then, in 199799, the Brazilian economy experi-
enced another growth deceleration because of the contagion efects from
the East Asian crises of 1997 and the Russian crisis of 1998, and because
of the recessive impact of their very own Brazilian crisis of 1999. The
economy resumed growth in 2000, but the expansion was quickly cur-
tailed by the Argentine crisis and the Brazilian energy rationing of 2001.
The Brazilian economy started another recovery in 2002, but the expan-
sion was short-lived because of the speculative attack to the Brazilian
currency during the presidential elections of that year. The exchange rate
depreciation and the restrictive macroeconomic measures adopted by the
Brazilian government pushed economic growth down in 2003, and only
in the beginning of 2004 did the economy start to recover. However,
diferently from all of the previous episodes, the 2004 expansion was
cut for domestic reasons, namely the fear of the BCB that the economy
was overheating at the end of that year. As we saw earlier, interest rates
were raised and economic growth decelerated in approximately two
2
0
2
4
6
8
10
1
9
9
4
-
1
1
9
9
5
-
1
1
9
9
6
-
1
1
9
9
7
-
1
1
9
9
8
-
1
1
9
9
9
-
1
2
0
0
0
-
1
2
0
0
1
-
1
2
0
0
2
-
1
2
0
0
3
-
1
2
0
0
4
-
1
2
0
0
5
-
1
2
0
0
6
-
1
2
0
0
7
-
1
INFLATION TARGETING EXCHANGE RATE
TARGETING
P
e
r
c
e
n
t

(
%
)
Note: The growth rate is the rate of change of the average GDP during period t through t-3,
in relation to the average GDP during period t-4 through t-7.
Source: IBGE (http://www.ibge.gov.br) and authors calculation.
Figure 7.5 Moving-average of the annual GDP growth rate of Brazil
percentage points in 2005, but it still remained well above the previous
troughs. Then, at the end of the period under analysis, in the second half
of 2006, the growth rate of the Brazilian economy started once again to
accelerate.
19
7.3 INFLATION TARGETING, REAL EXCHANGE
RATES AND POTENTIAL OUTPUT
Whether or not Brazil started a new growth cycle in recent years is an
open question that depends, among other things, on the management of
monetary policy. From the recent history of infation targeting in Brazil,
there are two main macroeconomic challenges to the Brazilian authori-
ties that will eventually have to be addressed. First, so far the success of
infation targeting depended heavily on a favorable behavior of exchange
rate and this cannot go on indefnitely. By defnition real exchange rate
appreciation cannot go on forever, otherwise the countrys currency will
become infnitely expensive and, more important, an excessively appreci-
ated real exchange rate can reduce the growth prospects of the economy.
Second, so far infation targeting has been characterized by a slow GDP
growth, both in comparison to Brazils previous history and to the growth
rate of the rest of the world. Since also by defnition any growth acceler-
ation results in a temporary increase in the level of economic activity, it
tends to create infationary pressures and requires compensatory meas-
ures by monetary policy. However, if monetary policy is too conservative
in setting a low infation target and a fast speed of convergence to it, the
economy may end up locked in a slow-growth equilibrium where the BCB
kills any growth acceleration for fears of rising infation. To complete
our analysis of infation targeting in Brazil, let us look at these two chal-
lenges in more detail.
First, regarding exchange rate, stable infation requires a stable real
exchange rate, but the level of the exchange rate is not determined a priori.
There are multiple equilibria and, therefore, the real exchange rate can
be stabilized at a competitive or an uncompetitive level. Thus, when the
infation target is ambitiously low and the speed of convergence to it too
fast, there will be a tendency to real exchange rate appreciation, which in
its turn may end up increasing the fnancial fragility of the economy in the
medium run, especially if appreciation is sustained by speculative capital
infows. The usual logical sequence here is that high real interest rate leads
to capital infows and exchange rate appreciation, which in their turn leads
to a reduction in the trade and current account balance of the economy,
which in their turn make the economy more vulnerable to shifts in the
international liquidity. So, when infation targeting is too ambitious and
fnanced by speculative capital infows, the result tends to be a sequence of
booms and busts, during which GDP growth fuctuates around a low rate.
Another possibility is that infation targeting stabilizes the real exchange
rate at a new low level without compromising the balance of payments of
the economy.
20
However, even in this case ambitious infation targeting
can still result in a slow growth rate because an appreciated real exchange
rate does not stimulate the development of the domestic tradable sector.
In other words, productivity growth is usually faster in the tradable sector
and, therefore, an appreciated real exchange rate may end up reducing the
overall growth rate of the economy.
21
The natural solution to the exchange rate challenge outlined above is to
combine infation targeting with an asymmetric dirty foating that main-
tains the countrys real exchange rate stable and competitive in the long
run. The dirty foating will have to be asymmetric because the instruments
to combat appreciation are diferent than the ones use to combat depreci-
ation in order to keep the real exchange rate competitive. In other words,
in face of depreciation the government should use its traditional restrictive
macroeconomic tools to stop the process and avoid an increase in infa-
tion, without selling much of its foreign reserves. However, in the face
of appreciation the government should buy the foreign currency to slow
down the process, but with no open compromise with a specifc exchange
rate. The auxiliary measures to combat appreciation are obviously to
increase imports and reduce domestic interest rates so that, at the end of
the day, the asymmetric dirty foating ends up creating a sliding foor for
the exchange rate.
22
Second, regarding growth, infation targeting is usually aimed at stabil-
izing economic growth at its potential level, so that it eliminates excess
demand pressures that tend to increase infation, and insuf cient demand
pressures that do the opposite. The logical form of the orthodox stand-
point is that the long-run growth rate of the economy is given from outside
of macroeconomic issues, by preferences and institutions, and the only
thing a responsible central bank can do is to make the economy grow at its
potential rate in the long run. So, according to the orthodox view of eco-
nomic policy, attempts to grow faster than such supply-determined limits
end up increasing infation, with either no or negative permanent real
efects on the economy. The main problem with such a view is that it fails
to recognize that the potential output of an economy is an endogenous
variable and, therefore, it can and usually is afected by macroeconomic
policy itself, including monetary policy.
More specifcally, potential output is an unobservable variable that is
estimated from the past and expected behavior of the economy.
23
Since
expectations are usually heavily infuenced by the recent past, the estimates
of potential output tend to extrapolate current trends to the near future
and, in this way, they may create a self-fulflling monetary policy. An
example helps to illustrate the point. Suppose that a conservative central
bank estimates a slow potential growth rate for the economy and, based
on such a pessimistic estimate, it combats a growth acceleration for fearing
that it will increase infation above the pre-specifed target. In doing so
the very own actions of the central bank reduce the efective growth rate
of the economy, especially of investment, and when potential output is
re-estimated again ex post, the data will confrm that the central bank was
right not necessarily because its initial estimates were right, but because
the growth deceleration produced by the central bank is automatically
translated in a lower estimate of the growth potential of the economy.
24

In short, since the estimates of potential output are highly uncertain at the
end of the sample and continuously revised every time a new observation
is incorporated into the analysis, the actions of the central bank may seem
to be correct ex post because its very own actions produced the scenario
that justifed it ex ante.
The endogeneity of potential output at the end of the sample is espec-
ially dangerous for an economy, as Brazil, that aims to accelerate its
growth rate. The danger here is that because the growth rate in the recent
past was low, any strong growth acceleration tends to generate an appar-
ent excessive output gap in the short run. So, if the central bank stops the
process too soon, there wont be enough time for the estimates of poten-
tial output to pick the structural change in the growth prospects of the
economy. The fnal result is either that the economy never takes of, or that
it takes of very slowly. On the other hand, the endogeneity of potential
output does not obviously mean that authorities live in a sort of feld of
dreams, in which, if you believe, growth will come.
25
There are objective
limits to the growth rate of the economy that cannot be avoided by opti-
mistic expectations like, for instance, the maximum supply of energy, the
stock of foreign reserves and a zero unemployment rate.
The endogeneity of potential output only means that there may be more
than one equilibrium position for the GDP growth rate and, therefore,
the speed of convergence to the infation target is also a relevant variable
for economic growth. As usually happens in economics, central banks
face a trade-of in this area: on the one hand, quick disinfation may lock
the economy in a slow-growth equilibrium, but, on the other hand, slow
disinfation may lock the economy in a high-infation equilibrium. Even
though this trade-of has yet to be emphasized by modern macroeconomic
theory, it is an unavoidable matter for policy makers who deal with the
real world.
7.4 CONCLUSION
Infation targeting represents an improvement in relation to the previ-
ous monetary policy regime adopted by Brazil, but it can and should be
improved in order to increase the growth rate of the economy. In general
terms the main conclusions from the previous sections can be summarized
in ten points. (1) Infation targeting managed to reduce infation in Brazil
after its 1999 and 2002 currency crises, with a substantial help of exchange
rate appreciation. (2) Economic growth was slower under infation target-
ing than under exchange rate targeting, but with a smaller volatility and
with an apparent upward trend. (3) Infation targeting reduced the real
interest rate of the economy, which nevertheless remained well above
international standards and more than three times higher than the GDP
growth rate of Brazil in 19992006. (4) The high real interest rate required
a substantial increase in fscal austerity by the Brazilian government, but
so far this has not been suf cient to stop the increase in the ratio of net
public domestic debt to GDP. (5) The high domestic real interest rates and
the favorable international trade and fnancial conditions in the rest of the
world allowed the Brazilian government to accumulate foreign reserves,
repay most of its foreign debt and reduce its dependence of foreign capital
in 200306. (6) Infation targeting can be combined with an asymmetric
dirty foating regime, in which the central bank combats exchange rate
depreciation with restrictive monetary policy, and accumulates foreign
reserves to slow down appreciation. (7) The asymmetric dirty foating
should aim at a stable competitive real exchange rate, in order to promote
the fast growth of the domestic tradable sector, and in this way push the
countrys exports and imports up without increasing the foreign fnancial
fragility of the economy. (8) Aggregate estimates of potential output are
not good guides for the demand pressures on infation in moments of
structural growth acceleration or deceleration. (9) Because of the endo-
geneity of potential output at the end of the sample, infation targeting
should be done with moderation in order to avoid a self-fulflling mon-
etary policy that locks the economy in a slow-growth equilibrium. (10) A
fast convergence to a low infation target may produce a permanently low
growth rate, but a slow convergence to a moderate infation target can
produce a permanently high infation.
NOTES
1. In this chapter we defne the exchange rate as the domestic price of foreign currency, so
that a depreciation of the exchange rate means an increase in the exchange rate, that is,
a devaluation of the domestic currency. By analogy, appreciation means a reduction in
the exchange rate, that is, a revaluation of the domestic currency.
2. For instance, according to the estimates for that time (Belaisch, 2003), 23 percent of
exchange rate variations tended to pass through to consumer prices (measured by the
IPCA index) in the long run, whereas the pass-through to the general price level (meas-
ured by the IGPDI index) was 71 percent.
3. For a more technical analysis of infation targeting in Brazil, see Bogdanski et al. (2000),
Minella et al. (2003), Tombini and Alves (2006) and Bevilaqua et al. (2007).
4. According to the Brazilian Presidential Decree 3088, of 21 June 1999, it is up to the
BCB to execute the necessary policies to meet the specifed targets.
5. In 19992005 the Copom met every month to determine interest rates. Starting in 2006
the Copom has been meeting eight times a year, that is, one meeting approximately every
six weeks. If necessary, an extraordinary meeting can be called by the president of the
BCB.
6. According to the IMF estimates, the average world GDP growth rate was 3.7 percent
in 199498 and 4.2 percent in 199906. In its turn, the infation targeting period can be
divided in two phases regarding the world GDP growth rate, that is, slow-growth, in
199906, when the average world annual GDP growth rate was 3.5 percent, and fast
growth, in 200306, when the average annual growth rate was 4.9 percent.
7. Unless stated otherwise, all numbers are annual fgures.
8. The infation rate of 1999 was high when compared to 1998, but modest when com-
pared to the magnitude of the exchange rate depreciation verifed in 1999. As we will
see later in this section, most of the small pass-through of the exchange rate to domestic
prices in 1999 can be credited to the high real interest rate and the abrupt fscal crunch
practiced by the Brazilian government during that year.
9. As we will also see later in this section, the high real interest rate practiced by the BCB
and the very own recessive impact of the two supply shocks on the level of economic
activity helped to reduce the magnitude of infation acceleration.
10. One of the frst measures of the Lula administration was to increase interest rates, cut
public spending and increase the infation target for 2003, from the 4.0 percent set by the
previous administration to 8.5 percent because of the adverse electoral shock of 2002. A
complete analysis of Lulas economic policy is beyond the scope of this chapter. The inter-
ested reader can fnd more information in Barbosa-Filho (2007) and Arestis et al. (2007).
11. Note that most of the real exchange rate appreciation was concentrated in the frst
semester of 2003 and in 2005, which coincided with the periods of increases in the
Brazilian base interest rate, as we will see later.
12. These are geometric averages and the real interest rate was calculated ex post.
13. At the time there was an intense debate in the media on whether or not the economy
was really overheating, since the increase in infation at the end of 2004 also refected an
increase in an important indirect tax rate, the so-known PIS-COFINS rate.
14. Indexation to the interest rate increased the impact of exchange rate depreciation on
domestic debt because the BCB usually increased its base rate abruptly after a devalu-
ation of the domestic currency.
15. The nominal defcit is the diference between the total government expenditures and
revenues, that is, it is the net interest payments minus the primary surplus.
16. Only in 2006 the primary surplus fell in relation to GDP, but it was still higher than the
level verifed in 2002.
17. Nevertheless, the combined reduction in public revenues and spending had a negative
impact on economic growth in 2003 through the balanced-budget multiplier.
18. In 200306 all of the increase in government expenditures in percentage of GDP was
directed to an increase in income transfers through social security, social assistance and
unemployment benefts (Barbosa-Filho, 2007).
19. The expansion, started in 2006, gained force in 2007, when the Brazilian GDP grew 5.4
percent.
20. For a formal model of this case, see Barbosa-Filho (2006).
21. For an analysis of the link between real exchange rates and development, see Frenkel
and Taylor (2006).
22. This idea has been frst proposed by Ros (1995) for Mexico. For a more recent analysis
of infation targeting in Mexico, see Galindo and Ros (2006).
23. For an analysis of the methods of estimation of potential output applied to Brazil, see
Barbosa-Filho (2005).
24. So far most of the mainstream studies on the uncertainty associated with estimates
of potential output have been focused on the reliability of forecasts based on real-
time data, rather than on the implications of a self-fulflling monetary policy. For an
example of the mainstream approach, see Orphanides and van Norden (2005).
25. Field of Dreams is a 1989 movie in which a farmer becomes convinced by a mysterious
voice that, if he constructs a baseball diamond in his corn feld, the ghosts of deceased star
baseball players will come to play. The voice repeatedly says to the farmer: If you build, he
will come, and the Hollywood story obviously ends up confrming the voices prediction.
REFERENCES
Arestis, P., L.F. de Paula and F. Ferrari (2007), Assessing the economic policies
of President Lula da Silva in Brazil: has fear defeated hope?, Oxford Centre for
Brazilian Studies working paper CBS-81-07.
Barbosa-Filho, N.H. (2005), Estimating potential output: a survey of the alterna-
tive methods and their applications to Brazil, IPEA working paper no. 1092.
Barbosa-Filho, N.H. (2006), Exchange rate, growth and infation, paper presented at
the Annual Conference on Development and Change, Campos do Jordo, Brazil.
Barbosa-Filho, N.H. (2007), An unusual economic arrangement: the Brazilian
economy during the frst Lula administration, 20032006, International Journal
of Politics, Culture and Society, 19, 193215.
Belaisch, A. (2003), Exchange rate pass-through in Brazil, International Monetary
Fund working paper no. 03-141.
Bevilaqua, A.S., M. Mesquita and A. Minella (2007), Brazil: taming infation
expectations, Central Bank of Brazil working paper no. 129.
Bogdanski, J., A.A. Tombini and S.R.C. Werlang (2000), Implementing infation
targeting in Brazil, Central Bank of Brazil working paper no. 1.
Frenkel, R. and L. Taylor (2006), Real exchange rate, monetary policy and
employment, United Nations, Department of Economic and Social Studies
Galindo, L.M. and J. Ros (2006), Alternatives to infation targeting in Mexico,
University of Massachusetts Political Economy Research Institute, alternatives
to infation targeting: central bank policy for employment creation, poverty
reduction and sustainable growth working paper no. 7.
Minella, A., P.S. de Freitas, I. Goldfajn and M.K. Muinhos (2003), Infation tar-
geting in Brazil: constructing credibility under exchange rate volatility, Central
Bank of Brazil working paper no. 77.
Orphanides, A. and S. van Norden (2005), The reliability of infation forecasts
based on output gap estimates in real time, Journal of Money, Credit and
Banking, 37 (3), 583601.
Ros, J. (1995), La crisis mexicana: causas, perspectivas, lecciones, Nexos, (May), 46.
Tombini, A.A. and S.A.L. Alves (2006), The recent Brazilian disinfation process
and costs, Central Bank of Brazil working paper no. 109.
158
8. Alternatives to infation targeting
in Mexico
Luis Miguel Galindo and Jaime Ros
1
8.1 INTRODUCTION
Infation targeting has become increasingly popular over the past decade.
As a nominal anchor for monetary policy with a public and explicit com-
mitment to maintain economic discipline, infation targeting is being pro-
moted as a general framework in order to reduce and control the infation
rate, improve infation predictability, accountability and transparency,
reduce the expected infation variability (Sheridan, 2001), improve the
output-infation trade-of (Clifton et al., 2001), help solve the dynamic
consistency problem and even reduce output variability (Svensson, 1997,
1998). Several Latin American countries such as Chile (1990), Peru (1994),
Mexico (1999), Brazil (1999) and Colombia (1999) have moved their
monetary regimes to an infation targeting framework (Corbo et al., 2002;
Schmidt-Hebbel and Werner, 2002).
There are also a number of concerns regarding the adoption of an infa-
tion targeting regime. There is an important concern about the ability
of the central bank to control the infation rate, in particular under the
presence of fscal or external shocks. There are also fundamental doubts
about the economic consequences of an infation targeting regime. For
example, Ball and Sheridan (2003) argue that the reduction of infation
and the reduction in output variability in the OECD countries is a general
trend that is not necessarily related to the instrumentation of infation
targeting and that this kind of regime does not afect output variability.
Furthermore, Newman and von Hagen (2002) claim that the evaluation
of infation targeting has been misleading because it has not properly con-
sidered the problem of regression to the mean given that some infation
targeting countries were worse of before their implementation than the
countries without an infation targeting framework.
In the case of emerging market economies there is a special concern
about the relationship of infation targeting with exchange rate move-
ments, imperfect and poorly regulated fnancial markets and weak
Alternatives to infation targeting in Mexico 159
monetary institutions. In efect, there are important concerns about the
efectiveness of an infation targeting regime under weak fscal conditions,
poorly regulated fnancial systems, large potential external shocks, low
institutional credibility and currency substitution phenomena (Fraga et.
al., 2003; Calvo and Mishkin, 2003). There is also a major concern that
important shifts in exchange rates will afect the general competitiveness
of the economy, the current account defcit and generate external shocks
on the infation rate. For example, abrupt changes in the exchange rate
might lead to infation paths that are inconsistent with the original infa-
tion target and therefore the central bank might try to use the nominal
exchange rate as a nominal anchor (Svensson, 1998). In this case exchange
rate shocks to the infation rate are followed by a general appreciation of
the real exchange rate afecting the general performance of the economy.
Goldfajn and Gupta (2003) argue also that an appreciation of the real
exchange rate is related with higher nominal interest rates or a tight mon-
etary policy that makes economic recovery more dif cult.
This chapter addresses Mexicos experience with infation targeting.
Section 8.2 presents some background on the operation of monetary
policy in Mexico since the 199495 peso crisis. Section 8.3 assesses empiri-
cally a number of important issues in the evaluation of infation targeting:
the efects of the real exchange rate on output, the question of the pass-
through of exchange rate movements into infation and the asymmetric
response of monetary policy in the face of exchange rate shocks. Section
8.4 considers alternatives to infation targeting as currently implemented.
Section 8.5 concludes.
8.2 INFLATION TARGETING IN MEXICO: SOME
BACKGROUND
In the 1990s Mexico experienced a variety of monetary and exchange rate
regimes. More precisely, the monetary regime went through three stages:
nominal exchange rate targeting with a crawling band regime before the
1994 crisis, monetary targeting for a short period after the crisis, followed
by a transition to infation targeting that by now has been largely com-
pleted. With the 1994 crisis the exchange rate regime shifted from a crawl-
ing band to foating.
The experience with the crawling band regime and its crisis has been
analysed widely (see, for example, Lustig and Ros, 1998; Ros, 2001). It
was during this period, in 1993, that the central bank was given indepen-
dence and a mandate to preserve price stability. After the 199495 peso
crisis, monetary policy focused on the growth of monetary aggregates and
limits to credit growth as a means to control the infationary efects of the
sharp peso devaluations and rebuild the damaged credibility of the Banco
de Mxico. More precisely, the central bank established as its nominal
anchor an intermediate target for the growth of the monetary base. As the
country moved to a fexible exchange rate regime, an assumption of no
international reserves accumulation was made. The ceiling on the growth
of the monetary base was thus in essence a growth ceiling on net dom-
estic credit. This monetary policy framework was soon abandoned as the
policy failed to stabilize infationary expectations, the exchange rate and
infation itself. The main reasons for this failure were the instability of the
relationship between the monetary base and infation and the fact that the
central bank has no control of the monetary base in the short run. The
demand for bills and coins in circulation largely determines the monetary
base and this demand is very interest inelastic in the short term (Carstens
and Werner, 1999).
The peso crisis had strong infationary efects, taking the infation rate
from single digits to over 50 percent. These infationary efects were largely
brought under control by means of a tight fscal policy and an incomes
policy negotiated with unions and business confederations. After this, the
main objective of monetary policy became the reduction of infation in a
gradual and sustainable way. At the same time, the monetary policy regime
moved towards infuencing the level of interest rates, establishing bor-
rowed reserves as its key instrument with the monetary base becoming less
relevant and the infation target more important in the conduct of policy.
Under this framework the Banco de Mxico establishes that banks
should seek a null balance in their accounts at the central bank. A penalty
rate (double the government treasury bill rate, the CETES rate) is applied
to overdrafts and positive balances are not remunerated. Thus, the banks
seek this balance in particular because if the daily average balance were
negative they would have to pay the penalty rate. By providing more
or less liquidity through daily monetary auctions, the overall net daily
average balance of all current accounts held by banks at the Banco de
Mxico may close the day at a predetermined amount. If negative, the
central bank puts the banking system in short (corto) in which case at
least one credit institution has to pay the penalty interest rate. This is the
way in which the central bank exerts pressure on interest rates. Although
the corto represents only a small amount of the total liquidity (normally
between 0.09 and 0.068 of the monetarty base), increasing it induces banks
to bid up interest rates to avoid paying overdraft charges and puts upward
pressure on market interest rates. The reverse mechanism applies when the
central bank targets a positive balance (largo) (Banco de Mxico, 1996,
Appendix 4; Carstens and Werner, 1999, pp. 1516; OECD, 2002, 2004).
The transition to infation targeting was accelerated in January 1999
when the Banco de Mxico announced a medium-term infation objective,
and since 2000 the central bank publishes quarterly infation reports to
monitor the infationary process, analyse infation prospects and discuss
the conduct of monetary policy. By now Mexico is considered to have
in place the main components of an infation targeting framework: an
independent monetary authority (since 1993) that has infation as its only
policy objective, a fexible exchange rate regime, the absence of other
nominal anchors and a transparent framework for the implementation
of monetary policy (Schmidt-Hebbel and Werner, 2002).
The implementation involves the choice of an infation index to be
adopted as target, the target range and the time horizon. The national
consumer price index (CPI) is used to determine the infation objec-
tive. Since the beginning of 2000 the bank also tracks a core CPI which
excludes volatile items such as agricultural and livestock products, and
also education (tuition fees) and prices controlled by or agreed with
the public sector. In the transition from high to moderate infation rates,
the infation objective was specifed in terms of a value that should not
be exceeded. Today policy specifes a target range of plus or minus a
percentage point. From January 1999 onwards the target of monetary
policy has been framed in terms of a medium-term infation objective of
bringing down infation to the rate prevailing in Mexicos main trading
partners by the end of 2003, interpreted as an annual growth of the CPI
of 3 percent.
Table 8.1 shows the infation targets since the 199495 crisis, actual
infation performance, the growth projections and performance as well as
the evolution of the real exchange rate. After a rough start, when in 1995
the infation objective was missed by over 30 percentage points, the central
banks record has been improving over time with its infation target being
met in fve years since 1999 and infation tending to converge towards
its medium-term target range of 3 percent plus or minus one percentage
point.
Along with the success in bringing down infation, the growth per-
formance has been disappointing. More precisely, in the second half of
the 1990s, under the stimulus of a very competitive exchange rate, the
economy rapidly recovered from the 1995 recession and grew at rates that
often surpassed the growth rate projected by the central bank and the
government. However, from 2001 to 2003, growth sharply decelerated to
rates that for three years in a row were below or barely above the rate of
population growth. That is, the economy recorded a decline in per capita
incomes from 2000 to 2003 before recovering in 2004. Overall perform-
ance since 1994 to 2004 has been unsatisfactory with GDP growth below 3
162
T
a
b
l
e

8
.
1

I
n
f

a
t
i
o
n
,

g
r
o
w
t
h

a
n
d

t
h
e

r
e
a
l

e
x
c
h
a
n
g
e

r
a
t
e
1
9
9
5
1
9
9
6
1
9
9
7
1
9
9
8
1
9
9
9
2
0
0
0
2
0
0
1
2
0
0
2
2
0
0
3
2
0
0
4
2
0
0
5
2
0
0
6
I
n
f

a
t
i
o
n

t
a
r
g
e
t

(
%
)
1
9
1
0
1
5
1
2
1
3
1
0
6
.
5
4
.
5
3

1
3

1
3

1
3

1
A
c
t
u
a
l

i
n
f

a
t
i
o
n

(
%
)
5
2
.
0
2
7
.
7
1
5
.
7
1
8
.
6
1
2
.
3
9
.
0
4
.
4
5
.
7
4
.
0
5
.
2
3
.
3
4
.
1
G
D
P

g
r
o
w
t
h

p
r
o
j
e
c
t
i
o
n
n
.
a
.
.

3
.

4
5
3
4
.
5
n
.
a
.
1
.
5
3
.
0
3
.
0
3
.
5
3
.
8
3
.
6
A
c
t
u
a
l

G
D
P

g
r
o
w
t
h
6
.
2
5
.
1
6
.
8
4
.
9
3
.
7
6
.
6
0
.
1
0
.
7
1
.
3
4
.
2
3
.
0
4
.
5
R
e
a
l

e
x
c
h
a
n
g
e

r
a
t
e

(
e
n
d

o
f

t
h
e

y
e
a
r
)

(
i
n
d
e
x
)
1
4
1
.
4
1
2
7
.
8
1
1
4
.
1
1
1
6
.
1
1
0
5
.
8
1
0
0
.
0
9
2
.
5
1
0
2
.
3
1
0
6
.
2
1
0
3
.
2
9
8
.
5
9
8
.
7
S
o
u
r
c
e
:

B
a
n
c
o

d
e

M
x
i
c
o
,

w
w
w
.
b
a
n
x
i
c
o
.
o
r
g
.
m
x
/
t
i
p
o
/
e
s
t
a
d
i
s
t
i
c
a
s
/
i
n
d
e
x
.
h
t
m
l
.

T
h
e

r
e
a
l

e
x
c
h
a
n
g
e

r
a
t
e

i
s

d
e
f

n
e
d

a
s

t
h
e

n
o
m
i
n
a
l

e
x
c
h
a
n
g
e

r
a
t
e

t
o

t
h
e

U
S

d
o
l
l
a
r
,

m
u
l
t
i
p
l
i
e
d

b
y

t
h
e

r
a
t
i
o

o
f

f
o
r
e
i
g
n

(
C
P
I

o
f

U
S
A
)

t
o

d
o
m
e
s
t
i
c

p
r
i
c
e

l
e
v
e
l
s

(
C
P
I

o
f

M
e
x
i
c
o
)
.
percent per year, well below the historical rates of the period 1940 to 1980
(6 to 6.5 percent).
8.3 THE EMPIRICAL EVIDENCE
In this section we address three relevant questions for the evaluation of
the infation targeting regime. First, what are the short-term and long-run
efects of the real exchange rate on output? Have these efects changed
with trade liberalization and integration with the US economy? Second,
how important is the pass-through of the exchange rate to prices? Has
infation targeting modifed the pass-through? Third, does infation tar-
geting have a bias towards exchange rate appreciation? If so, and the real
exchange rate has long-run efects on output, the regime is not neutral with
respect to economic growth.
However, infation targeting monetary policies are relatively recent
making it dif cult to use several econometric techniques. In this sense,
these results must be considered only preliminary. The database consists
of quarterly data for the period 1980 to 2003 and 1999 to 2003. The selec-
tion period was due to data limitations. A defnition of the variables is
included in the Appendix 8.1A.
8.3.1 The Real Exchange Rate and Output
The net efect of the real exchange rate on output is not clear-cut because
there are alternative transmission channels with opposite efects in the
short and long runs.
2
In order to assess the long-run impact of the real
exchange rate on output we proceed to estimate a vector autoregressive
model (VAR) model including output (Y
t
), investment (INV
t
), US output
(YUS
t
) and the real exchange rate (SR
t
).
3
The unit roots tests of these vari-
ables are summarized in Table 8A.1 in Appendix 8.1A. These tests indicate
that these variables are all I (1) and therefore it is necessary to consider
the option of possible cointegration
4
among them. The statistics of the
Johansen (1988) procedure
5
(Table 8.2) indicates the presence of at least
one cointegrating vector (lower case letters refer to the natural logarithm
of the series) already correcting for certain instability in the cointegrating
space (Johansen et al., 2000).
Normalizing this vector as an output equation (Equation 8.1) indicates
the presence of a positive relationship between output, investment, US
output and the real exchange rate. Therefore, a devaluation of the real
exchange rate has a positive impact on Mexican economic growth in the
long run.
y
t
5 0.478*inv
t
1 0.320*yus
t
1 0.019*sr
t
(8.1)
The series included in the estimated VAR are all considered initially
as endogenous variables. Under these circumstances, the causal relations
among these variables are not clear while the normalization of the cointe-
grating vector already implies a particular classifcation between endog-
enous and exogenous variables. Thus, it is important to take the results
with some caution due to the potential endogeneity problem (Ltkepohl
and Reimers, 1992). Therefore, we perform an impulse response exercise
in the VAR including, in the following order: yus
t
, y
t
, inv
t
and sr
t
.
The impulse response function incorporates information on contem-
poraneous efects since it ignores the contemporaneous correlations of
residuals. In order to avoid this pitfall it is common to orthogonalize inno-
vations. The orthogonalized innovation impulse response function should
be interpreted with caution since orthogonalization imposes a pre-order
which implies a semi-structural interpretation of the model. The variable
which enters frst in the system acts as the most exogenous. The move-
ments in these variables at one point in time precede the movements of
the rest of the variables which come afterwards in the system. The impulse
response analysis, using the Cholesky decomposition and the new order
of the variables in the system, confrms the positive efect of investment
and US output on Mexicos economic growth and indicates that the real
exchange rate has an initial negative efect that tends to disappear in the
long run (Figure 8.1).
The fnding that an appreciation of the real exchange rate has long-
run contractionary efects on economic growth in Mexico contradicts
Table 8.2 Statistics of the Johansen procedure including output,
investment, US output and the real exchange rate, period
1981:012003:04
y
t
5 b
1
*inv
t
1 b
2
*yus
t
1 b
3
*sr
t
**
H
0
Constant Trend Trace 95%
r 5 0 m0 0 63.09* 47.21
r 1 m0 0 20.57 29.68
r 2 m0 0 6.10 15.41
r 3 m0 0 0.01 3.76
Notes: * Signifcant at the 5 percent level; Trace 5 Trace test; r 5 number of cointegrating
vectors. Number of lags in the VAR 5 4; VAR includes constant unrestricted.
** y: output; inv: investment; yus: US output; sr: real exchange rate.
previous results of impulse response analysis that the long-run relation-
ship between output and the real exchange rate is negative (Kamin and
Rogers, 1997). It is also worth noting that the efect of the real exchange
rate on output has signifcantly increased after the beginning of the North
American Free Trade Agreement (Table 8.3). Therefore, the Mexican
economy seems to be more sensitive to exchange rate changes today than
in the past.
8.3.2 The Nominal Exchange Rate and Infation (Pass-through)
The positive impact of the nominal exchange rate on the infation rate
(pass-through) is one of the main concerns of the monetary authorities.
Under these conditions, an infation targeting regime is prone to sufer
from external dominance. That is, the high sensitivity of infation to any
0.02
0.01
0.00
0.01
0.02
0.03
1 2 3 4 5 6 7 8 9 10
Response of output (y) to
US output (yus)
0.02
0.01
0.00
0.01
0.02
0.03
1 2 3 4 5 6 7 8 9 10
output (y)
0.02
0.01
0.00
0.01
0.02
0.03
1 2 3 4 5 6 7 8 9 10
investment (inv)
0.02
0.01
0.00
0.01
0.02
0.03
1 2 3 4 5 6 7 8 9 10
real exchange rate (sr)
Note: The horizontal axis refers to the ten periods or quarters while the vertical axis shows
the response of the infation rate to a one standard deviation shock in each variable in the
system.
Figure 8.1 Impulse response analysis for yus
t
, y
t
, inv
t
and sr
t
exchange rate depreciation could cause external infation shocks that
will make it dif cult for the central bank to achieve its infation target.
It has also been argued that an infation targeting regime reduces the
pass-through problem due to its ability to increase the credibility of the
monetary authorities and therefore the importance of forward-looking
variables (Fraga, et al., 2003; Schmidt-Hebbel and Werner, 2002).
We evaluate the pass-through efect in Mexico. First, we estimate a
VAR including the infation rate, the output gap and changes in the
nominal exchange rate. The sample period covers quarterly data from
1986:01 to 2003:04. This specifcation is relatively similar to a tradit-
ional Phillips curve and includes some elements of the VAR specifca-
tion used in Schmidt-Hebbel and Werner (2002) or Fraga et al. (2003).
That is, the output gap should have a positive impact on the infation
rate due to cost pressures through the labor market and input costs
while a devaluation increases import costs and tradable goods prices
and has a positive impact on the infation rate. Additionally, it is pos-
sible to argue that this VAR includes variables that are all I (0) (Table
8A.1 in Appendix 8.1A). Therefore, this VAR, including the infation
rate, the output gap and changes in the nominal exchange rate, was
used to generate the impulse response (Figure 8.2) and indicates that
the output gap and the change in the exchange rate have both a positive
impact on the infation rate also refecting the relevance of the pass-
through efect.
Second, in order to evaluate the relevance of the pass-through in the
Mexican economy we can consider the rolling correlation coef cient
between the infation rate and exchange rate depreciation. This correlation
coef cient is estimated by adding one observation sequentially since 1989
(1). Figure 8.3 indicates the presence of a strong relationship between these
two variables. The initial reduction of the pass-through, arguably related
with the instrumentation of the infation targeting regime, is not continu-
ous and there is a persistence of a positive relationship between infation
and exchange rate movements.
Table 8.3 Cointegration vectors of the Johansen procedure including
output, investment, US output and the real exchange rate
y
t
5 b
1
*inv
t
1 b
2
*yus
t
1 b
3
*sr
t
Period inv
t
yus
t
sr
t
1982(1)1993(4) 0.852 0.328 0.475
1994(1)2003(4) 0.761 0.420 0.855
8.3.3 The Response of Monetary Policy to Exchange Rate Shocks
In this section we look for asymmetric efects of the exchange rate on
monetary policy. The hypothesis is that monetary policy shows an asym-
metric response to movements in the exchange rate. That is, the central
bank raises the interest rate in response to a depreciating exchange rate but
does not modify the interest rate in response to an appreciating exchange
rate. The fnal result of this monetary policy is an appreciation of the real
exchange rate.
Table 8.4 shows the evolution of the corto. This table shows the dates
of the changes and the quantities (with negative sign) of the monetary
base to which the penalty rate is applied. The dates without changes in
the corto are not reported in Table 8.4. Over the period between January
1996 and December 2004, the Banco de Mxico has increased the corto
0.02
0.01
0.00
0.01
0.02
0.03
0.04
1 2 3 4 5 6 7 8 9 10
0.02
0.01
0.00
0.01
0.02
0.03
0.04
1 2 3 4 5 6 7 8 9 10
0.02
0.01
0.00
0.01
0.02
0.03
0.04
1 2 3 4 5 6 7 8 9 10
Response inflation rate to
inflation rate
output gap
nominal exchange rate
Note: Period 1986:012003:04. The horizontal axis refers to the ten periods or quarters
while the vertical axis shows the response of the infation rate to a one standard deviation
shock in each variable in the system.
Figure 8.2 Impulse response analysis for the infation rate, the output
gap and changes in the nominal exchange rate
on 34 occasions and reduced it six times. Never over the period has the
banking system been put in a largo.
Increases in the corto were preceded by or were simultaneous to
depreciations of the peso on 18 occasions (53 percent of the number of
increases) and the increase in the corto was followed by appreciations of
the peso on 14 occasions. Only on four occasions (August and November
1996 and May and July 2001) were appreciations followed by a reduction
of the corto and only on one occasion did the central bank allow a sus-
tained depreciation of the peso to take place without increasing the corto
(from June through December of 2003).
We also explore the possibility of an asymmetrical response of mon-
etary policy to exchange rate movements with an econometric exercise. In
order to assess the relevance of this asymmetric efect we use a two-step
procedure similar to Cover (1992), Karras (1996) and Kim, Ni and Ratti
(1998). This procedure contains initially an equation in order to obtain an
equilibrium exchange rate trough the use of the purchasing power parity
(PPP) hypothesis (Hallwood and MacDonald, 2000). We consider that a
value above the equilibrium position is an undervaluation while a value
under the equilibrium position is an overvaluation. These values are, after-
wards, used to test for a possible asymmetric response of monetary policy
to movements in the exchange rate.
0.58
0.60
0.62
0.64
0.66
0.68
0.70
0.72
1990 1992 1994 1996 1998 2000 2002
C
o
r
r
e
l
a
t
i
o
n

c
o
e
f
f
i
c
i
e
n
t
Figure 8.3 Rolling correlation coef cient between the infation rate and
exchange rate depreciation, quarterly data 1989: 12003:4
Hence, Equation (8.2) describes the exchange rate using the purchasing
power parity condition (ibid.):
S
t
5 b
0
1 b
1
P
t
/P*
t
1 u
t
(8.2)
where S
t
denotes the nominal exchange rate, P
t
denotes the national price
index and P*
t
denotes the foreign price index.
The database includes quarterly information from 1995(1) to 2004(4)
on the nominal exchange rate and the price indexes of Mexico and the
USA. The estimation of Equation (8.2) using ordinary least squares (OLS)
Table 8.4 Changes in the corto, 19962004
Date of change Corto (millions
of pesos)
Date of change Corto (millions
of pesos)
1996 2001
23 January 5 12 January 400
25 January 20 18 May 350
7 June 30 31 July 300
21 June 40 2002
5 August 30 8 February 360
19 August 0 12 April 300
14 October 20 End of September 400
26 November 0 6 December 475
1998 2003
11 March 20 10 January 550
25 June 30 7 February 625
10 August 50 28 March 700
17 August 70 2004
10 September 100 20 February 29 (daily)
30 November 130 12 March 33
1999 27 April 37
13 January 160 23 July 41
2000 27 August 45
January 180 24 September 51
May 200 22 October 57
June 230 26 November 63
July 280 10 December 69
October 310
November 350
Source: Banco de Mxico, www.banxico.org.mx. Informe Anual and Informe sobre
Politica Monetaria, various issues.
indicates that the nominal exchange rate has a long-term relationship with
the price index diferentials.
The estimated long-run equation is as follows:
S
t
5 3.36 1 1.202 * P
t
/P*
t
(8.3)
ADF(1) 5 23.07
Then using Equation (8.3) we defne two shock functions based on the
diferences between the ftted values and the actual values of the exchange
rate (U).

Pos U
1
t
5 max (shock, zero)
Neg U
2
t
5 min (shock, zero)
The frst function is associated with an undervaluation and the second
one refers to an overvaluation. The second step of the procedure consists
in estimating an interest rate equation including the positive and negative
shocks given by the previous functions.
R
t
5 b
0
1 b
1
U
1
t
1 b
2
U
2
t
1 b
3
R
t21
1 e
t
(8.4)
Equation (8.4) describes the impact of exchange rate deviations on the
interest rate. This equation is estimated to fnd evidence of asymmet-
ric efects of the exchange rate on the interest rate. The interest rate is
the three months nominal interest rate on government bonds (CETES)
(Mexicos central bank database). Making use of such a procedure we get
the following estimates:
R
t
5 20.087*U
2
t21
1 0.145*U
1
t21
1 0.76*R
t21
2 23.6*D98(3)
220.7*D95(4)
(t) (21.01) (3.16) (25.70) (27.95)
(26.72)
( p) (0.26) (0.00) (0.00) (0.00)
(0.00)
R
2
5 0.82
Normality test:
6
Jarque-Bera: X
2
(2) 5 2.11(0.35)
Autocorrelation: LM(4) 5 0.169(0.952)
Heteroskedasticity: ARCH(4) 5 1.211(0.326)
Period: 1995:12004:4
The long-run solution is:
R
t
5 20.371*U
2
t21
1 0.615*U
1
t21
(8.5)
The main results indicate that only in the case of an undervalued
exchange rate do we fnd a statistically signifcant coef cient. According to
such estimates a tight monetary policy is carried out by the central bank
whenever there is an undervalued exchange rate; however an overvalued
exchange rate is not followed by an easy money monetary policy. The
previous estimates thus confrm that the exchange rate has an asymmetric
impact on monetary policy.
8.4 PROBLEMS WITH INFLATION TARGETING
AND ALTERNATIVES
As discussed in Section 8.1, the recent past has been characterized by
success on the infation front and a poor growth performance. Two of
our fndings in Section 8.2, the positive efect of the real exchange rate
on output and the signifcant pass-through of the exchange rate move-
ments on prices, suggest that those two aspects, success on the infation
front and poor growth performance, are linked through the evolution
of the real exchange rate (see Table 8.1). After the sharp depreciation
of 1995, the real exchange rate has been appreciating over time, a trend
that was only interrupted in 200203. From the end of 1995 to 2002,
the real exchange rate fell by 35 percent (and by 27 percent from 1995
to 2004). With a relatively stable nominal exchange rate, in an increas-
ingly open economy, government policy has provided a strong disinfa-
tionary pressure. At the same time, real appreciation, through the loss
of competitiveness of the economy, has contributed to a poor growth
performance.
Our third fnding in Section 8.2, on the asymmetric response of mon-
etary policy to exchange rate shocks, indicates that real appreciation may
be the result of a built-in bias in monetary policy towards real exchange
rate appreciation. This bias has to do with the high pass-through. Over the
past few years the central bank has been trying to break the link between
exchange rate and prices by increasing the short in response to sharp
depreciations. In doing so, it has reversed the depreciation itself. Because
the economy has been in a process of disinfation in which the central bank
has barely met its infation targets, the process is not symmetrical, that
is, there is no similar incentive to reverse the appreciations that may take
place as a result of shocks to the exchange rate.
8.4.1 Alternatives
Infation targeting is certainly a more fexible framework for monetary
policy than the previous monetary frameworks such as the use of some
monetary aggregate to control the infation rate. That is, infation target-
ing considers additional factors in order to control infation such as the
exchange rate. Additionally, the hypotheses that prices and the monetary
aggregate have a stable relationship and that money can be considered as
the exogenous variable do not appear to be valid. For example, Carstens
and Werner (1999) fnd that base money is essentially accommodating to
exogenous shocks in the Mexican case. Therefore, infation targeting rep-
resents a new framework with could include additional possible options.
A frst option is to move monetary policy towards a more neutral stance
(that is, a more symmetric response to exchange rate shocks). In this sense,
monetary authorities should try, at least, to use a reduction in the amount
of corto in the case of an appreciation of the real exchange rate. This is
more likely and feasible than in the past as infation tends to converge
towards the long-run target and, as a result of either an enhanced credibil-
ity of the central bank or a less infationary environment, the pass-through
of the exchange rate on infation tends to fall. In this case, monetary policy
has more degrees of freedom because the authorities have established a
reputation against infation. In this context, a neutral monetary policy
does not imply that the central bank will lose credibility in its commit-
ment to control infation. A neutral monetary policy will have at least two
main impacts. First, it will allow a monetary policy with less bias against
economic growth. That is, if the side efect of a contractionary monetary
policy is a reduction, at least in the short run, of economic growth, a
neutral monetary policy will contribute, in the margin, to a more dynamic
economic growth. Second, a neutral monetary policy will not contribute
to an overvaluation of the real exchange rate and, therefore, will not have
an additional bias against economic growth through the real exchange rate
channel.
A second option is to shift from a CPI target to a domestic infation
target (that is, a measure of infation purged from the direct efects of the
exchange rate on imported goods in the CPI). Such a move would further
reduce the pass-through efect of the exchange rate on the targeted price
index and contribute to eliminate the asymmetric response of monetary
policy to exchange rate shocks. In this case, a transitory shock on the
nominal exchange rate will not generate a direct response in the interest
rate. Under these circumstances monetary policy will be more focused on
the long-term path of the infation rate and therefore it will not overreact
to transitory shocks.
These two options preserve the infation targeting framework. A third,
more radical departure from the current framework, would be to combine
infation targeting with real exchange rate targeting. More precisely, the
central bank would promote a competitive exchange rate by establishing a
sliding foor to the exchange rate in order to prevent excessive appreciation
(an asymmetric band with a foor and no ceiling as in Ros 1995). This
would imply intervening in the foreign exchange market at times when
the exchange rate hits the foor but allow the exchange rate to foat freely
otherwise.
Such a proposal is free from some of the orthodox objections that have
been made to real exchange rate targeting. In particular, it does not require
knowledge of the adequate or equilibrium real exchange rate but only of
the danger zone in which overvaluation severely hurts the growth process.
This is because under our proposal the central bank does not target a par-
ticular real exchange rate but only establishes a foor on its value, leaving
the real exchange rate to move freely above this threshold. Moreover, there
is no problem with the amount of reserves required to defend the exchange
rate since the central bank only defends the foor (which requires to accu-
mulate rather than deplete reserves as would be the case if the central bank
defended a ceiling). Of course, there is an additional orthodox objection
when it comes to defending a foor to the exchange rate and this is that the
central bank may lose control of the money supply and this could imply
giving up the achievement of the infation target (for a fuller discussion,
see Frenkel and Rapetti, 2004). The problem arises at times of excess
supply of foreign currency as a result, in particular, of massive capital
infows. It is worth noting, however, that speculative capital infows will
tend to be deterred to the extent that the central bank clearly signals that
it will prevent the appreciation of the domestic currency, thus stabilizing
exchange rate expectations. If necessary, however, the central bank can
impose capital account regulations on short-term capital fows in order to
recover control over the money supply.
8.5 CONCLUSIONS
The main conclusions that emerge from our analysis can be summarized
as follows. Infation has declined, in part during the infation targeting
regime, from levels over 50 percent in 1995 to around 5 percent in 2004.
There has also been a reduction of the pass-through of exchange rate
movements into infation possibly as a result of an enhanced central bank
credibility (with its stabilizing infuence on infation expectations) or of
lower infation itself. These achievements have, however, come with a cost,
an almost continuous appreciation of the real exchange rate that has had
contractionary efects on output in the long run. Our empirical analysis
clearly supports these assertions (the reduction of the pass-through and
the negative efect of real appreciation on output). It also suggests that real
appreciation has been fed by an asymmetrical response of monetary policy
to exchange rate movements: depreciations are followed by a tightening
of monetary policy while appreciations are not reversed by a relaxation of
monetary conditions. The alternative to the current framework is to give a
more prominent place to the achievement of a competitive and stable real
exchange rate in the design of monetary and exchange rate policy.
NOTES
1. We are indebted for comments to two anonymous referees, Jerry Epstein, Roberto
Frenkel, Lance Taylor, Erinc Yeldan and other participants at the Amherst/CEDES
Conference on Infation Targeting, Buenos Aires, 13-14 May 2005. The usual caveat
applies.
2. The literature on the contractionary efects of devaluation is very large and includes,
among others, Diaz-Alejandro (1963), Cooper (1971), Krugman and Taylor (1978),
Kamin (1988), Edwards (1989), Edwards and Montiel (1989), Morley (1992) and Razmi
(2006). On the empirical evidence for Mexico, see Kamin and Rogers (1997), Lpez and
Guerrero (1998) and Kamin and Klau (1998).
3. The real exchange rate is defned as: SR
t
5 S
t
(Pus/P)
t
, where S is the nominal exchange
rate; Pus is the consumer price index of the USA; and P is the consumer price index of
Mexico.
4. The time span of the data is certainly not enough to consider the results of the cointegrat-
ing vector as a long-run solution. Therefore, it is not possible to make long-term infer-
ences on the basis of these results but it represents a valid approximation for the period.
5. The cointegration tests without any dummies are reported in Appendix 8.1A in Table
8A.2.
6. The non-normality can be attributed to an outlying value in 1995:2 when debt and fnan-
cial crisis took place in Mexico.
REFERENCES
Ball, L. and N. Sheridan (2003), Does infation targeting matter?, National
Bureau for Economic Research working paper no. 9577, March.
Banco de Mxico (1996), Informe Anual, Mxico: Banco de Mxico.
Calvo, G.A. and F.S. Mishkin (2003), The mirage of exchange rate regimes for
emerging market countries, Journal of Economic Perspectives, 17 (4), 98118.
Carstens, A. and A. Werner (1999), Mexicos monetary policy framework under
a foating exchange rate regime, Banco de Mxico, Documento de Investigacin
No. 90-05, May.
Clifton, E.V., H. Leon and C.H. Wong (2001), Infation targeting and the unem-
ployment-infation trade of, International Monetary Fund IMF working paper
WP/01/166.
Cooper, Richard N. (1971), Currency devaluation in developing countries, in
G. Ranis (ed.), Government and Economic Development, New Haven, CT: Yale
University Press, pp. 472513.
Corbo, V., O. Landerretche and K. Schmidt-Hebbel (2002), Does infation target-
ing make a diference?, in N. Loayza and R. Soto (eds), Infation Targeting:
Design, Performance, Challenges, Santiago: Banco Central de Chile, document
106, pp. 165.
Cover, J.P. (1992), Asymmetric efects of positive and negative money-supply
shocks, Quarterly Journal of Economics, 107 (4): 126182.
Diaz-Alejandro, Carlos F. (1963), A note on the impact of devaluation and the
redistributative efects, Journal of Political Economy, 71, 57780.
Dickey, D. and W.A. Fuller (1981), Likelihood ratio statistics for autoregressive
time series with unit root, Econometrica, 49 (4), 1057977.
Edwards, Sebastian (1989), Real Exchange Rates, Devaluation, and Adjustment,
Edwards, Sebastian and Peter J. Montiel (1989), Devaluation crises and macro-
economic consequences of postponed adjustment in developing countries, IMF
Staf Papers, 36, 875903.
Fraga, A., I. Goldfajn and A. Minella (2003), Infation targeting in emerging
market economics, National Bureau for Economic Research working paper no.
100019, pp. 150.
Frenkel, R. and M. Rapetti (2004), Polticas macroeconmicas para el crecimiento
y el empleo, Paper prepared for the International Labor Organization.
Goldfajn I. and P. Gupta (2003), Does monetary policy stabilize the exchange rate
following a currency crisis?, IMF Staf Papers, 50 (1), 90114.
Hallwood C. and R. MacDonald (2000), International Money and Finance,
Malden, MA: Blackwell Publishers.
Johansen, S. (1988), Statistical analysis of cointegrating vectors, Journal of
Economic Dynamics and Control, 12 (23), 23154.
Johansen, S., R. Mosconi and B. Nielsen (2000), Cointegration analysis in the pres-
ence of structural breaks in deterministic trends, Econometrics Journal, 3, 21649.
Kamin, S.B. (1988), Devaluation, external balance, and macroeconomic per-
formance in developing countries: a look at the numbers, Princeton Essays in
International Finance no. 62.
Kamin, S.B. and J.H. Rogers (1997), Output and the real exchange rate in devel-
oping countries: an application to Mexico, Board of Governors of the Federal
Reserve Bank international fnance discussion paper no. 580, May.
Kamin, S.B. and M. Klau (1998), Some multicountry evidence on the efects of
real exchange rates on output, Board of Governors of the Federal Reserve
Bank international fnance discussion paper no. 611, May.
Karras, G. (1996), Are the output efects of monetary policy asymmetric?
Evidence from a sample of European countries, Oxford Bulletin of Economics
and Statistics, 58 (2), 26778.
Kim, J., S. Ni and R. Ratti (1998), Monetary policy and asymmetric response in
default risk, Economics Letters, 60 (1), 8390.
Krugman, P. and L. Taylor (1978), Contractionary efects of devaluation, Journal
of International Economics, 8 (3), 44556.
Kwiatkowski, D., P.C.B. Phillips, P. Schmidt and Y. Shin (1992), Testing the
null hypothesis of stationary against the alternative of a unit root, Journal of
Econometrics, 54 (13), 15978.
Lpez, J. and C. Guerrero (1998), Crisis externa y competitividad de la economa
mexicana, El Trimestre Econmico, OctoberDecember, LXV (260), 58298.
Ltkepohl, H. and H.E. Reimers (1992), Impulse response analysis of co-inte-
grated systems, Journal of Economic Dynamics and Control, 16, 5378.
Lustig N. and J. Ros (1998), Reforma estructural, estabilizacin econmica y el
sndrome mexicano, Desarrollo Econmico, 37 (148), 50331.
Maddala, G.S. and I. Kim (1998), Unit Roots, Cointegration and Structural
Change, Cambridge: Cambridge University Press.
Morley, Samuel A. (1992), On the efect of devaluation during stabilization pro-
grams in LDCs, Review of Economics and Statistics, LXXIV, 217.
Neumann, M.J.M. and J. von Hagen (2002), Does infation targeting matter?,
Federal Reserve Bank of St Louis Review, (JulyAugust).
OECD (2002), Economic Surveys, Mexico City: OECD.
OECD (2004), Economic Surveys, Mexico City: OECD.
Phillips, P.C.P. and P. Perron (1988), Testing for unit root in the time series
regression, Biometrica, 75 (2), 33546.
Razmi Arslan (2006), The contractionary short-run efects of nominal devaluation
in developing countries: some neglected nuances, University of Massachusetts
Amherst, Department of Economics, working paper 2005-09.
Ros, J. (1995), Despus de la crisis: la poltica econmica, Nexos, accessed
October 1996 at www.nexos.com.mx/articulosEspeciales.php.
Ros, J. (2001), Del auge de capitales a la crisis fnanciera y ms all: Mxico en los
noventa, in Ricardo Ffrench-Davis (ed.), Crisis fnancieras en pases exitosos,
Santiago: CEPAL-McGraw Hill, pp. 11957.
Schmidt-Hebbel, K. and A. Werner (2002), Infation targeting in Brazil, Chile and
Mexico: performance, credibility, and the exchange rate, Central Bank of Chile
working paper no. 171, July.
Sheridan, N. (2001), Infation dynamics, PhD dissertation, Johns Hopkins
University.
Svensson, L.E. (1997), Infation forecast targeting: implementing and monitoring
infation targets, European Economic Review, 41 (6), 111146.
Svensson, L.E. (1998), Open economy infation targeting, National Bureau for
Economic Research working paper no. 6545, May.
APPENDIX 8.1A
Description of Variables
y 5 Real Gross Domestic Product (GDP) in millions of Mexican pesos of
1993, Instituto Nacional de Geografa, Estadstica e Informtica (INEGI),
http://www.inegi.gob.mx.
yus 5 Real Gross Domestic Product (GDP) in billions of US dollars at
Table 8A.1 Unit root tests
Variable ADF PP(4) KPSS(10)
A B C A B C h
m
h
t
y
t
3.19(8) 0.54(8) 1.99(8) 4.85 0.33 2.32 0.946 0.171
y
t
3.91(8) 3.64(8) 2.54(8) 21.52 21.54 18.91 0.096 0.047
yus
t
2.68(2) 0.64(2) 4.27(2) 2.69 0.12 6.98 0.967 0.065
yus
t
3.45(8) 4.04(2) 2.67(1) 7.57 7.63 4.51 0.072 0.065
inv
t
3.69(4) 0.97(5) 0.34(2) 2.44 1.12 0.51 0.663 0.148
inv
t
4.11(8) 3.88(8) 3.83(8) 8.82 8.81 8.82 0.156 0.066
sr
t
4.05(4) 3.26(4) 0.02(0) 2.77 2.49 0.03 0.262 0.101
sr
t
10.8(0) 10.90(0) 10.96(0) 10.81 10.82 10.87 0.108 0.071
p
t
1.86 (3) 3.50 (3) 0.55 (3) 0.37 3.91 0.99 0.903 0.238
p
t
4.01 (4) 2.01 (2) 1.30 (2) 3.59 2.60 1.79 0.613 0.074
p
t
9.09 (1) 9.08 (1) 9.14 (1) 9.80 9.85 9.91 0.068 0.063
p*
t
1.39 (3) 1.84 (3) 2.55 (3) 3.48 4.60 7.87 0.981 0.242
Dp*
t
7.64 (0) 3.66 (2) 2.63 (2) 7.75 6.50 3.42 0.627 0.087
r
t
2.64 (3) 0.96 (3) 0.82 (3) 2.70 1.10 0.72 0.615 0.097
r
t
5.28 (2) 5.14 (2) 5.12 (2) 9.82 9.69 9.72 0.236 0.077
r*
t
2.62 (3) 1.24 (3) 1.82 (3) 1.66 0.32 1.71 0.769 0.076
Dr*
t
2.67 (8) 2.49 (8) 2.17 (8) 10.63 10.44 10.31 0.140 0.101
s
t
2.10 (3) 3.35 (3) 1.71 (4) 1.02 3.21 1.62 0.868 0.229
s
t
4.18 (2) 3.17 (2) 2.42 (2) 7.73 6.89 5.57 0.512 0.068
gap
t
3.99 (8) 4.00 (8) 4.05 (8) 8.67 8.71 8.75 0.043 0.043
gap
t
4.75 (8) 4.79 (8) 4.81 (8) 23.38 23.42 23.54 0.052 0.048
Notes: Test statistics in bold indicate a rejection of the null hypothesis. Critical values at 5
percent signifcance level for the Augmented Dickey-Fuller and Phillips-Perron tests for a size
T5100 are 3.45 including constant and trend (model A), 2.89 including constant (model
B) and 1.95 without constant and trend (model C) (Maddala and Kim, 1998, p. 64). h
and
h
t
are the KPSS tests for the null hypothesis of stationarity around a level and deterministic
linear trend, respectively. Both tests are calculated with a lag window size equal to ten. The 5
percent critical values for the two tests are 0.463 and 0.146, respectively (Kwiatkowski et al.
1992, p. 166). Small letters represent the values in logarithms.
Sources: Dickey and Fuller (1981) and Phillips and Perron (1988).
prices of 2000, (the series is already seasonally adjusted), US Department
of Commerce: Bureau of Economic Analysis.
p 5 Mexican consumer price index (base 2002 5 100), Bank of Mexico,
http://www.banxico.org.mx.
p* 5 US consumer price index (base 198284 5 100), US Department of
Commerce, Bureau of Economic Analysis.
GAPY 5 Deviation of the real Gross Domestic Product (GDP) from
potential output obtained using the Hodrick-Prescott flter.
S 5 Exchange rate (pesos per US dollar), 48-hour interbank exchange
rate. The data is taken from the last day of each quarter, Bank of Mexico,
http://www.banxico.org.mx.
Sr 5 Real exchange rate defned as S (P*/P) where S is the nominal
exchange rate, P refers to domestic prices and P* to foreign prices.
r 5 Nominal interest rate on three-month treasury bills (CETES 91 days).
Average of the last month of the quarter, Bank of Mexico, http://www.
banxico.org.mx.
r* 5 Nominal interest rate three-month treasury bill, Board of Governors
of the Federal Reserve System
Table 8A.2 Mis-specifcations tests of the Vector Autoregressive Model:
s
t
, (p
t
2 p*
t
) and (r
t
2 r*
t
)
Variable Autocorrelation:
LM(4)
Heteroskedasticity
ARCH(4)
NormalityJ-B
s
t
F(3,14) 5 0.40 [0.75] F(3,11) 5 0.095 [0.96] c
2
(2) 5 8.39[0.02]*
p
t
2 p*
t
F(3,14) 5 1.59 [0.24] F(3,11) 5 0.31 [0.82] c
2
(2) 5 3.57[0.17]
r
t
2 r*
t
F(3,14) 5 2.05 [0.15] F(3,11) 5 0.09 [0.97] c
2
(2) 5 9.38[0.01]**
Note: *, ** indicate a rejection of the null hypothesis to 5 percent and 1 percent signifcance
level. Period 1995(1)2004(4).
179
9. Five years of competitive and stable
real exchange rate in Argentina,
200207
Roberto Frenkel and Martn Rapetti
1
9.1 INTRODUCTION
In 1991 Argentine authorities established the convertibility regime, which
implied the pegging of the peso (AR$) to the US dollar ($) by law and the
validation of contracts in foreign currencies. The new monetary arrange-
ment also stipulated that the central bank must fully back the monetary
base with foreign reserves,
2
what in practice turned the central bank into
a currency board. The convertibility regime was the pillar of a broader
stabilization program, intended to take the economy away from the high
infation regime settled since the mid 1970s, which had led to two brief
hyperinfationary episodes in 1989 and 1990. The program also included
an almost complete liberalization of trade fows and the full deregulation
of the capital account of the balance of payments. It was jointly applied
with an impressive process of market-friendly reforms, targeting the priva-
tization of a large proportion of state-owned frms.
The program successfully managed to stop infation and initially
spurred rapid growth. However, as happened with many other stabiliz-
ation programs in the region, it led to the appreciation of the real exchange
rate, which made economic growth highly dependent on external debt
accumulation.
3
Since the Asian and Russian crises, and especially after the
Brazilian devaluation in 1999, the deceleration of capital infows put the
economy into a defationary trend that ended up in a fnancial and exter-
nal crisis in 200102. Between the last days of 2001 and the beginning of
2002, Argentina declared the default of its international debt and devalued
the peso. The collapse of the convertibility regime implied a 21 percent
contraction in GDP with respect to the peak of mid-1998 and a rise in the
unemployment rate up to 21.5 percent, taking half of the population below
the poverty line.
However, only one quarter after the devaluation and default economic
activity gradually started to recover. By the end of 2002, once the govern-
ment managed to stabilize domestic fnancial markets, the recovery gained
momentum and since then the economy has shown an impressive perform-
ance. In fve years from the frst quarter of 2002 to the same period in
2007 GDP has been growing at 8.1 percent annual rate, reaching a peak
18.1 percent higher than the one in mid-1998. The investment rate rose to
22 percent of GDP (on seasonally adjusted basis), which is the maximum
range of the of cial time series beginning in 1993 and continues to grow at
a higher pace than GDP. During these fve years exports have expanded
at a slightly higher rate than GDP, but its rate of growth has substantially
increased from mid 2004 onwards.
Current economic evolution contrasts with Argentinas economic per-
formance of the last 60 years. Since the Second World War economic
growth has been low and very volatile, especially in the second fnancial
globalization period beginning in the mid 1970s. For the frst time in 30
years Argentina has grown fve years in a row. More importantly, current
expansion is based on solid macroeconomic fundamentals. The volatility
of Argentine growth has been typically associated with current account
and fscal defcits. Between the mid 1940s and the mid 1970s, macro-
economic evolution was characterized by stop-and-go cycles related to
external imbalances. During the second fnancial globalization period, the
availability of external funds momentarily relaxed the external constraint
to growth, but it led to two episodes of explosive fscal and external debt
accumulation,
4
one in the late 1970s and the beginning of the 1980s and the
other during the convertibility regime period.
In contrast to those traditional fscal and external imbalances, the
current macroeconomic confguration stands out with the existence of
external and fscal surpluses. Certainly, the debt restructuring in 2005
implying a $67 billion reduction in the nominal stock softened both
external and fscal requirements, releasing resources for private sector
spending. Similarly, favorable external conditions especially the high
prices of some commodities have also played a role. However, in our
view the main factor behind the current success is the of cial policy aiming
at preserving a stable and competitive real exchange rate (SCRER). The
SCRER has been a key factor explaining the current account adjustment,
which passed from a $14.5 billion defcit in 1998 to $7.6 billion surplus
in 2006. From this $22 billion adjustment, $20 billion came from the
improvement in the trade balance, which is mainly attributable to the
efects of the exchange rate depreciation.
The infuence of the SCRER on the fscal accounts performance has
also been important. After devaluation, the government introduced taxes
on traditional exports, mainly agricultural products and oil. In practice,
Five years of competitive and stable real exchange rate in Argentina 181
this measure implied the introduction of multiple exchange rates that
contributed to reduce the pass-through of devaluation to wage-goods
prices, but also to capture part of the rent that these sectors obtained from
the competitive real exchange rate. In 2006 the federal administration
recorded a primary surplus of 3.5 percent of GDP and a total surplus of
1.8 percent of GDP, from which taxes on exports accounted for 63 percent
of the former and 122 percent of the latter.
In our view, the positive efects of the SCRER policy are the princi-
pal factors explaining the rapid growth experienced so far. This policy
promotes economic growth not only by preserving external and fscal
accounts sustainability, but also by providing incentives to the tradable
sector and thus encouraging the expansion of its production, employment
and investment. Although the success of the SCRER strategy throughout
these fve years has undoubtedly had a persuasive impact among ana-
lysts, skepticism remains.
5
The SCRER policy collides with conventional
wisdom, particularly with the trilemma paradigm.
In this chapter, we argue that a macroeconomic regime based on a
SCRER is both desirable and manageable for a developing open economy.
The next section describes the evolution of monetary and exchange rate
policies in Argentina in the post-convertibility period. Section 9.3 discusses
the usual criticisms against the SCRER policy and shows the conditions
in which this policy is sustainable. Section 9.4 presents some concluding
remarks regarding the management of a macroeconomic regime with a
SCRER as an intermediate target.
9.2 MONETARY AND EXCHANGE RATE POLICIES
IN THE POST-CONVERTIBILITY PERIOD
The deceleration of capital infows that led to the 200102 crisis began
in mid 1998. This process took place simultaneously with a persistent
rise of the sovereign risk premium. However, the divergent trends in
the domestic fnancial market that triggered the collapse of the convert-
ibility regime only started in October 2000, associated with the political
turmoil caused by the Vice-Presidents resignation. The process followed
a simple dynamics. Devaluation expectations and the perception of a
higher risk of default led the private sector to withdraw deposits and
run against the central banks international reserves. There were no
bankruptcy reports of failing banks because the central bank supported
the liquidity of the banking system. Despite several signals issued by
the government aiming at changing the expectations, the intensifcation
of this process could not be stopped. In December 2001 restrictions on
capital outfows and on the withdrawal of deposits (the so-called cor-
ralito) were established.
After the abandonment of the convertibility regime, the government
aimed to restrain the capital outfow and stabilize the foreign exchange
(FX) market by introducing a dual exchange rate regime. The idea was
to use this scheme only temporarily, in order to stabilize the nominal
exchange rate while the domestic prices absorbed the impact of the devalu-
ation, and then pass to a foating rate regime. The authorities also decided
to convert to pesos most of the domestic debts contracted in dollars (bank
credits, rents and so on) at a AR$/$ 1 rate (plus indexation to consumer
price index (CPI) infation), thus neutralizing most of the efects of relative
price change on the debtors balance sheets. In contrast, banks deposits
originally denominated in dollars were pesoifcated at a AR$/$ 1.40 rate
(plus indexation to the evolution of CPI infation).
6
Together with the
pesoifcation, the authorities unilaterally decided to extend the maturity
and duration of all deposits, including those originally contracted in pesos.
In exchange, private depositors received certifcates for the reprogrammed
deposits.
In February 2002 the FX market was unifed and the peso started to
foat freely. Given the political and economic uncertainty, the exchange
rate skyrocketed fed by self-fulflling expectations. Interestingly, this
process developed in an illiquid environment because of the restrictions
on the withdrawal of cash from banks. The erratic monetary policy that
followed in the frst quarter of 2002 also failed to stabilize the exchange
rate. The authorities delayed the launching of a domestic asset that could
perform as a potential substitute for foreign currency. Given the distrust
in banks and in the Treasury, the economic depression and the growing
infation, the international currency appeared as the only asset available
to allocate fnancial savings. Only two and half months after the devalua-
tion the central bank started to issue notes (the Lebac) in order to supply
a fnancial instrument that could compete with the dollar.
All these factors contributed to deepen the perverse dynamics of the
fnancial variables during the frst semester of 2002. The capital fight from
domestic assets between March 2001 and mid 2002 is illustrated in Figure
9.1, which shows the large fall in private bank deposits
7
and international
reserves, while the nominal demand for cash remains stagnant. These
developments provide evidence for the substitution of local assets (cash
and deposits) in exchange for external assets (international reserves).
The result of the asset substitution afected the FX market. The nominal
exchange rate (NER) and real exchange rate (RER)
8
rose continu-
ously in the frst semester of 2002 (around 260 percent and 180 percent,
respectively). Their paths are shown in Figure 9.2. Real exchange rate
overshooting was so pronounced that in June 2002 its value was almost
193 percent higher than the 19802001 period average value, and 309
percent higher than the convertibility decade average.
These disruptive trends began to revert in July 2002. The turning point
was the exchange rate stabilization. Several factors contributed to this
outcome. Controls on FX transactions
9
had been introduced in November
2001 before the convertibility collapse and they were further tightened
in March 2002. Since June 2002 controls and interventions in the FX
market were strengthened in order to conduct a systematic policy to sta-
bilize the exchange rate. The decision that export revenues surpassing $1
million had to be sold directly to the central bank was especially important
in this regard. This became the main source of international reserves ac-
cumulation for the monetary authority, which in turn agreed to increase
the volume of its interventions in the FX market.
Financial market behavior itself also contributed to stop the bubble in
the exchange rate. On the one hand, local interest rates skyrocketed (Figure
9.3). In July 2002 the average time deposits annual interest rate reached a
0
10000
20000
30000
40000
50000
30000
40000
50000
60000
70000
80000
90000
100000
2001 2002 2003 2004 2005
Pesoifcation
Exchange rate
stabilization
Demand for cash
Central bank reserves
Private bank deposits
Lebac
Source: Central Bank of Argentina.
Figure 9.1 Demand for cash, central bank international reserves, Lebac
and private bank deposits (right axis) (in millions of pesos and
dollars)
76 percent peak, and the annual interest rate of the 14-day Lebacs reached
almost 115 percent. Thus local fnancial assets began to appear more
attractive as substitutes for the dollar. On the other hand, as mentioned
above, the real price of the dollar reached very high and abnormal levels
in historical terms (that is, the prices in dollars of domestic assets, non-
tradable goods and salaries were perceived as abnormally low). In this
context, once the authorities managed to stop the exchange rate bubble
in July, the public rapidly changed expectations and the market started to
show an appreciation trend.
Thus, in the second half of 2002 a phase of monetary and fnancial
variables normalization started. After reaching a peak of almost AR$/$
4 during the last days of June, the exchange rate began to experience
a smooth nominal appreciation trend. Although the infation rate was
already low and decelerating, the rise in domestic prices contributed to
the real appreciation. In that context, local assets became increasingly
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
2001 2002 2003 2004 2005 2006
MRER
NER
RER
Notes:
* Calculated using US and Argentina consumer price indexes.
** Elaborated using monthly data of the exchange rates of the 18 principal trading partners
of Argentina, adjusted by the respective consumer price indexes.
Figure 9.2 Bilateral nominal (NER) and real exchange rate (RER) with
the US and multilateral real exchange rate (MRER) (in pesos
and indexes 1 5 December 2001)
attractive. Bank deposits began to grow, as did the demand for Lebac,
local shares and the demand for cash (Figure 9.1). This portfolio substitu-
tion in favor of local assets resulted in a persistent drop in the interest rates
(Figure 9.3).
The normalization in fnancial activity dissipated, disrupting expecta-
tions and thus favored the recovery of private expenditure. Interestingly,
this recovery took place without signifcant contribution from bank
credits. Even though private deposits gradually improved allowing the
recuperation of banks liquidity, credit to the private sector continued
shrinking until late 2003. The fnancial crisis appeared to have persistent
efects on the behavior of bank credit, which at the beginning of 2007 was
still below the peak reached in 1998 (Figure 9.4).
Domestic expenditure was mainly fnanced by private sector cash
holdings. Figure 9.5 shows the increase in cash holdings since the fourth
quarter of 2001. Both the monetary base/GDP ratio and the monetary
base/total bank deposit ratio showed very high rates of growth and also
relatively high levels in comparison to the convertibility period. Although
the low interest rates on banks deposits (and the tax on fnancial transac-
tions) have contributed to that performance, this behavior seems to be
another persistent consequence of the fnancial crisis.
0
20
40
60
80
M
o
n
t
h
l
y

a
v
g
100
120
2001 2002 2003 2004
14d Lebac
91d Lebac
Time deposits 3059d
Prime 30d
Exchange rate stabilization
Figure 9.3 Interest rates in pesos: Lebac (14 and 91 days), time deposits
(30 to 59 days) and prime (30 days) (monthly average, in %)
8
12
16
20
B
a
n
k

c
r
e
d
i
t
/
G
D
P

i
n

p
e
r
c
e
n
t
a
g
e
24
28
96 97 98 99 00 01 02 03 04 05 06
Source: Central Bank of Argentina; Ministry of Economy of Argentina.
Figure 9.4 Bank credit to private sector in relation to GDP seasonally
adjusted
15
20
25
30
35
M
o
n
e
t
a
r
y

b
a
s
e
/
t
o
t
a
l

d
e
p
o
s
i
t
s
M
o
n
e
t
a
r
y

b
a
s
e
/
G
D
P
40
45
50
4
5
6
7
8
9
10
11
12
1994 1996 1998 2000 2002 2004
MB/GDP
MB/Deposits
Source: Central Bank of Argentina; Ministry of Economy of Argentina.
Figure 9.5 Monetary base in relation to total bank deposits and with GDP
seasonally adjusted
The nominal and real appreciation process stopped around mid 2003,
when the government decided to manage the fotation of the exchange rate
in order to preserve the SCRER. The impact of the SCRER on economic
activity, employment and external and fscal accounts was proving to be
highly favorable. Thus, the government gradually started to recognize and
make explicit reference to the importance of preserving the SCRER in
the of cial economic strategy. On the contrary, central bank authorities
never made any explicit statement regarding the existence of any exchange
rate target. According to their of cial statements and documents, the per-
manent intervention in the FX market has been oriented to accumulate
international reserves for a precautionary purpose, namely to protect
the economy from international capital markets volatility. Statements
aside, the joint intervention of the central bank and the Treasury in the
FX market actually controlled the price of the dollar in a narrow range
between AR$2.8 and AR$3.1. The resulting fuctuation of the exchange
rate in this range made the multilateral real exchange rate remain stable
around a level 129 percent higher than the one at the end of the converti-
bility regime. The bilateral real exchange with the US dollar also remained
stable for some years, but since early 2005 it has shown a soft appreciation
trend (see Figure 9.2).
In 2002, when the Congress passed a law revoking the currency board,
the government decided to keep the central banks independence with the
mandate of pursuing low infation rates as its primary mission. Given that
the economy was still absorbing the efects of the crisis and the devaluation
and that the domestic fnancial markets had shrunk signifcantly, the central
bank disregarded the option of following an infation targeting regime. The
transmission mechanisms through the interest rate on aggregate demand
were thought to be uncertain and weak.
10
Instead, the authorities opted
to follow a more pragmatic policy based on broad quantitative monetary
targets. From 2003 on, targets have been announced at the beginning of
every year throughout the central bank monetary programs, in which the
authorities commit themselves to maintain monetary aggregates within a
certain range. Given the uncertainty surrounding the efects of monetary
policy, the central bank has tended to set these ranges suf ciently broadly.
However, their upper bounds ended up being systematically lower than
the monetary expansion arising from the intervention in the FX market to
preserve the SCRER. Thus, since 2003 the central bank has dealt with two
conficting objectives: the preservation of a competitive exchange rate by
intervening in the FX market and at the same time the attainment of the
targets of monetary expansion announced in the monetary program.
The tension between these two policy objectives can be observed in Table
9.1, which shows the sources of variation of the monetary base. In the frst
188
T
a
b
l
e

9
.
1

S
o
u
r
c
e
s

o
f

v
a
r
i
a
t
i
o
n

o
f

t
h
e

m
o
n
e
t
a
r
y

b
a
s
e

(
m
o
n
t
h
l
y

a
v
e
r
a
g
e

v
a
r
i
a
t
i
o
n
,

i
n

m
i
l
l
i
o
n
s

o
f

p
e
s
o
s

o
r

d
o
l
l
a
r
s
)
C
e
n
t
r
a
l

b
a
n
k

F
X

i
n
t
e
r
v
e
n
t
i
o
n
M
o
n
e
t
a
r
y

b
a
s
e

v
a
r
i
a
t
i
o
n
E
x
c
e
s
s

o
f

m
o
n
e
t
a
r
y

e
x
p
a
n
s
i
o
n
C
e
n
t
r
a
l

b
a
n
k

s
t
e
r
i
l
i
z
a
t
i
o
n
A
s
s
i
s
t
a
n
c
e

t
o

b
a
n
k
s
A
s
s
i
s
t
a
n
c
e

t
o

t
h
e

T
r
e
a
s
u
r
y
O
t
h
e
r
s
T
r
e
a
s
u
r
y

F
X

i
n
t
e
r
v
e
n
t
i
o
n
a
2
0
0
2
:
0
1
b
1

4
5
0

4
0
6
1

8
5
6
2
1
6
1

4
2
6
1
2
4
5
2
2
n
.
a
.
2
0
0
2
:
0
2
c

1

2
8
1
1

6
7
4
3
9
3
2
7
0
8
6
2
5
0
3
2
7
n
.
a
.
2
0
0
3

1

3
7
4

8
0
9
5
6
5
4
2
0
1
2
5
5
2
3
2

2
8
2
0
0
4

1

9
3
1

4
8
3
1

4
4
7
3
2
3
6
0
1
5
4
3
1
9
1
1
2
2
0
0
5

2

3
5
2

1
8
6
2

1
6
6
8
3
6
9
3
9
3
5
3
3
9
3
4
3
2
0
0
6

3

5
8
4
2

1
1
3
1

4
7
1
8
1
2
2
0
4
4
4
6
8

4
0
N
o
t
e
s
:
a
.

I
n

m
i
l
l
i
o
n
s

o
f

d
o
l
l
o
r
s
.
b
.

C
a
l
c
u
l
a
t
e
d

f
o
r

t
h
e

p
e
r
i
o
d

F
e
b
r
u
a
r
y
J
u
n
e

2
0
0
2
.
c
.

A

c
a
n
c
e
l
l
a
t
i
o
n

o
f

a

B
a
n
c
o

N
a
c
i
s

r
e
d
i
s
c
o
u
n
t

b
y

t
h
e

T
r
e
a
s
u
r
y

w
i
t
h

a
s
s
i
s
t
a
n
c
e

o
f

t
h
e

c
e
n
t
r
a
l

b
a
n
k

i
n

S
e
p
t
e
m
b
e
r

2
0
0
2

f
o
r

a
b
o
u
t

A
R
$
3
5
0
0

m
i
l
l
i
o
n

i
s

o
m
i
t
t
e
d
.

S
o
u
r
c
e
:

C
e
n
t
r
a
l

B
a
n
k

o
f

A
r
g
e
n
t
i
n
a
.
semester of 2002 the central bank intervened in the FX market providing
dollars to contain the depreciation pressures. Thus, the FX intervention
operated as a source of monetary base contraction. Once the exchange
rate was stabilized, the accumulation of international reserves resulted,
on the contrary, in a source of monetary expansion. During the second
semester of 2002 this source of monetary expansion was easily absorbed
by the rapid growth in the demand for cash caused by the re-monetization
of the economy. However, since 2003 the gradual deceleration of money
creation established in the monetary programs in order to maintain infa-
tion expectation under control started to confict with the increasing
expansion of monetary base generated by central banks intervention in
the FX market aiming to preserve the SCRER. Since the amount of mon-
etary base created to intervene in the FX market (frst column in Table
9.1) exceeded the actual expansion of the monetary base to accomplish
the monetary targets (second column), an excess of monetary expansion
(third column) had to be absorbed.
This excess of monetary expansion has been absorbed through several
mechanisms. Throughout 2003 the sterilization operations implemented
by the issuing of central bank notes were especially relevant. The need for
sterilization increased during 2004 and 2005. However, the central bank
could limit the issuing of Lebac because other compensatory mechanisms
began to operate. In the frst place, as liquidity grew the banks started to
service the debt incurred with the central bank during the fnancial crisis.
Hence, banks capital payments and especially the payment of interests
operated as a source of monetary base contraction. In 2005 the central
bank launched a program allowing the acceleration of banks debts amort-
izations, reinforcing this contractionary mechanism. By early 2006 most
banks had cancelled their debts with the monetary authority.
The Treasury also helped to absorb the excess of monetary expansion.
While in 2002 a net fow of fnancing from the central bank to the Treasury
was observed, in 2003, and especially since 2004, the transactions between
the Treasury and the central bank operated as a source of contraction of
the monetary base. The Treasurys purchases of international reserves
with the proceeds of the primary surplus gave place to a monthly average
contraction of the monetary base of AR$543 million in 2004. The main
purpose of these operations was to continue servicing of the debt with the
multilateral fnancial institutions. The Treasury and other of cial agencies
also accumulated part of the fscal surplus in foreign currency and thus
intervened directly in the FX market to alleviate central banks manage-
ment of the conficting objectives. These operations started in late 2002
and gradually expanded afterwards, thus becoming an important policy
instrument (see last column of Table 9.1).
In 2005 the sterilization needs through the issuing of Lebac increased
again. Thus, in order to soften the appreciation pressures in the FX market,
controls on the capital account were introduced in June. Basically, the new
measures established that all capital infows excluding the issuing of new
private and public debt, international trade fnancing and foreign direct
investment would be subject to a 30 percent unremunerated reserve
requirement for at least 365 days. This strategy is inspired by that applied
in Chile in the early 1990s and attempts to reduce short-term capital
infows. However, controls left open ways to avoid the reserve require-
ments. For instance, capital infows can easily circumvent the reserve
requirement by operating through the stock exchange market (by buying
domestic assets abroad and selling them in the local market). There has
been no evidence of a reduction in the supply of dollars in the FX market
after the measures were implemented. Local analysts believe that controls
are inefective and even the authorities do not reject the idea that they were
introduced more as a signal of the of cial willingness to maintaining the
SCRER strategy rather than as an efective control mechanism.
As from 2006 monetary policy stopped targeting the monetary base
and started to focus on M2. The authorities argued that the change
in the target was due to the increasing monetization of the economy
and the gradual recovery of bank credit. In these conditions, it was argued,
the use of a larger monetary aggregate represented a step forward toward
the fne-tuning of monetary policy. In practice, the switch of the monetary
aggregate target helped to relax the conficting management of exchange
rate and monetary policies. The central bank was facing increasing dif-
fculties to accomplish the monetary base targets. As Table 9.1 shows, the
excess of monetary expansion had risen substantially between 2003 and
2005. The use of M2 as a target gave the authorities greater fexibility to
conduct the two-target policy, allowing for greater intervention in the FX
market and expansion of the monetary base.
In summary, during the post-convertibility period the central bank has
been able to conduct the two-target policy successfully. Moreover, while
doing so it has obtained quasi-fscal surpluses every year. Some analysts
have argued that the management of monetary policy focusing on two
targets has had an infationary bias. Certainly, infation accelerated during
2004 and 2005 and has remained stable around an annual rate of 10
percent since 2006. In our view, the acceleration of infation is due not to
inconsistencies in the management of monetary and exchange rate poli-
cies, but to the lack of coordination between these and the fscal policy.
The expansion of public spending well above the increase of tax revenues
since 2006 has implied an expansionary fscal impulse to an already fast-
growing aggregate demand. Given the fact that monetary and exchange
rate policies focus on preserving a SCRER which is intended to put the
economy in a high growth path fscal policy is the only macroeconomic
instrument available to moderate aggregate demand when infationary
pressures arise.
9.3 THE ECONOMICS OF THE SCRER
The notion that a SCRER favors economic development has a long tradit-
ion in economic theory. Advocates of the outward orientation approach
to development during the 1960s and 1970s pointed to the SCRER as
a key element for that strategy (Balassa, 1971; Daz Alejandro, 1979).
According to this view, a competitive real exchange rate boosts economic
growth because it softens the balance of payment constraint and favors
the development of tradable activities, which tend to be more dynamic.
The stability of the exchange rate is also important because low volatility
reduces the risk and uncertainty of investment in tradable sectors. These
arguments have been recently revitalized by modern scholars.
11
Besides the
traditional efects, modern advocates also emphasize that a development
strategy based on a SCRER is market friendly (avoiding rent-seeking
practices) and compatible with free trade agreements. In recent years
many studies have documented a statistically and economically positive
relationship between growth and real exchange rates.
12
The preservation of a SCRER has also been invoked for other reasons.
Maintaining a competitive real exchange rate typically involves interven-
tion in the FX market and the accumulation of international reserves. It is
a well documented fact that international fnancial integration may lead to
macroeconomic instability and increases the likelihood of external crises.
Some scholars argue that international reserves accumulation serves as
a shield against volatile capital fows, especially for developing countries
(Feldstein, 1999). Empirical studies show a positive relationship between
reserve accumulation and growth (Polterovich and Popov, 2002).
Another less studied motive is that competitive real exchange rates
promote job creation. Besides the above-mentioned growth efects, a
SCRER may impact on employment through a more intense use of labor
(Frenkel, 2004). A competitive parity favors labor-intensive activities and
sectors and also the substitution of expensive inputs (such as imports) in
favor of labor across sectors.
13
Probably because of all these reasons, the preservation of a SCRER
does not attract much criticism by itself. Few scholars deny the benefcial
aspects of stable and predictable relative prices and the positive efects on
growth. In some cases welfare arguments against public intervention in
the FX market are raised. But the optimality of the free market determi-
nation of the exchange rate and the argument that the public sector has
no informational advantage over the private sector are not very appeal-
ing ideas in the specialized discussion about exchange rate regimes and
policies. The apparent volatility of capital fows and the instability and
unpredictability of free-foating exchange rates greatly lessen the relevance
of those ideas (Frankel and Rose, 1995). Moreover, in some scholars view
the free-foating exchange rate indeterminacy and unpredictability is pre-
cisely the deeper foundation of the need for managing the exchange rate
(Blecker, 2005). This is particularly emphasized in countries in which the
real exchange rate plays a crucial role in the economic performance.
Skepticism towards the SCRER policy points to the ability of govern-
ments to conduct it. The main objection is that the real exchange rate as
any real variable is not under the governments control, at least in the
long run. However, given the weak empirical support of real exchange rate
determination models,
14
the objections relevant for economic policy for-
mulations are based on the trilemma or impossible trinity argument. The
trilemma says that it is impossible for a country to simultaneously main-
tain free capital mobility, active monetary policy and the ability to manage
the exchange rate. One of these features must necessarily be given up. In
other words, the trilemma says that in an economy open to capital fows
it is impossible for the authorities to simultaneously control the exchange
rate and the interest rate (or the monetary base).
There are at least two ways to express the objection to the SCRER
policy based on the trilemma. One of them argues that targeting the
exchange rate implies a central bank intervention in the foreign exchange
market. In doing so, the central bank loses its ability to control money
supply. Targeting the exchange rate and controlling the money supply can
be simultaneously pursued only if capital fows are regulated. However,
the efectiveness of capital regulation tends to decrease, because the
private sector innovative capacity is greater than the public sector regula-
tory ability. The conclusion is that central banks have to choose between
two poles (Fischer, 2001): active monetary and foating exchange rate or
hard peg cum passive monetary policy.
The second way to express the objection focuses on the argument of
controlling infation. If the interventions in the exchange market target
the real exchange rate (instead of the nominal exchange rate), no nominal
anchor remains for the public to confgure infationary expectations. Since
the central bank cannot control the money supply, the infation rate is
completely out of control.
The trilemma is essentially a policy argument, logically derived from
interest rate parity theorems for open economies. When forwards exchange
markets are not fully developed the relevant theorem is the uncovered
interest parity (UIP) condition. The UIP states that the returns of two
perfect substitutes assets nominated in diferent currencies should be equal.
This implies that the domestic interest rate (i) should equalize the sum of
the foreign interest rate (i*) and the expected variation of the nominal
exchange rate (E(S
#
) 5 S
E
t11
2 S
t
/S
t
). With the additional assumptions of
small country (i*
t
5 i*) and perfect foresight (S
E
t11
5 S), the UIP condition
formally implies:
i
t
5 i* 1
S
S
t
2 1 (9.1)
Equation (9.1) is a simple model with two unknowns: the domestic
interest rate and the exchange rate (S
t
). Under a credibly fxed exchange
rate regime (S
t
5 S), the model is solved by determining the interest rate
endogenously equal to the international rate. In other words, the govern-
ment is able to set the exchange rate but loses control of the monetary
policy. When the exchange rates foats freely, Equation (9.1) is solved by
setting the domestic interest rate exogenously. This is the case in which
governments have an active monetary policy at the cost of letting the
exchange rate foat. If both the interest rate and the exchange rate are
exogenous, Equation (9.1) is overdetermined. The only way to avoid this
situation is to consider the imposition of capital controls, which prevent
arbitrage forces to make the parity hold.
In any model conclusions critically depend on the assumptions. In
the case of the trilemma one crucial assumption is that assets are perfect
substitutes. If this assumption is relaxed the validity of the trilemma as
a general theorem characterizing the performance of economies open to
capital fows no longer holds.
15
Moreover, it has been recognized for a long
time in open economy macroeconomics that in the context of free capital
mobility central banks have room to conduct active monetary policy and
control the nominal exchange rate when assets are imperfect substitutes.
16

The degrees of freedom of monetary policy vary inversely with the degree
of assets substitutability.
The degrees of freedom of monetary policy also depend on the insti-
tutional characteristics of the central bank, and the situation of the FX
market. In a case of excess supply of foreign exchange at the targeted
exchange rate, if the central bank is allowed to issue bonds to sterilize, it
can control both the prevailing exchange and interest rates by purchasing
all the excess supply of international currency in the FX market and ster-
ilize the monetary efect of that intervention through the issuing of bonds
in the monetary market. The central bank has two available instruments to
perform its two targets: the intervention in the exchange market to control
the exchange rate and the intervention in the money market to control the
interest rate. Tinbergens maxim is fulflled. The excess supply of internat-
ional currency, at the exchange rate targeted by the central bank, implies
an excess demand for domestic assets at the prevailing domestic interest
rate. The fully sterilized intervention in the exchange market can be imag-
ined as a policy implemented in two steps. In the frst one, before steriliz-
ation, the central bank intervention generates a monetary base expansion.
The resulting situation would show a higher amount of monetary base,
the same amount of domestic bonds and an interest rate lower than the
initial one. In the second step, the complete sterilization fully compensates
for the change in the private portfolio that took place in the frst step. The
central bank absorbs the increment in the monetary base and issues an
amount of domestic assets equal to the initial excess demand for domestic
assets (the excess supply of international currency) turning the domestic
interest rate to its previous level (Bofnger and Wollmerhuser, 2003).
Therefore, if assets are imperfect substitutes and sterilization is allowed,
the central banks ability to simultaneously manage the exchange rate and
the interest rate critically depends on the existence of an excess supply of
international currency at the targeted exchange rate. In this setting the
trilemma is invalid. It seems that this conclusion is not generally acknowl-
edged because the literature discussing monetary autonomy and exchange
regimes rarely considers situations of excess supply of international cur-
rency. It is mostly focused on balance of payments defcit situations.
17
Certainly, in excess demand contexts the predictions of the trilemma are
generally valid. Even when assets are imperfect substitutes, in these situ-
ations even powerful central banks have a limited capacity to intervene
in the FX market. The limit is determined by the stock of international
reserves. Consequently, it may be argued that even powerful central
banks cannot simultaneously control the exchange rate and the interest
rate in contexts of excess demand for international currency. But there is
no symmetry between excess demand and excess supply situations. In the
frst case the trilemma is valid while not necessarily in the second one. The
asymmetry lies in the fact that in the frst case sterilization is constrained
by a fxed stock (that is, the international reserves), while in the second
sterilization may be done indefnitely because of a variable stock (that is,
central banks bonds). The central banks ability to issue bonds but not
international reserves is the key diference.
This ability raises the question whether it is possible to carry the fully
sterilized intervention policy under excess supply of foreign currency situ-
ations permanently. In order to do so, the central bank has to fulfll a sus-
tainability condition: its net worth should not follow an explosive trend.
Sustainability, therefore, depends on the magnitudes of the international
and the domestic interest rates and on the rate of variation of the nominal
exchange rate. Taken as given the international interest rate and the trend
of the nominal exchange rate, the sustainability condition depends on the
domestic interest rate. The central bank enjoys autonomy to determine the
domestic interest rate, but in order to be sustainable the policy must deter-
mine domestic interest rates lower than a certain upper limit. This limit
can be formally determined as follows.
18
Assume a central bank that holds
international reserves (R) as its unique asset and issues monetary base
(H) and remunerated liabilities (L) yielding the domestic interest rate set
by the monetary authority (i
t
). Therefore, the central banks net worth
(N) at any point in time would be:
N
t
5 S
t
R
t
2 (H
t
1 L
t
) (9.2)
In each period the central bank earns the yielding of international reserves
which for simplicity we assume are invested at the international interest
rate and serves the interest payments of its remunerated liabilities. There
is also a valuation efect on the international reserves due to the variation
of the exchange rate (S
#
). Since the changes in the stocks cancel out, central
banks quasi-fscal result is equal to the variation of its net worth.
dN 5 SR(i* 1 S
#
) 2 iL (9.3)
A simple (although restrictive) condition for the central banks net worth
not to follow an explosive trend is to assume that the quasi-fscal result
has to be non-negative (dN $ 0). Under this sustainability condition, we
obtain the maximum domestic interest rate that makes the fully sterilized
intervention policy sustainable:
i
max
t
5
i* 1 S
#
L
t
/S
t
R
t
(9.4)
It follows that there is a range of interest rates from zero to i
max
that makes
the fully sterilized intervention policy sustainable. Given that central
banks typically beneft from seigniorage and infation tax revenues, the
case in which L
t
, S
t
R
t
does not seem unlikely. In these cases, the upper
limit of this range would be greater than the sum of the international inter-
est rate and the rate of variation of the nominal exchange rate.
It is important to notice that since i
max
depends on the behavior of
R
t
and L
t
, the range of sustainable interest rates also evolves over time.
Given a set of variables and parameters of the economy (such as the infa-
tion rate, the elasticity of money demand and the rate of variation of the
exchange rate), i
max
would tend to decrease as the interest rate set by the
central bank increases. Thus, in order to keep the policy in a sustainable
trend, the cumulative sterilization cost should be bounded and manage-
able. A key point for sustainability is, therefore, that the domestic interest
rates set by the central bank should be moderate in the mentioned sense.
9.4 CONCLUDING REMARKS
In this chapter we show that monetary and exchange rate policies target-
ing a SCRER are viable for developing open economies. We illustrated
our argument with recent Argentine experience, which is just one of
many other economies like China or India following this strategy. It is
important to notice, however, that when a country is trying to preserve
a SCRER, monetary, exchange rate and fscal policies should be coor-
dinated. Otherwise, potential conficts between domestic goals such as
the exchange rate, infation rate and employment might arise. Recent
infationary pressures in Argentina could be an example of conficts arising
from the lack of coordination between economic policies.
An outline of a macroeconomic regime targeting a SCRER in which
monetary, exchange rate and fscal policies are coordinated is briefy
described as follows.
19
First, it is important to mention that such a macro-
economic regime does not imply segmentation between objectives and
instruments. The preservation of a SCRER, the level of employment
and the control of infation set the priorities and the restrictions that the
economic policy must fulfll. Monetary, exchange rate and fscal policies
should be coordinated in order to guarantee the consistency between the
multiple objectives.
The exchange rate policy should focus on signaling the stability of the
real exchange rate in the medium and long term, in order to set in motion
the positive feedbacks mentioned in Section 9.3. In particular, the emer-
gence of appreciation trends should be avoided to prevent self-fulflling
bubbles that increase the monetary costs of buying FX interventions and
also because real exchange rate appreciation may harm the proftability of
tradable activities, making many of them non-viable and forcing frms to
close.
The preservation of a SCRER does not mean short-run indexation of
the nominal exchange rate to domestic prices. The fexibility and advan-
tages of foating the nominal exchange rate in the short run should be pre-
served. Central bank interventions in the FX market have to achieve two
conficting goals: they have to prevent expectations of real exchange rate
appreciation and allow the nominal exchange rate to foat in order to dis-
courage short-term speculative capital fows. The interval of interventions
has to be narrow enough to perform the frst function and wide enough to
perform the second.
The FX market behaves like an asset market. Buying and selling decis-
ions are mostly based on expectations. If central bank interventions and
signals stabilize expectations around the SCRER a necessary condition
for that is the consistency of monetary, exchange rate and fscal policies
and the robustness of the external sector accounts the market forces by
themselves will tend to stabilize the exchange rate. The monetary costs
of central bank interventions will be lower and fewer interventions will
be required. For this reason, interventions should be frm, in order to
clearly show to the market the willingness and strength of the monetary
authority.
It is implicit in the above presentation of the exchange rate policy that
the buying and selling fows of international currency are manageable.
This means that the central bank manage to keep the policy in a sustain-
able path. If capital infows are massive the cost of sterilization may turn
the monetary policy unsustainable. It is important to notice, however, that
such a situation might arise as an endogenous consequence of the exchange
rate policy itself. Massive capital infows may result from an excessively
stable nominal exchange rate in the short run, which turns speculative
investments in one-way bets. Short-run volatility in the exchange rate
increases the uncertainty of speculative investments and thus may reduce
capital infows, diminishing the amount and cost of interventions. This
is the main reason why the exchange rate policy under a SCRER regime
should preserve short-run volatility.
However, massive capital infows that make the policy unsustainable
may occur even when the central bank induces short-run volatility in
the exchange rate. This could happen in the context of high liquidity in
international capital markets. In this kind of situations, it would make
little sense to risk macroeconomic stability in order to preserve the capital
account full openness principle. The preservation of the macroeconomic
regime requires in this case capital account regulations, intended to restrict
capital infows and facilitate the management of exchange and monetary
policies. There is a menu of measures able to accomplish this function that,
even when they do not work perfectly well, evidence suggests that they
contribute to soften capital infows during booms.
20
The need for controls
is not permanent; they have to do their job only in a booming phase, and
we now know well that booming phases do not last forever.
When there is an excess demand for international currency that turns
exchange and monetary policies unmanageable, FX interventions would
cause an excessive monetary contraction and the consequent rise in the
interest rate would trigger a recession. The defense of some nominal
exchange rate may risk a speculative attack on the central bank reserves.
The situation has similarities with a fxed exchange rate regime crisis.
But there is an important diference. If there are no fundamental reasons
for depreciation generated, for instance, by expectation of balance of
payments defcit fscal and monetary policies are consistent with the
targeted real exchange rate, and infation is under control, then the macro-
economic regime should be preserved. This would only be possible if
exchange controls and restrictions on capital outfows were imposed. If
there are no fundamental reasons inducing the excess demand for inter-
national currency, there is no need for the controls and regulations to
last for long. As described in Section 9.2, Argentina successfully imposed
exchange controls and capital outfow regulations in mid 2002, when the
run into foreign currency was mainly caused by a self-fulflling bubble in
the exchange rate. The measures were gradually softened when the buying
pressure in the FX market diminished.
In a SCRER macroeconomic regime, monetary policy should not be
exclusively focused on infation. It is important to emphasize, however,
that this regime performs a preventative role with respect to infation
acceleration. In most developing economies the exchange rate is the main
transmission mechanism of monetary impulses to the infation rate. The
SCRER precisely encourages the central bank to implement monetary
policies that avoid excessive fuctuations in both the nominal and real
exchange rates. In contrast, for the same reason, an exclusive infation
focus of monetary policy generates incentives towards real exchange rate
appreciation.
In coordination with the other policies, monetary policy should be
managed in order to attain multiple objectives. To manage monetary
aggregates or set the interest rate to accomplish this goal, the central bank
may have to compensate for the interventions in the FX market. Out of
the extreme situations discussed above, this can be done through diferent
instruments. The most common is the sterilization operations. In Section
9.3 we showed that fully sterilized interventions can be carried out in the
context of excess supply of foreign exchange and we derived the range of
interest rates that make the policy sustainable in time.
Apart from sterilization, there are other instruments at hand to conduct
the monetary policy under the SCRER regime. For instance, a central
bank in possession of a signifcant amount of bank debt can manage it as
an instrument for monetary control (Lavoie, 2001). Public sector deposits
in the central bank can be used in an analogous way. Some prudential
regulations can be oriented to the same target, particularly when the
problem is to constraint money expansion. The central bank, for instance,
can raise the cash requirements of the banking system and thus lower the
banking multiplier. Other prudential regulations can be directly focused
on smoothing the selling pressure in the exchange market. For instance,
if local banks are not allowed to back credits in domestic currency with
liabilities in international currency and credits in international currency
are limited, there are fewer incentives to the banks procuring international
funding. The existence of public banks with a signifcant share of the fnan-
cial market can facilitate the monetary management. Public banks can be
coordinated in order to help the central bank in both the management of
the liquidity and the FX interventions. Through the management of these
instruments the central bank should be able to keep money expansion
under control.
Finally, fscal policy should complement monetary policy in attaining
infation and employment targets in the short run. It should focus on
the management of nominal aggregate demand: moderating it in cases
of infationary pressures and expanding it in the opposite situations.
However, it is important to notice that since a SCRER regime is meant
to promote development and growth, infationary pressures may be more
likely than defationary ones. Since in this regime monetary policy typi-
cally has an expansionary bias through the buying interventions in the FX
market, the role of moderating the aggregate demand to avoid infation
and real exchange rate appreciation would tend to rely on the fscal policy.
Conservative fscal policy could also be necessary in the SCRER regime to
ease the central bank needs to intervene in the FX market. Public sector
surpluses may be used to buy part of the excess supply in the FX market.
An anti-cyclical fscal fund intended to perform this role could be a good
institution to develop in a SCRER regime.
NOTES
1. The authors would like to thank Nelson Barbosa-Filho, Erinc Yeldan, Jan Kregel and
the participants in the workshop on Alternatives to Infation Targeting Monetary
Policy for Stable and Egalitarian Growth in Developing Countries held at CEDES
13-14 May 2005 for their comments to previous versions of this chapter. We also thank
Martn Fiszbein for his collaboration. The usual caveats apply.
2. In 1992 the new central bank law slightly relaxed this constraint by setting narrow
margins to the possibilities of purchasing public bonds and lending to the commercial
banks.
3. For an analysis of the macroeconomic performance during the convertibility period, see
Damill and Frenkel (2007).
4. An analysis of the evolution of Argentine debt, default and restructuring can be found
in Damill et al. (2005).
5. By mid 2007 energy supply shortages raised concerns about the sustainability of high
rates of economic growth. These concerns, however, are not related to the SCRER
policy.
6. Later on, the government issued new debt to compensate the banks for the balance
sheet efect of the asymmetric pesoifcation.
7. Figure 9.1 shows a jump in the private bank deposit series in January 2002. It refects
the accounting efect of the pesoifcation at 1.40 pesos per dollar of deposits issued in
foreign currencies, previously valued at a AR$/$ 1 rate. If we put this mere accounting
efect aside, it is easy to see the drop in deposits.
8. Exchange rates are defned so that a rise in this variable implies a nominal or real
depreciation.
9. They included the obligation to surrender the proceeds from exports in the local FX
market.
10. For instance, the efects of the interest rate through the credit channel are very weak in
an economy where the bank credit to private sector remains below 13 percent of GDP
as in Argentina.
11. The literature emphasizing the positive efects of the SCRER on development increases
day by day. See, for instance, Williamson (2003), Rodrik (2005), Dooley et al. (2004),
Frenkel (2004), Frenkel and Taylor (2005) and Subramanian (2007).
12. See Hausman et al. (2005) and Prasad et al. (2007).
13. See Frenkel and Ros (2006) for an analytical and empirical study of the positive rela-
tionship between real exchange rate and employment in Latin America.
14. Evidence regarding short-run indeterminacy of the real exchange rate seems to be
conclusive. Its behavior is almost completely determined by the nominal exchange
rate. Although most scholars agree about the existence of an equilibrium real
exchange rate in the long run, there is no consensus regarding the factors afecting
its determination. The purchasing power parity (PPP) is the most accepted hypoth-
esis (Taylor and Taylor, 2004). However, evidence regarding the PPP shows time
series reverting to their means in very long periods (that is, half life of 35 years)
and results are highly sensitive to data sets and estimation techniques. On the other
hand, mean-reverting time series is no suf cient condition for the validity of the PPP
hypothesis.
15. Another key assumption is that exchange rate expectations are formed with perfect
foresight. If departures from the perfect foresight-rational expectation paradigm are
considered, the predictions of the trilemma may no longer hold. For a critique of the
trilemma in this vein, see Frenkel and Rapetti (2007).
16. See, for instance, Chapter 10 of Dornbusch (1980).
17. See, as an example, Canales-Kriljenko (2003).
18. The complete model is in Frenkel (2007).
19. For a detailed description of a macroeconomic regime proposal with a SCRER as an
intermediate target, see Frenkel (2006).
20. See Epstein et al. (2003).
REFERENCES
Balassa, B. (1971), Trade policies in developing countries, American Economic
Review, 61 (2), 17887.
Blecker, R. (2005), Financial globalization, exchange rates and international
trade, in G. Epstein (ed.), Financialization in the World Economy, Cheltenham,
UK and Northampton, MA, USA: Edward Elgar, pp.183209.
Bofnger, P. and T. Wollmershuser (2003), Managed foating as a monetary
policy strategy, Economics of Planning, 36 (2), 81109.
Canales-Kriljenko, J.I. (2003), Foreign exchange intervention in developing
and transition economies: results of a survey, International Monetary Fund
working paper no. 03/95.
Damill, M. and R. Frenkel (2007), Argentina: macroeconomic performance and
crisis, in R. Ffrench-Davis, D. Nayyar and J. Stiglitz. (eds), Stabilization Policies
for Growth and Development, Initiative for Policy Dialogue, Macroeconomics
Policy Task Force, New York, forthcoming.
Damill, M., R. Frenkel and M. Rapetti (2005), The Argentinean debt: history,
default and restructuring, Columbia University Initiative for Policy Dialogue
working paper series, New York.
Daz Alejandro, C. (1979), Algunas vicisitudes histricas de las economas abier-
tas en Amrica Latina, Desarrollo Econmico, 19 (74), 14759.
Dooley, M., D. Folkerts-Landau and P. Garber (2003), An essay on the revived
Bretton Woods system, International Journal of Finance and Economics, 9,
30713.
Dooley, M., D. Folkerts-Landau and P. Garber (2004), The US current account
defcit and economic development: collateral for a total return swap, National
Bureau Economic Research working paper no. 10727.
Dornbusch, D. (1980), Open Economy Macroeconomics, New York: Basic Books.
Epstein, G., I. Grabel and K.S. Jomo (2003), Capital management techniques in
for the future, Political Economy Research Institute, working paper no. 56,
University of Massachusetts Amherst.
Feldstein, M. (1999), Self-protection for emerging markets, National Bureau for
Economic Research, working paper no. 6907.
Fischer, S. (2001), Exchange rates regimes: is the bipolar view correct?, Journal of
Frankel, J. and A.K. Rose (1995), Empirical research on nominal exchange rates,
in: G.M. Grossman and K. Rogof (eds), Handbook of International Economics,
vol. III, Amsterdam: Elsevier, pp. 1689709.
Chile and Mexico, paper prepared for the G-24, Washington, DC, 24
August.
Frenkel, R. (2006), An alternative to infation targeting in Latin America:
macroeconomic policies focused on employment, Journal of Post Keynesian
Economics, 28 (4), 57391.
Frenkel, R. (2007), La sostenibilidad de la poltica de esterilizacin, CEPAL
Review, 93 (December), 2936.
America, World Development, 34 (4), 63146.
employment, paper prepared for the High-Level United Nations Development
Conference, 14-15 March, New York.
Frenkel, R. and M. Rapetti (2007), Poltica cambiaria y monetaria despus del
colapso de la Convertibilidad, Ensayos Econmicos, 46, 13766.
Hausman, R., L. Pritchett and D. Rodrik (2005), Growth accelerations, Journal
of Economic Growth, 10 (4), 30329.
Lavoie, M. (2001), The refux mechanism in the open economy, in L.P. Rochon
and M. Vernengo (eds), Credit, Interest Rates and the Open Economy: Essays on
Horizontalism, Cheltenham, UK and Northampton, MA, USA: Edward Elgar,
pp. 21542.
Polterovich, V. and V. Popov (2002), Accumulation of foreign exchange reserves
and long term growth, New Economic School, working paper, Moscow.
Prasad, E., R. Rajan and A. Subramanian (2007), Foreign capital and economic
growth, Brooking Papers on Economic Activity, (1), 15320.
Rodrik, D. (forthcoming), Growth strategies, Brookings Papers on Economic
Activity.
Subramanian, A. (2007), Capital account convertibility: a neglected consider-
ation, Economic and Political Weekly, 42 (25), 241318.
Taylor, A. and M. Taylor (2004), The purchasing power parity debate, Journal of
Economics Perspectives, 18 (4), 13558.
Williamson, J. (2003), Exchange rate policy and development, paper prepared
for the project Capital Market Liberalization Task Force Initiative for Policy
Dialogue, Barcelona, 2 June, 2003.
203
10. A general equilibrium assessment
of twin-targeting in Turkey
Cagatay Telli, Ebru Voyvoda and
A. Erin Yeldan
1
10.1 INTRODUCTION: MACROECONOMICS OF
TWIN-TARGETING IN TURKEY
After a decade of failed reforms and deteriorated macroeconomic per-
formance, Turkey entered the millennium under a Staf Monitoring
Program signed with the International Monetary Fund (IMF) in 1998,
and put into efect in December 1999. The program currently sets the macro-
economic policy agenda in Turkey and relies mainly on two pillars: (1)
fscal austerity that targets a 6.5 percent surplus for the public sector in its
primary budget
2
as a ratio to the gross domestic product (GDP); and (2)
a contractionary monetary policy (through an independent central bank)
that exclusively aims at price stability (via infation targeting). Thus, in
a nutshell the Turkish government is charged to maintain dual targets:
a primary surplus target in fscal balances (at 6.5 percent to the GDP);
and an infation targeting central bank whose sole mandate is to maintain
price stability and is divorced from all other concerns of macroeconomic
aggregates hence the terms in the title: macroeconomics under twin-
targeting.
According to the logic of the program, successful achievement of the
fscal and monetary targets would enhance credibility of the Turkish
government ensuring reduction in the country risk perception. This would
enable reductions in the rate of interest that would then stimulate private
consumption and fxed investments, paving the way to sustained growth.
Thus, it is alleged that what is being implemented is actually an expansion-
ary program of fscal contraction.
On the monetary policy front, the Central Bank of Turkey (CBRT) was
granted its independence from political authority in October 2001. In what
follows, the central bank announced that its sole mandate is to restore and
maintain price stability in the domestic markets and that it will follow a
disguised infation targeting until conditions are ready for full targeting.
Thus, over 2002 and 2003 the CBRT targeted net domestic asset position of
the central bank as a prelude to full infation targeting. Finally on 1 January
2006 the CBRT announced that it will adopt full-fedged infation targeting.
The purpose of this chapter is to provide an assessment of the key macro-
economic developments in Turkey over the post-2001 crisis period and
to provide a general equilibrium analysis of the macroeconomic policy
alternatives of the twin-targeters. We focus on three sets of issues: frst,
we study the macroeconomics of the expanded foreign capital infows in
resolving (temporarily) the macroeconomic impasse between the disinfa-
tion motives of the CBRT and imperatives of debt sustainability and fscal
credibility of the Ministry of Finance. Second, we study the reduction
of the central banks interest rates. Third, we implement a labor market
reform and study the implications of reducing/eliminating payroll taxes
(paid by the employers). To these ends we construct a macroeconomic
general equilibrium model with a full-fedged fnancial sector in tandem
with a real sector. Across all policy simulations we exclusively focus on
both the fscal and fnancial adjustments and study the possible dilemmas
of gains in ef ciency in the labor markets versus the loss of fscal revenues
to the state.
Our fnding is that the current monetary strategy followed by the CBRT
that involves a heavy reliance on foreign capital infows along with a rela-
tively high real rate of interest is efective in bringing infation down; yet
it sufers from increased cost of interest burden to the public sector and
strains fscal credibility. In contrast, our simulation results suggest that,
given the ex ante constraints of the domestic economy in the short run, an
alternative heterodox policy of reduction of the central bank interest rate
and lowering of the payroll tax burden in labor markets may have strong
employment and growth efects. The policy also achieves signifcant gains
in fscal credibility in the short run. Yet it sufers from increased infation-
ary pressures in the commodity and the fnancial markets. Even though
observed to prevail at a modest scale in our simulation experiments, the
ex ante constraints of maintaining infationary expectations may lead to
intolerance of the CBRT and render the policy inefective. Thus, maintain-
ing an integrated and coherent policy framework between the monetary
and fscal authorities is seen of prime importance for the success of the
policy formulation at the macro scale.
Our premise in this chapter is that a proper modeling of the general equi-
librium linkages between the production-income generation and aggregate
demand components across individual sectors as well as responses of
the real macro aggregates to fnancial decisions are essential steps to
understand the impact of the current austerity program on the evolution
of output, fscal, fnancial and external balances, and on employment.
A general equilibrium assessment of twin-targeting in Turkey 205
Accordingly, we develop a computable general equilibrium (CGE) model
with a relatively aggregated productive sector, a segmented labor market
and a full-blown public sector with a detailed treatment of fscal balances
and fnancial fows.
The current model shares many of the analytical structure of the Agnor
et al. (2006) design in the dynamics of fnancial transactions, especially
with respect to formation of expectations and fragility. It is explicitly
designed to capture the relevant linkages between the fscal policy decis-
ions, fnancialization constraints and external balances that we believe
are essential to analyse the impact of disinfation and fscal reforms on
labor market adjustment and public debt sustainability. We pay particular
attention to fscal issues such as a high degree of debt overhang and fscal
dominance; the link between real and fnancial sector interactions, and
interactions between external (current account) defcits private saving-
investment defcits and the public (primary balance) surpluses.
We organize the chapter under four sections. First, we provide a broad
overview of the recent macroeconomic developments in Turkey in Section
10.2. Here we study, exclusively, the evolution of the key macroeconomic
prices such as the exchange rate, the interest rate and price infation. Here
we also comment on the external balances, the dynamics of external debt,
fscal policy issues and the labor market. In Section 10.3, we introduce
and implement our CGE modeling analysis of the alternative policy scen-
arios to depict the short-run macroeconomic adjustments of the Turkish
economy under the conditionalities of the IMF program targets on
primary surplus to GDP ratio and on infation rate. Finally, we provide a
brief summary with concluding comments in Section 10.4.
10.2 MACROECONOMIC DEVELOPMENTS UNDER
IMFS STAFF MONITORING
The growth path of the Turkish economy over the post-1998 period had been
erratic and volatile, mostly subject to the fows of hot money. Following the
contagion efects of the Asian, Russian and Brazilian turmoil, the economy
frst decelerated in 1998 with a growth rate of 3.1 percent, and then con-
tracted in 1999 at the rate of 5.0 percent. The boom of 2000 was followed
by the 2001 crisis. The recovery was sharp as the economy has grown at an
average rate of 7.1 percent over the 200206 period. Price movements were
also brought under control through the year and the 12-month average
infation rate in consumer prices has receded from 45 percent in 2002 to 7.7
percent in 2005, and from 50.1 percent to 5.9 percent in producer prices.
The post-2003 period has also meant a period of acceleration of exports,
where export revenues reached $91.7 billion in 2006. Nevertheless, with
the rapid rise of the import bill over the same period, the defcit in the
current account reached $31.7 billion (or about 7.9 percent of the gross
national product (GNP) in 2006). The current account defcit continued to
widen in 2007 and reached $34 billion over 12 months cumulative period
in the frst quarter. On the public sector front one witnesses a very strong
fscal discipline efort. The ratio of central government budget defcit to
the GNP was reduced from its peak of 16.2 percent in 2001 to 0.8 percent
by 2006. Consequently, the public sector borrowing requirement (PSBR)
as a ratio to the GNP fell from 16.1 percent to negative 3 percent, indicat-
ing a surplus, in 2006. Table 10.1 documents the main macro indicators of
the post-1998 Turkish economy under close IMF supervision.
10.2.1 Macroeconomic Prices and the Monetary Policy
The CBRT initiated an open infation targeting framework starting 1
January 2006. The CBRTs current mandate is to set a point target of
5 percent infation of the consumer prices. Given internal and external
shocks, the CBRT has recognized an internal (of 1 percent) and an external
(of 2 percent) uncertainty band around the point target. Thus, the CBRT
will try to keep the infation rate at its point target; however, recognizing a
band of maximum 2 percentage points below or above the 5 percent target
rate. The CBRT has announced that it will continue to use the overnight
interest rates as its main policy tool to reach its target. It is stated explicitly
that the sole objective of the CBRT is to provide price stability, and that
all other possible objectives are out of its policy realm.
3
Despite the positive achievements on the disinfation front, rates of inter-
est remained slow to adjust. The real rate of interest remained above 10
percent for much of the post-2001 crisis era, and generated heavy pressures
against the fscal authority in meeting its debt obligations (Table 10.1). The
persistence of the real interest rates, on the other hand, had also been condu-
cive in attracting heavy fows of short-term speculative fnance capital over
2003 and 2006. This pattern continued into 2007 at an even stronger rate.
Inertia of the real rate of interest is enigmatic from the successful macro
economic performance achieved thus far on the fscal front. The credit
interest rate, in particular, had been constrained by a lower bound of 16
percent despite the deceleration of price infation. Consequent to the fall
in the rate of infation, the inertia of credit interest rates translates into
increasing real costs of credit.
High rates of interest were conducive in generating a high infow of hot
money fnance to the Turkish fnancial markets. The most direct efect of
the surge in foreign fnance capital over this period was felt in the foreign
207
T
a
b
l
e

1
0
.
1

B
a
s
i
c

c
h
a
r
a
c
t
e
r
i
s
t
i
c
s

o
f

t
h
e

T
u
r
k
i
s
h

e
c
o
n
o
m
y

u
n
d
e
r

t
h
e

I
M
F

s
u
r
v
e
i
l
l
a
n
c
e
,

1
9
9
8
2
0
0
6
S
t
a
f

m
o
n
i
-
t
o
r
i
n
g

p
r
o
g
r
a
m

i
n
i
t
i
a
t
e
d
C
o
n
t
a
-
g
i
o
n

o
f

e
m
e
r
g
i
n
g

m
a
r
k
e
t

f

n
a
n
c
i
a
l

c
r
i
s
e
s
I
M
F
-
d
i
r
e
c
t
e
d

d
i
s
i
n
-
f

a
t
i
o
n

p
r
o
g
r
a
m
F
i
n
a
n
c
i
a
l

c
r
i
s
i
s
I
M
F
-
D
i
r
e
c
t
e
d

P
o
s
t
-
C
r
i
s
i
s

A
d
j
u
s
t
m
e
n
t
s
U
n
d
e
r

t
h
e

t
h
r
e
e
-
p
a
r
t
y

c
o
a
l
i
t
i
o
n

g
o
v
e
r
n
-
m
e
n
t
U
n
d
e
r

t
h
e

p
r
a
g
m
a
t
i
c

a
n
d

w
e
s
t
e
r
n
-
f
r
i
e
n
d
l
y

I
s
l
a
m
i
s
m

o
f

t
h
e

A
K
P
1
9
9
8
1
9
9
9
2
0
0
0
2
0
0
1
2
0
0
2
2
0
0
3
2
0
0
4
2
0
0
5
2
0
0
6
R
e
a
l

R
a
t
e

o
f

G
r
o
w
t
h
G
D
P
3
.
1
5
.
0
7
.
4
7
.
4
7
.
6
5
.
8
8
.
9
7
.
4
6
.
1
C
o
n
s
u
m
p
t
i
o
n

E
x
p
e
n
d
i
t
u
r
e
s

P
r
i
v
a
t
e
0
.
6
2
.
6
6
.
2
9
.
2
2
.
0
6
.
6
1
0
.
1
8
.
8
5
.
2

P
u
b
l
i
c
7
.
8
6
.
5
7
.
1
8
.
6
5
.
4
2
.
4
0
.
5
2
.
4
9
.
6
I
n
v
e
s
t
m
e
n
t

E
x
p
e
n
d
i
t
u
r
e
s

P
r
i
v
a
t
e
8
.
3
1
7
.
8
1
6
.
0
3
4
.
9
7
.
2
2
0
.
3
4
5
.
5
2
3
.
6
1
7
.
4

P
u
b
l
i
c
1
3
.
9
8
.
7
1
9
.
6
2
2
.
0
1
4
.
5
1
1
.
5
4
.
7
2
5
.
9
0
.
2
E
x
p
o
r
t
s
1
2
.
0
7
.
1
1
9
.
2
7
.
4
1
1
.
0
1
6
.
0
1
2
.
5
8
.
5
8
.
5
I
m
p
o
r
t
s
2
.
3
3
.
7
2
5
.
4
2
4
.
8
1
5
.
7
2
7
.
1
2
4
.
7
1
1
.
5
7
.
1
M
a
c
r
o
e
c
o
n
o
m
i
c

B
a
l
a
n
c
e
s

(
a
s

r
a
t
i
o

t
o

t
h
e

G
N
P
,

%
)
A
g
g
r
e
g
a
t
e

D
o
m
e
s
t
i
c

S
a
v
i
n
g
s
2
2
.
7
2
1
.
2
1
8
.
2
1
7
.
5
1
9
.
2
1
9
.
3
2
0
.
2
1
7
.
1
1
6
.
6
A
g
g
r
e
g
a
t
e

F
i
x
e
d

I
n
v
e
s
t
m
e
n
t
s
2
4
.
3
2
2
.
1
2
2
.
8
1
9
.
0
1
7
.
3
1
6
.
1
1
8
.
4
2
0
.
3
2
3
.
1
B
u
d
g
e
t

B
a
l
a
n
c
e
7
.
0
1
1
.
6
1
0
.
9
1
6
.
2
1
4
.
3
1
1
.
2
7
.
1
2
.
0
0
.
8
P
u
b
l
i
c

S
e
c
t
o
r

B
o
r
r
o
w
i
n
g

R
e
q
u
i
r
e
m
e
n
t
9
.
3
1
5
.
5
1
1
.
8
1
6
.
4
1
2
.
7
9
.
3
4
.
7
0
.
4
3
.
0
C
u
r
r
e
n
t

A
c
c
o
u
n
t

B
a
l
a
n
c
e
1
.
0
0
.
7
4
.
8
2
.
4
0
.
8
3
.
4
5
.
2
6
.
2
7
.
9
S
t
o
c
k

o
f

F
o
r
e
i
g
n

D
e
b
t
5
5
.
4
7
1
.
0
6
3
.
4
9
2
.
7
7
7
.
5
5
7
.
1
5
0
.
4
4
6
.
9
5
0
.
4
208
T
a
b
l
e

1
0
.
1

(
c
o
n
t
i
n
u
e
d
)
S
t
a
f

m
o
n
i
-
t
o
r
i
n
g

p
r
o
g
r
a
m

i
n
i
t
i
a
t
e
d
C
o
n
t
a
-
g
i
o
n

o
f

e
m
e
r
g
i
n
g

m
a
r
k
e
t

f

n
a
n
c
i
a
l

c
r
i
s
e
s
I
M
F
-
D
i
r
e
c
t
e
d

d
i
s
i
n
-
f

a
t
i
o
n

p
r
o
g
r
a
m
F
i
n
a
n
c
i
a
l

c
r
i
s
i
s
I
M
F
-
D
i
r
e
c
t
e
d

P
o
s
t
-
C
r
i
s
i
s

A
d
j
u
s
t
m
e
n
t
s
U
n
d
e
r

t
h
e

t
h
r
e
e
-
p
a
r
t
y

c
o
a
l
i
t
i
o
n

g
o
v
e
r
n
-
m
e
n
t
U
n
d
e
r

t
h
e

p
r
a
g
m
a
t
i
c

a
n
d

w
e
s
t
e
r
n
-
f
r
i
e
n
d
l
y

I
s
l
a
m
i
s
m

o
f

t
h
e

A
K
P
1
9
9
8
1
9
9
9
2
0
0
0
2
0
0
1
2
0
0
2
2
0
0
3
2
0
0
4
2
0
0
5
2
0
0
6
M
a
c
r
o
e
c
o
n
o
m
i
c

P
r
i
c
e
s
R
a
t
e

o
f

C
h
a
n
g
e

o
f

t
h
e

N
o
m
i
n
a
l

E
x
c
h
a
n
g
e

R
a
t
e

(
T
L
/
S
)
7
1
.
7
6
0
.
6
2
8
.
6
1
1
4
.
2
2
3
.
0
0
.
6
4
.
9
5
.
7
6
.
9
I
n
f

a
t
i
o
n

(
P
P
I
)
7
1
.
8
5
3
.
1
5
1
.
4
6
1
.
6
5
0
.
1
2
5
.
6
1
4
.
6
5
.
9
9
.
4
I
n
f

a
t
i
o
n

(
C
P
I
)
8
4
.
6
6
4
.
8
5
4
.
9
5
4
.
4
4
4
.
9
2
5
.
3
1
0
.
6
7
.
7
9
.
6
R
e
a
l

I
n
t
e
r
e
s
t

R
a
t
e

o
n

G
D
I
s
a
2
9
.
5
3
6
.
8
4
.
5
3
1
.
8
9
.
1
1
5
.
4
1
3
.
1
1
0
.
4
7
.
9
R
e
a
l

W
a
g
e

G
r
o
w
t
h

R
a
t
e
s
b

P
r
i
v
a
t
e

S
e
c
t
o
r
0
.
9
8
.
6
2
.
6
1
4
.
4
5
.
0
0
.
5
4
.
8
1
.
6
1
.
9

P
u
b
l
i
c

S
e
c
t
o
r
5
.
5
1
8
.
3
1
5
.
6
1
1
.
5
0
.
5
5
.
3
4
.
7
7
.
9
3
.
0
N
o
t
e
s
:
a
.

D
e
f

a
t
e
d

b
y

t
h
e

P
r
o
d
u
c
e
r

P
r
i
c
e

I
n
d
e
x
.
b
.

B
a
s
e
d

o
n

r
e
a
l

w
a
g
e

i
n
d
e
x
e
s

(
1
9
9
7

5

1
0
0
)

i
n

m
a
n
u
f
a
c
t
u
r
i
n
g

p
e
r

h
o
u
r

e
m
p
l
o
y
e
d
,

T
u
r
k
s
t
a
t

d
a
t
a
.
S
o
u
r
c
e
:

S
P
O

M
a
i
n

E
c
o
n
o
m
i
c

I
n
d
i
c
a
t
o
r
s
;

U
n
d
e
r
s
e
c
r
e
t
e
r
i
a
t

o
f

T
r
e
a
s
u
r
y
,

M
a
i
n

E
c
o
n
o
m
i
c

I
n
d
i
c
a
t
o
r
s
;

C
B
R
T

d
a
t
a

d
i
s
s
e
m
i
n
a
t
i
o
n

s
y
s
t
e
m
.
exchange market. The overabundance of foreign exchange supplied by the
foreign fnancial arbiters seeking positive yields led signifcant pressures
for the Turkish lira to appreciate. As the CBRT has restricted its mon-
etary policies only to the control of price infation, and left the value of
the domestic currency to be determined by the speculative decisions of the
market forces, the lira appreciated by as much as 40 percent in real terms
against the US dollar and by 25 percent against the euro (in producer price
parity conditions, over 200206).
The overvaluation of the lira was the most important contributor in
reducing the burden of an ever-expanding foreign indebtedness. While the
aggregate foreign debt stock has increased from US$113.6 billion in 2001
to US$206.5 billion by the end of 2006, as a ratio to the GNP it has created
an illusionary tendency to fall when measured in the overvalued lira units.
10.2.2 Fiscal Policy and Debt Management
The current fscal policy stance in Turkey relies primarily on expenditure
restraint. On the revenue side one witnesses a signifcant efort in raising
tax revenues, both in real terms and also as a ratio to the GNP. Much of
this efort can be explained by the rise in the share of indirect/excise taxes
on goods and services (to 21 percent as a ratio to the GNP, or about 70
percent of total tax revenues), while the contribution of direct income
taxes to the budgetary revenues are observed to fall especially after 2000.
Data reveal a secular fall in the budget defcit through the post-2001
crisis adjustments and is now reduced to less than 1 percent to the GNP.
As discussed above, much of the aggregate budget expenditures can be
explained by the high costs of debt servicing, and the main logic of the
current austerity program rested on maintaining the debt turnover via
only primary surpluses. As a result, the boundaries of the public space are
severely restricted, and all fscal policies are directed to securing debt ser-
vicing at the cost of extraordinary cuts in public consumption and invest-
ments. Within total expenditures, public investments share has fallen
from 12.9 percent in 1990 to 5.1 percent in 2003. As a ratio to the GNP,
public investments stand at less than 2 percent currently.
All of these painful adjustments on the fscal front can be contrasted
against the gains over the existing debt burden of the public sector. Data
from the Ministry of Finance
4
reveal that, as a ratio to the GNP, gross
public debt of the aggregate public sector has fallen from 68.1 percent in
2000 to 63.1 percent by the end of 2006, a decline of only 5 percentage
points. This could have been achieved despite the very rapid rise in the
rate of growth of GNP (7.2 percent per annum over the whole period),
and the very strict fscal austerity measures of primary surplus targets (of
6.5 percent to the GNP for 2002 and beyond). Furthermore much of this
decline has come only after 2005, and all of it is due to the decline in the
ratio of foreign debt to the GNP. As a ratio to the GNP, public external
debt declined from 25.2 percent in 2000 to 16.9 percent in 2006; while the
domestic debt burden increased from 43.1 percent to 46.2 percent over the
same period. It is a clear fact that the illusion of falling foreign indebted-
ness is a direct outcome of the real appreciation of the Turkish lira. As the
increased external indebtedness of the public sector from $47.6 billion in
2000 to $69.6 billion in 2006, its ratio to the GNP had the efect of a fall
when denominated in appreciated liras.
In fact, appreciation of the lira disguises much of the fragility associated
with both the level and the external debt induced fnancing of the current
account defcits. A simple purchasing power parity (PPP) correction of
the real exchange rate, for instance, would increase the burden of external
debt to 76.8 percent as a ratio to the GNP in 2005.
5
This would bring the
debt burden ratio to the 2001 pre-crisis level. Under conditions of the
foating foreign exchange regime, this observation reveals a persistent
fragility for the Turkish external markets, as a possible depreciation of the
lira may severely worsen the current account fnancing possibilities.
10.2.3 Persistent Unemployment and Jobless Growth
Yet the most striking observation on the Turkish labor markets over the
post-2001 crisis era is the sluggishly slow performance of employment gen-
eration capacity of the economy. Despite the very rapid growth perform-
ance across industry and services, employment growth has been meager.
This observation, which actually is attributed to many developing econ-
omies as well,
6
is characterized by the phrase jobless growth in the litera-
ture. The rate of open unemployment was 6.5 percent in 2000, increased to
10.3 percent in 2002, and remained at that plateau despite the rapid surges
in GNP and exports. Open unemployment is a severe problem, in particu-
lar, among the young urban labor force reaching 24.5 percent in 2005.
On the other hand, the participation rate fuctuates around 48 percent
to 50 percent, due mostly to the seasonal efects. It is known, in general,
that the participation rate is less than the EU averages. This low rate is
principally due to women choosing to remain outside the labor force, a
common feature of Islamic societies, but its recent debacle depends as
much on the size of the discouraged workers who had lost their hopes for
fnding jobs. According to Turkish Statistical Institute (Turkstat) data,
the excess labor supply (unemployed plus underemployed) is observed to
reach 13.6 percent of the labor force by the end of 2006 (Figure 10.1).
Thus, to conclude, two important characteristics of the post-crisis
adjustment path stand out. First, the post-2001 expansion is observed to
be concomitant with a deteriorating external disequilibrium, which in turn
is the end result of excessive infows of speculative fnance capital. Second,
the output growth contrasts with persistent unemployment, warranting
the term jobless growth.
The foregoing facts bring the following tasks to our agenda: (1) What
are the viable policy choices in combating unemployment in the short run,
and under the conditionalities of the twin targets? (2) Given our assess-
ments of fragility conditions currently prevailing in Turkey, what are the
short-run efects of a reduction in the interest cost of the central bank
credit in terms of output, employment, foreign indebtedness and other
macro aggregates?
We now turn to the analytics of general equilibrium with the aid of our
CGE model to study these questions.
10.3 COMPUTABLE GENERAL EQUILIBRIUM
MODELING ANALYSIS
Given the overview of the recent macroeconomic developments, we now
develop a real-fnancial CGE model for Turkey. In what follows, we
provide a birds-eye overview of the model, and invite the interested reader
to the Political Economy Research Institute website for a full algebraic
description.
7
0
2
4
6
8
10
12
14
16
18
20
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
U
n
e
m
p
l
o
y
m
e
n
t

(
%
)
45
46
47
48
49
50
51
52
53
54
P
a
r
t
i
c
i
p
a
t
i
o
n

r
a
t
e

(
%
)
Total unemployment rate (%)
Labor participation rate
Source: Authors calculation using Turkish Statistical Institute (Turkstat), Household
Labor Force Surveys database.
Figure 10.1 Labor participation rate and total unemployment
10.3.1 The Algebraic Structure of the Model and Adjustment Mechanisms
Product markets
The model is fairly aggregate over its microeconomic structure but accom-
modates a relatively detailed treatment of the public accounts, and of real-
fnancial sector linkages. There are four production sectors as agriculture,
industry, private services and public services. There is a fnancial sector
with a full-fedged banking segment, a central bank, enterprises, govern-
ment and household portfolio instruments.
Sectoral production is modeled via multilevel functions. At the top level
total output is given as a Leontief specifcation of value-added and inter-
mediate inputs. The value-added in each sector is generated by combining
labor, as well as public and private physical capital. At the last stage of
this multilevel production a sector-specifc public capital combines with
the composite input under a Cobb-Douglas specifcation. The composite
primary input, in turn, is defned to be a combination of private capital
and labor aggregate L
i
through a constant elasticity of substitution (CES)
type of production function.
Public capital is assumed to be fxed and sector-specifc. Private capital
is mobile across sectors and the movement is directed by the diference
in the diferentiated private proft rates. Labors wage rate is fxed in the
short run and the labor market clears through quantity adjustments on
employment.
Households save a fraction 0 , s
P
, 1 of their disposable income. The
saving rate is considered to be a positive function of the expected real
interest rate in domestic currency denominated deposits:
s
P
5 s
P
0
a
1 1 intD
1 1 E[Inf ]
b
s
H
SAV
(10.1)
with E[Inf ], the expected infation rate and s
p
0
is a scaling parameter. The
portion of income that is not saved is allocated to consumption and that
total fow of savings of the household is channeled to the accumulation of
household fnancial wealth.
Private capital investment is assumed to depend on a number of factors.
The frst is the growth rate of real GDP, which captures the regular accel-
erator efect. This efect is positive. The next one is the negative efect of
the expected real cost of borrowing from the domestic banks. Specifcally,
private investment demand is represented by:

PK
#
P
INV
NomGDP
5 a1 1
DRealGDP
21
DRealGDP
21
b
s
ACC
a
1 1 intLD
1 1 E[INF]
b
2s
INTL
(10.2)
where NomGDP and RealGDP are the nominal and real values of the
gross domestic product, respectively, valued at market prices.
Financial markets, asset allocation and risk premia
Households fnancial wealth is typically allocated to fve diferent cate-
gories of assets: domestic money, H
D
, domestic currency denominated
bank deposits held at home, DD
H
, foreign currency denominated deposits
held domestically,
8
FDDom
H
, holdings of government bonds, GDI
H
and
port folio investments abroad, PFI
H
.
9
The household demand function for currency is positively related to
consumption and negatively related to expected infation and interest on
domestic currency denominated deposits, intD. It also depends negatively
on the interest on foreign currency denominated deposits, intDF, adjusted
for the expected rate of depreciation (1 1 e
exp
):
H
D
5
H
0
PRIVCON
q
H
CON (1 1 E[Inf ])
2q
H
Inf (1 1 intD)
2q
H
DD
[ (1 1 De
exp
) (1 1 intDF ) ]
q
H
DF
(10.3)
Household allocation on domestic versus foreign currency deposits is a
function of the interest rate on domestic currency denominated deposits as
a ratio to the rate of return on foreign currency denominated deposits held
at home. Total portfolio investments of households abroad is taken to be
a fxed fraction of total household fnancial wealth, and the demand for
government bonds by households is regarded as a function of the expected
bond interest rate, E[intB].
A crucial decision of the enterprise sector is how to allocate their profts
between funds to private investment and funds to government bonds. This
decision depends on the average proft rate expected from production
activities and, on the other hand, expected returns on government debt
instruments:

PK
#
P
INV
DGDI
E
5 m
E
GDI
c
(1 1 E[intB])
(1 1 avgRPR
21
)
d
2s
E
GDI
(10.4)
Banks set both deposit and lending interest rates. The deposit rate on
domestic currency denominated deposits, intD, is set equal to the borrow-
ing rate from the central bank, intR. The deposit rate on foreign currency
deposits at home, on the other hand, is set on the basis of the (premium
inclusive) marginal cost of borrowing on world capital markets:
(1 1 intDF ) 5 (1 1 intFW ) (10.5)
Following Agnor et al. (2006), the risk-premium inclusive foreign
interest rate is formulated as a function of the (risk-free) world interest
rate, intFW
RF
, and an external risk premium:
(1 1 intFW ) 5 (1 1 intFW
RF
) (1 1 riskpr) (10.6)
in which the risk premium is assumed to be a function of total foreign debt
to exports ratio:
riskpr 5 contag 1
k
2
a
a
ForDebt
a
E
i
b
2
(10.7)
In Equation (10.7) contag is an exogenous coef cient used to capture
the characteristic changes in the sentiments in world capital markets.
The bank lending rate, intLD, in the last analysis is defned as a weighted
average of the cost of borrowing from the central bank and the cost of
borrowing from foreign capital markets. It also takes into account the
(implicit) cost of holding required reserves.
Public sector, credibility and expectations
Since the government debt instruments constitute a relatively signifcant
share of the assets in the domestic fnancial markets in Turkey, modeling
the interactions between the public sector and the central bank (the so-
called fscal dominance) is one of the crucial concerns of this study. With
a mandated target of the primary surplus-GNP ratio, the governments
fscal policy is basically centered around the primary balance. A fscal
defcit is still realized if interest costs on the outstanding public debt exceeds
the primary surplus. The public sector borrowing requirement, PSBR, is
fnanced by either an increase in foreign loans or by issuing bonds. Of
crucial importance is the realization of the interest rate on government
bonds. The expected rate of return on this instrument is defned as:
E[intB] 5 (1 2 PR
default
) intB (10.8)
where PR
default
denotes the subjective probability of default on the current
stock of public debt as perceived by the markets. This variable is set to
depend on, among various alternative measures, the current debt stock to
tax revenues ratio with a one-period lag:
PRdefault 5 1 2 e
2g
0

(DomDebt
G
1 ForDebt
G
)
GTaxRev
(10.9)
The probability of default, PR
default
, has also a further efect on infation
expectations in such a way that the less the probability of default that
is perceived, the higher the chances for the declared infation target to
materialize. Following Agnor et al. (2006) the expected infation rate is
formulated as a function of the governments credibility infcator, that is,
the inverse of the probability of default, PR
default
, and the targeted rate of
infation in the previous period:
E[Inf ] 5 (1 2 PR
default
) Inf
trgt
1 PR
default
Inf
21
(10.10)
Note that, under such a setting, the demand for government bonds is
afected by the probability of default. Private investors assign a non-zero
probability of default in the current period. The expected rate of return
will refect the probability and will demand compensation in the form of
higher nominal interest rates on government bonds. On the other hand,
the larger the stock of debt, the higher the probability of default, and the
higher the interest rate.
For a given probability of default, a continued increase in the supply of
bonds will require an increase in interest rates to evoke investors demand.
Next an increase in the stock of debt will lead to a rise in the probability
of default, which will also rise the prevailing interest rate on government
bonds. Such a mechanism in the model tries to capture the structure of
government trying to provide a signal of confdence to the markets under
the current measures of the program.
10.3.2 General Equilibrium Analysis of Alternative Policy Environments
Now we utilize our CGE apparatus to provide a general equilibrium
analysis of the macroeconomic policy alternatives under twin-targeting. In
what follows we will focus on three sets of issues to depict three alternative
policy environments: frst, we highlight the important role of the expanded
foreign capital infows in resolving (temporarily) the macroeconomic
impasse between the disinfation motives of the CBRT and imperatives
of debt sustainability and fscal credibility of the Ministry of Finance.
Second, we implement a fscally benign monetary policy of reducing the
interest rate charged by the CBRT. Third, we complement the interest rate
reduction policy with a labor market reform and study the implications of
reducing/eliminating payroll taxes (paid by the employers). In all of the
policy simulations we exclusively focus on both the fscal and fnancial
adjustments, and study the possible dilemmas of gains in ef ciency in the
labor markets versus the loss of fscal revenues to the state. Our simulation
experiments are implemented as one-shot, comparative-static exercises.
The results are tabulated in Table 10.2. As valid for all types of modeling
exercises of the current genre, the simulation results should not be taken as
Table 10.2 Experiment results
Base year
(2003) data
EXP-1:
Efects of
increased
foreign
capital
infows
EXP-2A:
Reduce
central
bank
interest
rate
EXP3:
Reduce
central
bank interest
rate and
reduce
payroll taxes
Macroeconomic
Aggregates
Real GDP (Bill 2003 TL) 369.700 369.765 370.283 375.631
Real Private
Consumption (Bill
2003 TL) 255.022 263.259 250.255 256.549
Real Private Investment
(Bill 2003 TL) 66.212 75.138 72.931 74.535
Merchandise Imports
(Bill US$) 69.378 76.307 70.240 70.910
Merchandise Exports
(Bill US$) 47.215 43.475 47.866 48.354
Current Account
Balance (Bill US$) 9.201 22.491 9.234 9.511
Unemployment Rate (%) 10.55 10.35 10.69 7.63
Average Proft Rate (%) 16.15 16.20 16.20 16.80
As Ratios to the GDP
Private Consumption 68.98 71.10 67.60 68.80
Private Investment 17.91 20.30 19.70 20.00
Imports 28.06 29.70 28.39 28.35
Exports 19.09 16.92 19.34 19.33
Current Account
Balance 3.72 8.75 3.73 3.80
Financial Rates and Prices
Infation Rate (CPI) 25.30 18.72 27.72 28.23
Expected Infation Rate 17.65 18.83 16.58 16.79
Expected Depreciation
Rate 41.66 41.78 41.56 41.58
Realized Depreciation
Rate 1.01 9.52 0.98 0.97
Central Bank Interest
Rate (intR) 40.27 40.27 20.00 20.00
Interest Rate on
Domestic Deposits
(intD) 40.27 40.27 20.00 20.00
Interest Rate on Private
Domestic Debt
(intLD) 46.50 47.16 37.65 37.65
Interest Rate on
Government Bonds
(intB) 36.56 60.37 5.49 5.92
Expected Interest
Rate on Gov. Bonds
(E[intB]) 18.28 25.63 3.13 3.29
Risk Premium Incl.
Foreign Int. Rate
(intFW) 3.41 4.09 3.24 3.23
Fragility Indicators
Ratio of Gov. Dom.
Debt to Tax
Revenues 156.98 196.26 122.34 127.76
Governments Fiscal
Credibility Index 0.50 0.42 0.57 0.56
Perceived Probability
of Default on Gov.
Debt 0.50 0.58 0.43 0.44
Ratio of Foreign
Debt to Central
Bank Foreign Reserves 235.87 264.88 223.08 223.77
Ratio of Foreign
Debt to GDP 48.93 52.89 46.25 45.92
Risk Premium on
Private Foreign
Borrowing 1.39 2.10 1.20 1.20
Currency Substitution
(FX deposits/Tot.
Deposits) 90.86 90.99 93.85 93.85
Monetary Aggregates
(as ratio to the GDP)
Money Demand by HH 2.64 2.75 2.68 2.68
Domestic Deposits of
HH 21.27 20.76 20.75 20.60
FX Deposits of HH 19.32 18.89 19.47 19.33
Central Bank Foreign
Reserves 0.21 0.20 0.21 0.21
a forecast of the future, but rather ought to be regarded as quantitative
insights on the relevant macroeconomic outcomes of alternative policy
environments.
EXP-1: Macroeconomics of foreign capital infows
The post-2001 Turkish economy has benefted from the recent surge of
fnancial fows quite extensively. The increased buoyancy in the global
fnancial markets led both to a fall in the rates of interest in the global
markets and also served for provision of expanded liquidity, propelling
consumption and investment expenditures. Mostly driven by the private
portfolio fows, the net annual infow of fnance capital into the new emerg-
ing market economies totaled $456 billion in 2005, before receding to $406
billion in 2006.
10
These magnitudes exceeded the previous peaks hit in the
global fnancial markets before the eruption of the 1997 Asian crisis.
As outlined in Section 10.2, Turkey too had been one of the major
benefciaries of this fnancial glut. Balance of payments data indicate
that the fnance account has depicted a net surplus of $103.3 billion over
the AKP (Justice and Development Party) period, 2003 through 2006.
Base year
(2003) data
EXP-1:
Efects of
increased
foreign
capital
infows
EXP-2A:
Reduce
central
bank
interest
rate
EXP3:
Reduce
central
bank interest
rate and
reduce
payroll taxes
Fiscal Results (as ratios
to the GDP)
Government Aggregate
Revenues 39.13 39.44 38.76 37.03
Government Tax
Revenues 33.35 33.79 33.15 31.48
Government
Consumption Exp. 11.95 12.05 11.84 11.31
Government
Investment Exp. 4.36 4.53 4.16 3.20
Government
Interest Exp. 16.60 42.40 3.99 4.13
PSBR 14.28 28.01 1.00 1.09
Primary Balance 6.50 6.50 6.50 6.50
About half of this sum ($151.2 billion) was due to credit fnancing of the
banking sector and the non-bank enterprises, while a third ($32.8 billion)
originated from non-residents portfolio investments in Turkey. It is also
observed that 64 percent of the total infows (net fnancial fows plus errors
and omissions) was used for fnancing the current account defcit which
had totaled $71.8 billion over the same period; while 36 percent had been
used for reserve accumulation of the CBRT.
In this frst policy experiment we frst study the macroeconomic adjust-
ment mechanisms against this continued infow of fnance capital into the
Turkish economy. To this end, we exogenously increase the total infow
of portfolio investments from abroad, PFI
ROW
, by a factor of $30 billion
(roughly the realized net cumulative fow over 200306). No change in the
CBRTs current monetary policy stance is envisaged with respect to the
level of interest rates and/or exchange rate administration.
11
The exchange
rate was left to full foat to be determined by the free play of foreign
exchange market transactors. Our results are tabulated under the column
EXP-1 in Table 10.2.
The immediate efects of the increased infow of foreign capital are felt
in the currency markets. The exchange rate appreciates by 9.5 percent and
cost savings on the import side leads to a fall in the infation rate to 18.7
percent, from 25.3 percent. Appreciation of the exchange rate leads to a
rise in imports and the current account defcit widens to increase by about
four-fold to reach $22.5 billion in 2003 prices. As a ratio to the GDP, it
increases to 8.7 percent from its base value of 3.7 percent. The domestic
counterpart of the widening current account defcits is the expansion of
private investment (by 2.4 percentage points as a ratio to GDP) and of
private consumption (by 2.1 percentage points as a ratio to GDP). The
monetary base expands by 20 percent and serves for the liquidity require-
ments of this expansion.
The aforementioned expansion of the economy is limited, however, only
to the private sector. Given the fscal constraint on the primary surplus
target, the governments room for maneuver is limited on the expenditure
side. This constraint becomes even more binding as the domestic economy
continues to operate with a signifcantly high real interest burden. It has
to be remembered that a critical feature of the simulated policy environ-
ment is that the central bank continues to maintain its interest rate at
the already high level. As the economy disinfates, however, the real cost
of credit increases even further. The interest cost on the governments
debt instruments, in particular, expands to 60 percent from 36.3 percent.
The governments interest expenditures as a ratio to the GDP rises to 25
percent and that of the public sector borrowing requirement increases
to 28 percent. Increased interest expenditures lead to a widening of the
fscal defcit. Consequently there is a worsening of the fscal credibility of
the government. The credibility index falls to 0.43 from its value of 0.50.
The loss in fscal credibility leads to a rise in the subjective probability of
default as perceived by the private markets.
Thus, the main result of the scenario unveils an important dilemma for
the post-2001 Turkish economy: a policy of maintaining high real rates of
interest along with heavy reliance on foreign fnance proves to be disinfa-
tionary, and it also has expansionary efects on the private sector. The net
result is that the CBRT achieves relative success in controlling infationary
pressures. In the meantime, however, the increased debt burden strains the
already fragile fscal balances and results in further loss of fscal credibility.
The predicament of controlling price infation via high real interest rates
and enhanced foreign capital infows, on the one hand, and the imperatives
of debt turnover and fscal credibility, on the other, remains unresolved.
This impasse is further accentuated with the rise of foreign indebtedness
and consequent external fragility. As the results of EXP-1 suggest, stock of
external debt increases both as a ratio to the GDP (from 48.9 percent to 52.9
percent) and to the foreign reserves of the central bank (from 235 percent to
264 percent). As a result of these adverse developments on external fragil-
ity, the risk premium on the Turkish liabilities in the world markets increase
by 6 percentage points. Clearly, the realized quandary is not to be resolved
by reliance on foreign capital and tight macro management alone, and
postponing the necessary adjustments on the domestic front simply lead to
culminated pressures on the fscal side as well as on the external balances.
EXP-2: Reduce the central bank interest rate
Given the rather high costs of disinfation in terms of high fscal and external
fragility in the previous experiment, the natural policy question is to study
the efects of a reduction in the interest rate charged by the central bank.
In general, the burden of the interest rates has a signifcant contractionary
efect on the Turkish fnancial sector. The cost of central bank liquidity is
held responsible by many scholars for the external debt cycle and intensi-
fed infows of speculative short-term fnance into the Turkish economy.
There is a general call for a reduction of the central banks rate of interest to
escape the trap of speculative infows of fnance leading to appreciation and
more infows, with the consequent widening of the current account defcit
and the rise of external indebtedness. Thus, in this experiment we reduce
the central bank interest rate by half. Again, as above, no further change
is envisaged, when experimenting with this simulation, in the policy instru-
ments or in the parameterization of the exogenously set variables.
Note that given the algebraic characterization of the loanable funds
market, the central bank interest rate has a direct efect on the determination
of the deposit interest rate of the domestic banks. Thus, intD is reduced by
the same magnitude (50 percent) immediately. With declining rates of
interest on deposits the banks fnd it possible to lower their credit interest
rate charged to the enterprises (intLD falls by 8 percentage points). Private
investment expenditures increase by 1.9 percentage points as a ratio to the
GDP.
The distinguishing adjustment mechanism at work is the expenditure
switching of private consumption with investment. As private consump-
tion expenditures are reduced by 2 percentage points as a ratio to the
GDP, domestic savings could be generated to sustain the expansion in
investments; thus, the net efect on the external balances remains modest.
In other words, with a given level of domestic disposable income, an
expanded level of investment demand could have been sustained. Thus,
the current account balance is afected only marginally and the (nominal)
adjustments in the exchange rate are revealed to be modest as well, with a
realized depreciation of less than 1 percent.
What brings forth this adjustment in the private expenditure patterns is
the infation tax. Price infation accelerates by 2.7 percentage points, and
causes a downward shift of aggregate private consumption. The accelera-
tion of infation further strengthens the decline of the real interest cost for
all agents of the economy, private and public. In fact, the relative buoyancy
of the economy, along with the decline of the public interest expenditures,
leads to an improvement in the fscal balances. Of particular interest is the
decline in the public sector borrowing requirement to less than 1 percent of
the GDP, and the increase of the credibility index by 7 percentage points.
It is not clear, however, whether the central bank would be willing to
tolerate the resultant increase of the infation rate, which turns out to be
the crucial adjusting variable to bring forth the warranted adjustments in
the real economy. Yet a further issue is that even though the fscal results
of the policy are observed to be benign, the employment gains remain
quite meager. Unemployment rate persists at above the 10.5 percent level,
and it is this problem we aim to tackle in the next experiment.
EXP-3: Complement central bank interest rate reduction with labor tax
reform
In this experiment we continue on the policy environment of the previous
experiment and complement the central banks interest reduction strategy
with a labor tax reform. Keeping the central banks interest rate at its
reduced level (at half of the base run value), we now implement a further
reduction on taxes paid by employers of labor.
Turkey has one of the highest tax burdens on the labor markets.
Employer-paid social security contributions averaged about 36 percent
of total labor costs during 19962000; it has been argued that these high
social security taxes create strong disincentives to job creation. More gen-
erally, many observers have called for a thorough overhaul of Turkeys
social insurance system. Ercan and Tansel (2006) too state that both the
red tape and non-wage labor costs are higher in Turkey relative to, for
instance, OECD averages. The authors consider the high tax burden on
employment and high social security contributions among the institutional
factors that contribute to the high level of unemployment and high level
of undeclared work. Tunali (2003) indicates that employee contribution to
the social security system can be as high as 15 percent while an employer in
a typical risk occupation contributes as much as 22.5 percent.
Thus in this experiment we study the implications of lowering the payroll
tax paid by the employers on employment, production and fscal balances.
Maintaining the central bank rate of interest at half of its base run value,
we reduce the payroll tax by half, from its base rate of 19 percent. The
lower tax revenues are not compensated by any other taxes. The results of
the experiment are depicted under column EXP-3 in Table 10.2.
Clearly, the most important variable of this experiment is its efects on
unemployment rate and the fscal balances. Unemployment rate falls by
around 3 percentage points, and the real GDP expands by 1.7 percent
upon impact. We fnd, however, that the main adjustment falls on public
investments and then on the price infation. The frst outcome is the direct
result of the fscal administration under the current austerity program.
The logic of the fscal balances is that, given the tax revenues and interest
costs, the public sector is to maintain a primary surplus (of 6.5 percent)
as a ratio to the GDP. Once this constraint is met the rest of the public
expenditures are calculated. Thus, within the context of our experiment,
as tax revenues are curtailed, the government fnds it necessary to adjust
public investments downwards. As a percentage of GDP, public invest-
ments are observed to fall to 3.2 percent from its base value of 4.4 percent
(a signifcantly low rate itself).
The efect on fscal accounts is also emphasized in the ratio of govern-
ment tax revenues against public debt stock. The aggregate tax revenues
fall by almost 2 percentage points; yet, given the cost savings on the
interest expenditures, the overall solvency of the public sector remains
improved. Thus, the lower interest rate policy is enacted here as an import-
ant component of the labor tax reform policy. With a reduced interest
burden over the public sector, the CBRT facilitates the fscal authority
to alleviate pressures on the fscal balances that would have emerged as a
result of reduced labor tax revenues. With accelerated growth in GDP and
lower interest costs, the fscal balances improve, with consequent gains in
fscal credibility.
12
At the outset, the trade-ofs as suggested by the simulation exercise
under EXP-3 seem modest and not severely binding: at a loss of 2.9
percentage point increase of the infation rate (from 25.3 percent to 28.2
percent), the gains in fscal credibility and employment are found to be
relatively robust. Given that under the macroeconomic adjustments of the
experiment the external balances were not strained any further, equilib-
rium in the foreign exchange market seems to be maintained, as well.
10.4 CONCLUDING COMMENTS AND POLICY
DISCUSSION
In this chapter, we reported on the current state of the macroeconomic
policy environment in the Turkish economy throughout the 2000s and
studied the general equilibrium efects of two widely discussed policy
changes in the current context: reduce payroll taxes and reduce the central
bank interest rate. The current IMF-led austerity program operates with
a dual targeting regime: a primary surplus target in fscal balances (at 6.5
percent to the GDP); and an infation targeting central bank whose sole
mandate is to maintain price stability. Accordingly both policy questions
are analysed within the constraints of the aforementioned dual targets set
as outer conditionalities of Turkish macroeconomic decision making.
Our policy experiments reveal that the current monetary strategy fol-
lowed by the CBRT that involves heavy reliance on foreign capital infows
along with a relatively high real rate of interest, is efective in bringing
infation down, yet it sufers from increased cost of interest burden to the
public sector and strains fscal credibility. It also leads to excessive foreign
indebtedness with increased external fragility. In the medium to long run
the increased fscal and external fragilities along with a persistent and high
unemployment manifest a severe impasse, whose resolution will likely lead
to onerous adjustments in the labor markets and the real sector. Against
this background, we utilized the CGE model to search for applicable alter-
native policy regimes starting from the immediate short run. Our results
indicate that a heterodox policy of (1) reducing the central banks interest
rates along with (2) lowering the (payroll) tax burden in the labor markets
ofers a viable environment in the short run, with accelerated growth and
improved employment outcomes. The frst arm of the policy, viz. reduc-
tion of the CBRT interest rate, is important to facilitate the improvement
in fscal balance (and fscal credibility) at a time when tax monies from
labor taxes are expected to be reduced. The second critical element is low-
ering of the tax burden on employers. With lower payroll taxes levied on
employment in production, the employers are led to increase employment
demand (and also most probably be more willing to employ formal
labor, and reduce the unrecorded activities along with informalization of
the labor markets; issues that our model is not well-equipped to address).
With increased credibility of the public sector and lower rates of
unemployment, the returns to the heterodox policy reform agenda are
quite benign. However, all these come at visible opportunity costs; in
particular on the infation side. As lower interest rates boost domestic
investment expenditures and the domestic economic activity is revived
due to expanded employment, infationary pressures accumulate in the
commodity and fnancial markets. It is not so clear at the outset how toler-
ant would the CBRT be to the acceleration in infation. Even though our
results are quite modest on the pace of both realized and expected rates of
infation, it is clearly an important constraint that merits close observation
in the Turkish macroeconomic environment. It is, in fact, mainly for this
reason that we maintain some of the key features of the current austerity
program with respect to expectations management in the short run. Of
particular importance among these is the signaling efect of the primary
surplus target. The policy environment of EXP-3 sets the fscal balances
with the programmed target of 6.5 percent primary surplus ratio to the
GNP, rather than proposing a drastic break away from it. In fact, with
a proper emphasis on dynamics, a direct case can clearly be proposed to
stimulate domestic investment expenditures with a policy of lower interest
rates, and advocating a fscal policy of high public investments towards
enhancing human capital formation and social infrastructure. Rather than
cutting public investments on health, education and social infrastructure,
a case can be forwarded to disregard the rising public sector borrowing
requirement to GDP ratio in the short run, and implement a fscal policy
to maintain a level of household income capable of addressing the tasks of
accumulating human capital. Yet, given the short-run framework of our
current modeling framework, we choose to abstain from making ad hoc
statements regarding the dynamic consequences of such a policy environ-
ment, an issue that had been dealt with elsewhere more efectively (see,
for example, Gibson, 2005; ISSA, 2006; Voyvoda, 2003; Voyvoda and
Yeldan, 2005).
Above all, our simulation experiments clearly underscore the importance
of maintaining an integrated and coherent policy framework between
the monetary and fscal authorities. Given the acuteness of the perceived
dilemmas on disinfation and fscal credibility, the resolution of the current
impasse will surely necessitate a more tolerant view over the programmed
targets (on both infation and the primary surplus ratio) as well as a coher-
ent and a mutually supportive macro policy design. Furthermore, there
is a clear case for a acute need to design viable policies to diminish the
exposure of the domestic economy (in particular of the fnancial markets)
to short-term, speculative foreign capital. This, in turn, may necessitate
implementation of capital management techniques to gear infows towards
longer maturities.
NOTES
1. Author names are in alphabetical order and do not necessarily refect authorship
seniority. We are indebted to Korkut Boratav, Yilmaz Akyz, Jerry Epstein, Bill
Gibson and to the members of the Independent Social Scientists Alliance for their valu-
able comments and suggestions on previous versions of the chapter. Previous versions
of the chapter were presented at the Istanbul Conference of the EcoMod (June 2005);
the 9th Congress of the Turkish Social Sciences Association (December 2005, Ankara);
the Ankara congress of the Turkish Economics Association (September 2006); and
seminars at Bilkent, METU, Bogazici, Utah, Massachusetts-Amherst, Connecticut and
the central bank of Turkey. Research for this chapter was completed when Yeldan was
a visiting Fulbright scholar at the University of Massachusetts-Amherst for which he
acknowledges the generous support of the J. William Fulbright Foreign Scholarship
Board and the hospitality of the Political Economy Research Institute at the University
of Massachusetts-Amherst. Needless to mention, the views expressed in the chapter are
solely those of the authors and do not implicate in any way the institutions mentioned
above.
2. That is, balance on non-interest expenditures and aggregate public revenues. The
primary surplus target of the central administration budget was set 5 percent to the
gross national product.
3. Further institutional details of the central banks infation targeting framework can
be found in the December 2005 document, General framework of infation targeting
regime and monetary and exchange rate policy for 2006, accessed December 2006 at:
www.tcmb.gov.tr/yeni/announce/2005/ANO2005-45.pdf.
4. http://www.maliye.gov.tr.
5. Measured in 2002 producer prices. If the PPP-correction is calculated in 2000 prices, the
revised debt to GNP ratio reaches to 82.3 percent.
6. See, for example, UNCTAD, Trade and Development Report (2002 and 2003), New
York and Geneva.
7. http://www.peri.umass.edu/Alternatives-to.382.0.html.
8. By allowing households to hold foreign currency denominated deposits in the domestic
banking system, we try to represent the high level of dollarized liabilities in the Turkish
fnancial system (see Table 10.1).
9. Both residents portfolio investments abroad, PFI
H
, and non-residents portfolio
investments at home, PFI
ROW
, are incorporated in the model in order to capture any
real-economy efects of these speculative means, which we believe are important in
understanding the growth pattern of the Turkish economy in the last decade.
10. See, for example, Institute for International Economics, http://www.iie.com.
11. It has to be noted, as a reminder, that the current rate of interest set by the CBRT is
already signifcantly high in real terms. Maintaining high real rates of interest was but
one of the discretionary measures of the CBRT in an attempt to reduce infation by
curtailing domestic demand expansion, as well as to sustain the infow of foreign capital
to cover the widening current account defcit.
12. Note, however, that this comparison is valid against the base run. One witnesses a slight
loss in fscal credibility relative to EXP-2.
REFERENCES
Agnor, P.R., H.T. Jensen, M. Verghis and E. Yeldan (2006), Disinfation,
fscal sustainability, and labor market adjustment in Turkey, in Richard
Agnor, A. Izquierdo and H.T. Jensen (eds), Adjustment Policies, Poverty
and Unemployment: The IMMPA Framework, Oxford: Blackwell Publishing,
Chapter 7, pp. 383456.
Ercan, H. and A. Tansel (2006), How to approach the challenge of reconciling
labor fexibility with job security and social cohesion in Turkey, in Reconciling
Labour Flexibility with Social Cohesion: Facing the Challenge, Strasbourg:
Council of Europe Publishing.
Gibson, B. (2005), The transition to a globalized economy: poverty, human capital
and the informal sector in a structuralist CGE model, Journal of Development
Economics, 78, 6094.
ISSA (Independent Social Scientists Alliance) (2006), Turkey and the IMF:
Macroeconomic Policy, Patterns of Growth and Persistent Fragilities, Penang,
Malaysia: Third World Development Network.
Tunali, . (2003), Background study on the labour market and employment in
Turkey, paper prepared for the European Training Foundation, Turin, Italy,
June.
Voyvoda E. (2003), Alternatives in debt management: investigation of Turkish
debt in an overlapping generations general equilibrium framework, unpub-
lished PhD thesis, Bilkent University.
Voyvoda E. and E. Yeldan (2005), IMF programs, fscal policy and growth:
investigation of macroeconomic alternatives in an OLG model of growth for
Turkey, Comparative Economic Studies, 47, 4179.
World Bank (2000), Turkey country economic memorandum structural
reforms for sustainable growth, vols. I and II, report no. 20657TU, September,
Washington, DC.
227
11. Employment targeting central
bank policy: a policy proposal for
South Africa
Gerald Epstein
1
11.1 INTRODUCTION
The story of South Africas struggle with apartheid is a dramatic and
heroic one, a story that has culminated in a hugely successful politi-
cal transition from a minority dominated, authoritarian and repressive
government organized along racial lines before 1994, to an inclusive,
democratic government, in which the majority rules, subject to important
protections for minority rights.
The story of South Africas economic transition, on the other hand, is
not such a rosy one. Income and wealth are still highly unequally distrib-
uted, largely along ethnic lines with the white population still command-
ing the bulk of the national income, and still controlling the bulk of the
national wealth. Economic growth has been moderate during most of the
period since the creation of the new South Africa in 1994, running at an
average of 3.4 percent and unemployment rates are hovering somewhere
between 25 percent and 36 percent depending on exactly how one counts.
Among the black population, in 2006 unemployment was over 30 percent,
while among the white population it was less than 5 percent.
Such problematic economic trends are in no small part due to the highly
unequal economy inherited from the past. But some of the dif culties with
the economic transition have been due to the ideologies and the policies,
including macroeconomic policies, chosen by the new government in the
last decade or so. The government has adopted many pages from the
neoliberal play book, including fnancial liberalization, capital account
liberalization, austere fscal policy and, most relevant to this chapter,
formal infation targeting. These policies have achieved certain gains, but
have done little to reduce unemployment and generate more economic
equality.
Many in South Africa are looking for some alternatives to these policies.
While the government is still strongly committed to infation targeting,
recognition of its limitations are becoming more widespread.
In this chapter I present employment targeting (ET) as an alternative
to infation targeting monetary policy framework for South Africa. It is
a framework for monetary policy which attempts to incorporate some
of the advantages normally claimed for a targeting framework namely
enhancing transparency and accountability while focusing the goals of
monetary policy more directly on critical macroeconomic problems facing
the South African economy, namely, employment. Of course, no govern-
ment, including that of South Africa, can ignore the infationary impacts
of policy, and any ET approach will have to include a goal of stabilizing
infation at sustainable levels.
2
As will be seen as the argument develops, reorienting monetary policy
toward generating employment will require that monetary policy rede-
velop and utilize a multiplicity of monetary policy and credit tools,
including tools for credit allocation and capital management. While these
policies had been in the toolkit of many central banks for decades, they
have recently fallen out of favor, or have been made impotent by fnancial
liberalization or legal changes (Epstein, 2007). So one will fnd a bit of
going back to the future in some of the proposals found in this chapter.
But, as there is no simple return to the past, the ET monetary policy frame-
work for South Africa takes into account important changes at both the
national and international levels, especially in the operations of fnancial
markets, which must be accommodated in any policy framework.
11.2 INFLATION TARGETING IN SOUTH AFRICA
IN MACROECONOMIC CONTEXT
At the time of the 1994 victory in the struggle against apartheid a major
debate ensued over the future direction of economic policy, which culmi-
nated in the adoption of the Department of Finances macroeconomic
strategy, the Growth, Employment and Redistribution (GEAR) policy.
GEAR was similar to the neoliberal macroeconomic frameworks devel-
oped by the International Monetary Fund (IMF), and it is based on
similar premises: the key to achieving employment growth and a rapid
increase in living standards is to win the confdence of both domestic and
foreign investors, by engaging in fscal austerity and fnancial liberaliz-
ation, among other policies.
With respect to monetary policy, more specifcally, soon after being
elected in 1994, the South African government decided that price stabil-
ity should be the central concern of monetary policy. In 1998 the Reserve
A policy proposal for South Africa 229
Bank adopted, for the frst time, an informal infation target range of 15
percent. Then in February 2002 the Reserve Bank adopted a formal infa-
tion targeting approach. Under infation targeting the Ministry of Finance
establishes the target range and the reserve bank chooses the instruments
to achieve the target. It chose a target of 36 percent rate of increase of
CPI, which is a consumer price index excluding interest costs on mortgage
rates and the Reserve Bank chose the repo rate as the main monetary
policy instrument.
Looking back over the initial years, however, one can see that the
GEAR policies were implemented with mixed results (Table 11.1).
The GEAR policy delivered on its promises in the early years with
respect to fscal defcits, infation and the current account, but failed in
terms of economic growth and employment generation.
Real GDP growth has been higher since that time but still not large
enough to substantially reduce the rate of unemployment. Figure 11.1
shows the rate of real GDP growth before and after the transition to
demo cratic rule. While GDP growth has gone up since the confict-ridden
and sanctions-laden period between 1984 and 1994, the 3.45 percent per
year average growth rate is still relatively modest, though in the most
recent years, largely due to the global commodity boom, growth rates have
been above 4.5 percent.
Given this relatively low per-capita growth in real GDP, it should not be
surprising that employment performance has been fairly weak.
Table 11.2 shows the of cial unemployment rate broken down by popu-
lation group in 2001 and 2007. The contrast among the population groups
is striking with African unemployment, by the of cial measure showing
26.8 percent unemployment in 2007, in contrast to the overall white rate
of 5.8 percent. Both rates have come down somewhat since 2001 due again
to the higher than average rate of economic growth during the latter years
Table 11.1 Comparison of GEAR projections and actual outcomes
(19962000 averages)
GEAR projections Actual outcomes
GDP Growth (%) 4.2 2.8
Infation (CPI) (%) 8.2 6.7
Current Account Defcit (% GDP) 2.4 1.1
Employment Growth (%) 2.9 0.7
Non-Gold Exports 8.4 7.1
Fiscal Defcit (% GDP) 3.7 3.2
Source: Du Plessis and Smit (2005).
of that period. This shows that in all likelihood higher economic growth
does, in fact, lead to lower unemployment rates among all population
groups.
Table 11.3 shows the of cial unemployment rate in September 2001 and
September 2007, broken down by gender and population group. These
Table 11.2 Of cial unemployment, September 2001 and September 2007
(percentage of labor force)
September 2001 September 2007
African 35.7 26.8
Coloured 21.2 20.6
Indian 18.8 8.2
White 5.8 3.8
Average 29.4 22.7
Source: Statistics South Africa.
3
2
1
0
1
2
P
e
r
c
e
n
t
3
4
5
6
84 86 88 90 92 94 96 98 00 02 04 06
After democratic transition
Average growth: 1.03 Average growth: 3.45
Source: Reserve Bank of South Africa.
Figure 11.1 Real GDP growth
show that unemployment rates among women are higher than among
men for all population groups, but they too have gone down somewhat in
response to higher economic growth rates.
These data refect the of cial unemployment statistics. Table 11.4
presents data that include discouraged workers and therefore may be a
more accurate measure of unemployment. They indicate that the unof-
fcial unemployment rate might be at 34 percent overall and over 35
percent for women in 2007.
As mentioned earlier, since 2002, if not before, the Reserve Bank of
South Africa has followed an infation targeting regime for monetary
policy; the policy has been directed, most recently toward keep CPI
Table 11.3 Of cial unemployment by gender and ethnic group, September
2001 and September 2007 (percentage of labor force)
September 2001 September 2007
Male
Black African 31.5 23.3
Coloured 19.5 20.0
Indian/Asian 15.7 7.4
White 4.7 3.5
Average 25.8 19.8
Female
Black African 40.7 30.9
Coloured 23.1 21.3
Indian/Asian 23.5 10.2
White 7.4 4.2
Average 33.8 26.1
Source: Labour Force Survey, March 2006, March 2007 (Pretoria: Statistics South Africa,
September 2006 and 2007).
Table 11.4 Of cial unemployment plus discouraged workers, September
2001 and 2007 (percentage of labor force)
2001 2007
Male 33.8 31.8
Female 47.0 34.9
Total 40.0 34.0
Source: Authors calculations from Labor Force Survey, March 2006 and March 2007
(Pretoria: Statistics South Africa).
infation within a target range of 36 percent. Indeed, infation has been in
that range since mid 2003 or so (Figure 11.2).
Whether it has caused the decline is more dif cult to answer. As Ball
and Sheridan (2003) point out, however, infation fell in many countries
over the last decade, independently of whether the countries were adopt-
ing infation targeting or not. What is clear is that this infation targeting
regime has led for most of the period to very high rates of real interest rates
which, in turn, have contributed to relatively modest economic growth
and, until the recent period of the global commodity boom, to anemic
growth in employment.
11.3 AN EMPLOYMENT TARGETING
FRAMEWORK FOR SOUTH AFRICA
In light of the problems with the infation targeting regime in South Africa,
and especially given South Africas major problems with unemployment
and underemployment, an alternative framework for monetary policy is
2
4
6
8
P
e
r
c
e
n
t
10
12
2001 2002 2003 2004 2005 2006
CPI inflation
(Monthly, on an annual basis)
Inflation
target
range
Source: Reserve Bank of South Africa.
Figure 11.2 CPI infation
badly needed. Here I propose a real targeting framework for monetary
policy, and then describe a specifc type of real targeting framework that
applies to South Africa, namely an ET framework for monetary policy.
11.3.1 A Real Targeting Framework for Monetary Policy
A real targeting framework for monetary policy should adhere to the fol-
lowing principles:
Context-appropriate monetary policy. Central bank policy goals
and operating procedures must be based on the structure and needs
of South Africa: no generic, one-size fts all approach is likely to be
appropriate to every situation.
Real economy-oriented monetary policy. Policy makers should recog-
nize that very high rates of infation can have signifcant costs, but that
short of that, policy must also be oriented toward promoting invest-
ment, raising employment growth and reducing unemployment.
Transparency and accountability. Taking a leaf from the targeting
approach, central banks should be made more accountable to the
public by making their objectives and approaches more transparent.
They should tell the public what their targets for monetary policy
are, describe the economic assumptions underlying their plans to
reach those targets and, if they do not reach them, explain why while
describing their plans for achieving them in the next period. And,
most importantly, the goals of the central bank should be deter-
mined by a democratic process.
Policy fexibility. A fundamental fact is that there is a great deal of
uncertainty concerning the underlying structure of the economy
and about the nature of national and international shocks at any
particular time. Hence, adherence to any target has to be somewhat
fexible.
Suf cient tools to reach the targets. Monetary policy has been
stripped by fnancial liberalization and neoliberal conceptions of
policy tools needed to achieve a multiplicity of targets: credit allo-
cation techniques, which used to be very important components
of central bank policy in many countries, have been systematically
eliminated, capital controls (capital management techniques) like-
wise, which in the past have been part of the arsenal of monetary
policy, have been dramatically liberalized. These techniques need to
be revitalized and modernized so that, where and when appropriate,
they can be used as instruments to help central banks reach their
goals.
Supporting institutions. Central bank policy is no panacea. Other
important supporting institutions are also required, including
strong tax institutions to enable the government to raise the
revenue it needs to fund important public investments, and public
fnancial institutions to channel credit in support of productive
investment.
In addition to the direct advantages of a real targeting framework, there
are several extremely important indirect advantages of the ET framework
which themselves will contribute in a crucial way to the frameworks
success. These include the accumulation of new knowledge about the con-
nections between monetary policy and employment, and the implementa-
tion of new policies to generate more employment and economic growth.
Both of these positive outcomes will be facilitated by focusing the atten-
tion of the central bank, with its enormous human and fnancial resources,
on the key issue of employment growth.
Making employment growth a target of monetary policy might also
induce a profound shift in the attitude and the activities of the central
bank. It might begin to assign its economists to study the relationships
between monetary policy and employment growth; to study what mon-
etary policy instruments are best used to achieve employment growth; and
to organize conferences on employment growth and monetary policy. It
might even give promotions and more resources to members of its staf
that make breakthroughs in the understanding of these connections. In
the case of South Africa, it will lead the Reserve Bank to link up with
others outside the bank who have knowledge and experience with respect
to employment generation and its relationship to fnancial variables.
It could also lead the bank to design new programs to help it reach its
targets.
11.4 EMPLOYMENT TARGETING MONETARY
POLICY IN SOUTH AFRICA
In this section I describe the outlines of an ET policy for the Reserve
Bank of South Africa. This policy has been developed in the context of
an overall ET program for the South African economy developed by
economists at the Political Economy Research Institute.
3
This larger ET
program includes fscal stimulus, public credit allocation and develop-
ment banking, capital management techniques, mechanisms of infation
control, tax reforms, including mechanisms such as an enhanced securities
transactions tax to raise more revenue to fnance employment policies, and
other sectoral policies, for example, anti-trust and competition policy to
correct infrastructure bottlenecks preventing more growth and employ-
ment generation.
In this chapter I focus on the role of the Reserve Bank and monetary
policy within this ET framework.
11.4.1 The Role of Monetary Policy and the Reserve Bank in this ET
Program
The basic structure of the proposed central policy is as follows: the Reserve
Bank would set its interest rates to achieve an overall real growth rate con-
sistent with the plan which has an employment target at its core. As part
of its mandate, the Reserve Bank would try to reach an infation constraint
that is mutually decided upon as part of the overall program. In addition,
the Reserve Bank would manage some of the credit allocation programs
that are part of the targeted components of the overall ET macroeconomic
policy. Finally, the Reserve Bank would manage the capital account as
needed to maintain the exchange rate and exchange rate stability needed
to implement the program.
If the ET policy is to be fully successful, in the long run it will help con-
siderably if the Reserve Bank makes certain institutional commitments.
These might include some of the following. The Reserve Bank will launch
a set of research programs both within the bank and outside to improve
understanding of the relationship between monetary policy tools and
employment growth, both in the formal sector and informal sector. The
Reserve Bank will work with fnancial institutions, including development
banks and major commercial banks, as well as smaller fnancial institu-
tions, to develop instruments and programs to facilitate the allocation
of credit for efective employment generating activities. If the rates of
economic growth achieved through more expansionary monetary policy
do not generate as much employment as projected, the Reserve Bank is
committed to fnding new policies and mechanisms, along with the rest of
the government, to achieve these targets.
11.4.2 Targets and Instruments and All That
In the terminology of targets and instruments used in the monetary policy
literature, in this framework the ultimate target of monetary policy is
employment growth, while the intermediate targets are real GDP growth,
infation and exchange rates. The instruments are short-term interest
rates (the repo rate in the case of current practice in South Africa), capital
management techniques and credit allocation policies.
11.5 SIMULATING THE IMPACT OF MONETARY
POLICY: A VAR-BASED SIMULATION MODEL
In this section I present a simple exercise to show the impact of an expan-
sionary monetary policy as part of an ET framework. Following this
section, I consider the other instruments and intermediate targets to be
utilized by the Reserve Bank in its ET policy.
11.5.1 Monetary Policy Experiments
In this section I present some simulations based on a series of simple struc-
tural vector auto-regression (VAR) models I built of the South African
economy. These simulations are designed to estimate the impacts of
employment targeted monetary policy on real GDP growth a key deter-
minant of employment generation and on infation and exchange rate
stability which are crucial other variables of likely concern to the central
bank.
VARs-based simulation models are widely used in monetary policy
analysis (see, for example, Aron and Muellbauer, 2002; Bernanke et al.,
1997; Gallardo and Ros, 2008). This approach uses a minimum number of
assumptions about the structure of the economy to estimate a small, sim-
plifed model of the macroeconomy. Based on that estimation, we simulate
the model with changes in various policy tools and estimate the impact of
those changes in policy on the economy. I do this by estimating the difer-
ence between the baseline path of economic variables, such as infation
and economic growth, and the path these variables take under the new
(hypothetical) settings of the monetary policy tools. The diference between
these is taken to be the impact of the policy change on the economy.
It is important to point out that, despite their widespread use, these
VAR models have many drawbacks. They are highly simplifed models
of very complex economies; and in situations where there are either short
data series, or a large amount of structural change, these models may
not be able to form good estimates of economic relationships or forecast
with a great deal of accuracy. Still, for our purposes, we hope that these
results can give us some ball-park idea of the impact of the approach I am
proposing.
This analysis addresses the impact of changes in the South African
Reserve Bank monetary policys rule on exchange rates, infation and
economic growth. The goal is to estimate the quantitative impact of dif-
ferent interest rate settings on economic growth, infation and exchange
rate variability, and indirectly, then, on employment. It incorporates some
important exogenous variables as well.
Before describing the estimates and the simulations, I frst discuss our
data variables and sources.
11.5.2 Data and Preliminary Statistical Tests
All data are either originally quarterly data, or have been transformed into
quarterly data from monthly data.
Prime rate: the prime rate is the prime lending rate. It tracks
closely the changes in the repurchase (or repo) rate, which is the
main tool of monetary policy. The reason we use the prime rate
rather than the repo rate is that it has a longer data series.
4
Exchange rate: the nominal rate relative to the US dollar. Our
measure is the four quarter rate of change of the rand, measured so
that an increase means an increase in the rate of depreciation of the
rand.
5
Infation rate: we use the CPI variable, as calculated by Aron and
Mullbauer (2002). This is the consumer price level, excluding mort-
gage interest. We use the four quarter rate of infation.
6
GDP growth: the four quarter rate of growth of real GDP. Source:
South African Reserve Bank.
7
US Tbill: the three-month US Treasury bill rate.
8
Private credit: the four quarter rate of growth of private credit
created by all monetary institutions; monthly data transformed to
quarterly data. Source: South African Reserve Bank.
9
Terms of trade: four quarter change in the terms of trade. Source:
South African Reserve Bank.
10
11.5.3 Stationarity Tests on Variables
All variables were tested for stationarity using the Augmented Dickey-
Fuller tests. The results of these tests are presented in Table 11.5.
For the estimates that follow, we report on results using the growth
variable (of real GDP). We have done the estimates with de-trended GDP
growth as well and the results are roughly the same. We report on the non-
de-trended variable because of greater ease of interpretation.
11.5.4 VAR Estimates of Monetary Policy
I estimate a VAR model with four endogenous variables (prime rate,
exchange rate change, infation and growth) and one exogenous variable,
the US Treasury bill rate (US Tbill), over the period 1989, frst quarter
(1989q1) to 2004, fourth quarter (2004q4), with four quarter lags. I choose
1989q1 as a starting date because of concerns of signifcant structural
breaks at that time. I then estimated the impulse response function using
the Choleski de-composition, with the following ordering (prime, change
in exchange rate, infation, growth). I tried the results with diferent order-
ings and the results did not seem very sensitive to these changes. A positive
shock in the prime reduces economic growth, has a modest, cyclical impact
on infation and is associated with increased variability in the exchange
rate.
Figure 11.3 presents the impulse response functions from this model.
Note that in this VAR estimate, an increase in the prime rate has a
stable impact on infation, but it generates an increase in the infation
rate. This probably refects the so-called cost channel or Patman Efect,
named after US Congressman Wright Patman, who warned that interest
rate increases were infationary.
5.5 Simulation Results
The next step is to transform this VAR model into a simulation model to be
used for monetary policy experiments. The goal of the simulation exercise
is to estimate the impacts of a medium- to long-term decline in the prime
lending rate on exchange rate variability, infation and economic growth.
The basic idea is to estimate the likely impacts of a looser monetary policy
by the Reserve Bank which attempts to expand real GDP growth in order
to generate more employment, as part of an ET policy.
In order to estimate these impacts, I undertook the following steps.
I frst solved the estimated VAR model using a dynamic simulation,
over the period 1994q12004q4 and call the results of this dynamic
simulation the baseline results. These baseline estimates are then used
in combination with data generated by policy experiments to estimate
Table 11.5 Stationarity tests
Variable Stationary
(signifcance level)
Intercept Intercept
and Trend
Exchange Rate Change Yes (5%) Yes No
Credit Change Yes (5%) Yes No
Growth Yes (1%) No Yes
Infation Yes I (1) (1%) Yes No
Prime Yes I (1) (1%) Yes No
US Tbill Yes I (1) (1%) Yes
the impacts of changing the prime lending rate in the steps described
below.
As seen in Figure 11.4, the baseline estimates, when compared to the
actual data, capture basic trends in the data, for the most part, but they
do not do a very good job of tracking all the turning points in the data.
Unfortunately, this is not an uncommon problem in such models which
have a limited number of degrees of freedom.
The next step is to create alternative scenarios in which the prime
lending rate is adjusted to refect an employment-oriented monetary
policy. Here I report on one scenario: lowering the prime rate so that it
remains 4 percentage points below the baseline and keeping it there for
fve years. Table 11.6 presents the numerical summary of the main efects
of this policy experiment.
Figure 11.5 describes the impacts by showing the diference from the
baseline. The results show that GDP growth goes up on average by about
0
2
4
6
8
10
12
1 2 3 4 5 6 7 8 9 10
Accumulated Response of
Prime to Prime
30
20
10
0
10
20
30
1 2 3 4 5 6 7 8 9 10
Exchange Rate to Prime
4
3
2
1
0
1
2
3
4
5
1 2 3 4 5 6 7 8 9 10
Inflation to Prime
5
4
3
2
1
0
1
1 2 3 4 5 6 7 8 9 10
Growth to Prime
Figure 11.3 Accumulated response to Cholesky one SD innovations 2
SE
6 percent for the period 200104 (about 0.5 on average for the whole fve-
year period), infation increases by about 1 percentage point on average,
while the exchange rate changes become more variable.
Table 11.6 summarizes these results. In this simulation on average
economic growth is raised by 0.6 of 1 percent, with only modest increases
in infation and exchange rate instability. The exchange rate instability
40
30
20
10
0
10
20
30
40
94 95 96 97 98 99 00 01 02 03 04
Exchange rate change
Exchange rate (baseline)
Exchange rate change
0
1
2
3
4
5
6
94 95 96 97 98 99 00 01 02 03 04
Growth
Actual
Growth (baseline)
3
4
5
6
7
8
9
10
11
94 95 96 97 98 99 00 01 02 03 04
Inflation
8
12
16
20
24
28
94 95 96 97 98 99 00 01 02 03 04
Prime rate
Prime rate
Prime rate (baseline)
Actual
Inflation (baseline)
0
1
2
3
4
5
6
7
94 95 96 97 98 99 00 01 02 03 04
US Tbill
Actual
Baseline
Figure 11.4 Baseline estimate versus actuals 1994Q12004q4
created is relatively small given the swings in exchange rates that South
Africa has experienced in the last decade. And the increase in infation
stays well within the single digits. This increase in the rate of economic
growth is itself modest, but accumulated over time, it could have a signif-
cant impact on employment growth.
It should be obvious that, by itself, lowering interest rates will not solve
South Africas unemployment problem. The Reserve Bank, along with
other economic institutions, will need to participate in more aggressive
and creative initiatives to enhance employment opportunities.
11.6 CREDIT ALLOCATION AND CAPITAL
MANAGEMENT POLICIES TO SUPPORT
EMPLOYMENT GENERATION
11.6.1 Credit Allocation Policies
As the empirical exercise in the previous section suggests, manipulation
of interest rates will certainly have some positive impact on employment
in South Africa, but ultimately, other tools will be needed to encourage
employment at suf cient levels. Credit allocation tools can be an important
set of policies to contribute to employment growth in South Africa if they
are properly implemented and monitored. While some of this operation
might occur outside of the purview of the South African Reserve Bank, the
bank can and should play a signifcant role in this area. (See Pollin et al.
(2006) for much more detail on credit allocation tools in South Africa.)
In doing so, the Reserve Bank has many examples to draw on both
from its own history and from the central banks of many highly success-
ful developed and developing countries, including South Korea, Japan,
France and the Taiwan Province of China and the Peoples Republic of
China (Amsden, 2001; Epstein, 2007). Still, experiences are not simply
transferable in a turnkey way, and even the earlier experience of directed
Table 11.6 Impact of 4 percent point decline in prime rate
2000 2001 2002 2003 2004
Exchange Rate Change 6.16 5.74 1.91 0.35 1.29
Growth 0.22 0.79 0.44 0.61 0.55
Infation 0.12 1.29 0.96 0.76 0.85
Prime Rate 4.00 4.00 4.00 4.00 4.00
US Tbill 0.00 0.00 0.00 0.00 0.00
credited under the apartheid system is not something the current regime
would want to replicate. So, the Reserve Bank, while learning from past
experience, will need to develop credit allocation mechanisms to ft the
current situation. These could include concessionary loans and increased
capitalization for development banks, loan guarantees and asset-based
reserve requirements.
4
2
0
2
4
6
8
10
12
2000 2001 2002 2003 2004
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
2000 2001 2002 2003 2004
Exchange rate change Growth
0.4
0.0
0.4
0.8
1.2
1.6
2000 2001 2002 2003 2004
4.2
4.1
4.0
3.9
3.8
3.7
2000 2001 2002 2003 2004
Inflation Prime rate
1.0
0.5
0.0
0.5
1.0
2000 2001 2002 2003 2004
US Tbill
Figure 11.5 Impact of 4 percent point drop in prime relative to baseline,
2000Q12004Q4
To avoid corruption and misuse of these tools, careful monitoring and
sharp incentives need to be put in place.
11.6.2 Capital Management Techniques
Another vulnerability of these programs is that, in a fnancially integrated
economy, fnancial arbitrage and capital fight might undermine them. As
just discussed, an expansionary monetary policy might induce unstable
exchange rates and overshooting as well. Capital controls or capital
management techniques have been widely used to manage exchange
rates and create running room for monetary policy (see, for example,
Epstein et al., 2005).
South Africa has a long history of capital management techniques,
including exchange and capital controls, dating back to at least 1939 (see
Bruce-Brand (2002) for a brief history and overview). The current govern-
ing legislation was set out in 1961. Their application has ebbed and fowed
over the years. With the reintegration of South Africa into the world fol-
lowing the abolishment of the apartheid government in 1994, after consid-
erable discussion, the government decided to progressively liberalize the
capital and exchange controls.
Capital management techniques can help countries achieve more
autonomy in the making of monetary policy, including reducing the
variability and misalignment of the exchange rates, though there is still
a great deal of debate on this point. A large literature has attempted to
assess the efectiveness of capital controls (see, for example, Lee and
Jayadev (2005), Henry (2006) and Epstein et al. (2005), among many
others, for surveys). Studies which have analysed the efectiveness of
capital controls have looked primarily at two things: frst, at their ability
to maintain a wedge between domestic interest rates and foreign rates;
or, second, at their ability to avoid currency crises, that is, either large-
scale changes in exchange rates or the forcing of countries of of a peg
through reserve loss; more recent studies have looked at the ability of
controls (especially on infows) to prevent overvalued currencies and
currency instability and the ability of controls to provide autonomy for
central bank expansion.
Many studies fnd that capital controls do seem to be efective in insulat-
ing domestic interest rates and exchange rates from international factors,
but usually only to a moderate degree, especially if the time period is a
long one; and only if the overall set of macroeconomic policies are intern-
ally consistent. More recent evidence suggests, however, that capital con-
trols can help to avoid fnancial instability is stronger (Epstein et al., 2005).
Countries that had controls on outfows and/or infows were more likely
to be able to avoid the contagion from the Asian fnancial crisis. In the
long literature on controls on infows for Chile, there is evidence that Chile
was able to avoid an overvalued exchange rate and tilt toward a longer
maturity structure of borrowing. In the case of Malaysia (Kaplan and
Rodrik, 2002) controls on outfows temporarily allowed more expansion-
ary policy. Hence, the evidence can be read that, properly implemented,
controls can reduce instability and somewhat enhance macroeconomic
autonomy, at least to a limited extent. South Africa has a lot of experience
in implementing capital controls and should build on that experience to
tailor these capital management techniques to modern circumstances.
11.7 CONCLUSION
This chapter has tried to develop an ET monetary policy framework for
South Africa. I have emphasized that while monetary policy by itself
cannot solve the unemployment problem in South Africa, it must make a
contribution to doing so, and that simply stabilizing infation will not do
the trick. An ET framework will work best if it is part of an overall ET
macroeconomic strategy and if the central bank, as an institution, is com-
mitted to cooperating with the government to implement such a policy. I
have also argued that the central bank itself must bring to bear a coordi-
nated set of tools, including interest rate policy, credit allocation policy,
capital management techniques, and institutional development, to make
an ET approach efective.
NOTES
1. Thanks to K.S. Jomo Terry McKinley, Butch Montes and Jose Antonio Ocampo for
their encouragement, to my colleagues at PERI, Bob Pollin, James Heintz and Leonce
Ndikumana for their indispensable contributions on our larger project on employment
targeting in South Africa and to members of the Alternatives to Infation Targeting
group for their suggestions. A fnal thanks to my project co-organizer, Erinc Yeldan for
his contributions. All errors, of course, are mine.
2. See Robert Pollin et al. (2006) for more details on a general macroeconomic ET frame-
work for South Africa of which this central bank policy can be seen as a part.
3. Ibid.
4. International Financial Statistics (IFS), lending rate.
5. IFS.
6. http://www.csae.ox.ac.uk/resprogs/smmsae/datasets.html.
7. http://www.reservebank.co.za/.
8. IFS.
9. http://www.reservebank.co.za/.
10. Ibid.
REFERENCES
Amsden, A. (2001), The Rise of the Rest: Challenges to the West from Late-
Industrializing Economies, Oxford: Oxford University Press.
Aron, J. and J. Muellbauer (2002), Interest rate efects on output: evidence from
a GDP forecasting model for South Africa, IMF Staf Papers, 49 (November),
185213, accessed at (www.csae.ox.ac.uk/conferences/2002-UPaGiSSA/papers/
Aron-csae2002.pdf); non-technical summary at http://www.csae.ox.ac.uk/res-
progs/smmsae/nontechs/nontech07.html.
Ball, L. and N. Sheridan (2003), Does infation targeting matter?, International
Monetary Fund working paper no. 03/129.
Bernanke, B. et al. (1997), Systematic monetary policy and the efects of oil price
shocks, Brookings Papers on Economic Activity, 1, 91157.
Bernanke, B.S., T. Laubach, A.S. Posen and F.S. Mishkin (1999), Infation Targeting:
Lessons from the International Experience, Princeton, NJ: Princeton University Press.
Bruce-Brand, A.M. (2002), Overview of exchange controls in South Africa, state-
ment to the Commision of Inquiry into the Rapid Depreciation of the Exchange
Rate of the Rand, Reserve Bank of South Africa.
Dooley, M.P. (1995), A survey of academic literature on controls over internat-
ional capital transactions, National Bureau for Economic Research working
paper no. 5352.
Du Plessis, S. and B. Smit (2005), Economic growth in South Africa since 1994,
conference paper for the Economic Policy under Democracy: A 10 year Review
Conference, Stellonbosch University, 28-29 October.
Epstein, G. (2007), Central banks as agents of development, in Ha-Joon Chang
(ed.), Institutional Change and Economic Development, Helsinki: United Nations
University Press, accessed at www.wider.unu.edu/publications/working-papers/
research-papers/2006/on-GB/rp2006-54/_fles/78091779622373109/default/rp2006-
54.pdf.
Epstein, G., I. Grabel, K.S. Jomo (2005), Capital management techniques in
for the future, in Gerald Epstein (ed.), Capital Flight and Capital Controls in
Elgar, pp. 30133.
Farrell, G.N. (2001), Capital controls and the volatility of South African exchange
rates, South African Reserve Bank occasional paper no. 15, July.
Galindo, L.M. and J. Ros (2008), Alternatives to infation targeting in Mexico,
International Review of Applied Economics, forthcoming.
IMF (2003), South Africa: Selected Issues, International Monetary Fund country
report no. 03/18, January.
International Monetary Fund (IMF) (2004a), International Financial Statistics,
CD ROM.
IMF (2004b), South Africa: Selected Issues, International Monetary Fund country
report no. 04/379, December.
Jayadev, A. and K.K. Lee (2004), The efects of capital account liberalization
on growth and the labor share of income: reviewing and extending the cross-
country evidence, in Gerald Epstein (ed), Capital Flight and Capital Controls in
Elgar, pp. 1557.
Kaplan, E. and D. Rodrik (2002), Did the Malaysian capital controls work?, in
Sebastian Edwards and Jefrey A. Frankel (eds), Preventing Currency Crises in
Emerging Markets, Chicago, IL: University of Chicago Press, pp. 393441.
Lee, K. and A. Jayadev (2005), Capital account liberalization, growth and the
labour share of income: reviewing and extending the cross-country evidence,
in G. Epstein (ed.), Capital Flight and Capital Controls in Developing Countries,
Cheltenham, UK and Northampton, MA, USA: Edward Elgar, pp. 1557.
Pollin, R.N., G. Epstein, J. Heintz and L. Ndikumana (2006), An employment-
targeted economic program for South Africa, United Nations Development
Program Poverty Center, Brazil.
APPENDIX 11.A1 VARIABLES, DEFINITIONS AND
SOURCES FOR VAR MODEL
Variable Defnition Units Frequency Source Properties
RGDP
Growth
Growth rate
of real GDP,
annualized
Percent Quarterly South
African
Reserve
Bank
Stationary
Prime Prime lending
rate
African
Reserve
Bank
Stationary
Infation The infation
rate based on
CPI
metropolitan,
total
African
Reserve
Bank
First
diference is
stationary
Exchange
Rate
Nominal rand/
US dollar
exchange rate
(an increase is
depreciation)
Rand per
US dollar
Quarterly IMF,
Inter-
national
Financial
Statistics
(IFS)
Log frst
diference is
stationary
248
12. Infation targeting and the design
of monetary policy in India
Raghbendra Jha
1
12.1 INTRODUCTION
With around 20 central banks adopting it as their basic monetary policy
framework, infation targeting (IT) has emerged as an important mon-
etary policy framework. Over time IT has become more fexible in its inter-
pretation of target and permitted other goals to be included in the basic
framework. Central banks have enhanced their communication with their
respective publics about their targets and modus operandi.
Some authors have argued that for transition economies under going
sustained fnancial liberalization and integration in world fnancial
markets, IT is an attractive monetary policy framework. Consequently
there is pressure for such economies to adopt IT.
This chapter evaluates the case for IT in India. It begins in Section
12.2 by stating the objectives of monetary policy, especially that infation
control cannot be an exclusive concern of monetary policy in a country
with mass poverty. An evaluation of the rationale for IT and nuances
of implementation are spelt out in Section 12.3. Section 12.4 provides
some evidence on the efects of IT in developed and transition econo-
mies. Section 12.5 discusses Indias experience with using nominal targets
whereas Section 12.6 discusses some recent developments in Indian mon-
etary policy. Section 12.7 reviews some reasons why India is not ready
for IT. Section 12.8 shows that even if the Reserve Bank of India (RBI)
Indias central bank wanted to, it could not pursue IT since the short-
term interest rate (the principal policy tool used to afect infation in coun-
tries using IT) does not have signifcant efects on infation. Section 12.9
concludes and sketches the contours of an alternative to an IT policy. A
central feature of this policy is the use of capital controls. Appendix 12.A1
revisits the role that such capital controls played in ensuring that India did
not get severely afected by the East Asian fnancial crisis of the late 1990s,
even though Indias macroeconomic fundamentals at that time were com-
parable to those of countries that did experience such a crisis.
Infation targeting and the design of monetary policy in India 249
12.2 THE OBJECTIVES OF MONETARY POLICY IN
INDIA
An overriding short-term concern of monetary policy is stabilization of the
price level. However, India has had a long-standing problem of poverty
and its alleviation has to be the cornerstone of the success of any policy,
including monetary policy. Higher economic growth, along with some
supporting redistributive measures has a crucial role in reducing poverty,
even more than governance. Dollar and Kraay (2001) show for a broad
cross-section of countries including India, that the incomes of the poorest
20 percent of the population rise in proportion to average income.
2
China is an important example of the poverty reducing efects of eco-
nomic growth (Table 12.1).
For almost three decades Chinese per capita GDP has grown at more
than 8 percent per annum. Poverty has declined at an average of 9 240 000
persons per year.
In contrast Indias growth and poverty reduction record has been less
spectacular (Table 12.2).
Because of lower growth the reduction in poverty in India has been far
lower than in China notwithstanding the fact that inequality in China has
grown more sharply than in India (Jha, 2004). Indias national poverty
headcount ratio fell only by about 12 percentage points between 195152
to 1997
3
and the rate of poverty reduction was higher in the 1980s than in
the reform period, post-1991, indicating that the quality of growth in the
Table 12.1 Growth and poverty alleviation in China
Year Annual
poverty
reduction
announced
by the
government
(10 thousand)
Average
annual
growth
rate of GDP
per capita (%)
Average
annual
growth rate
of farmers
consumption
level (%)
Average
annual
growth rate
of farmers
net income
per capita (%)
197885 1786 8.3 10.0 15.1
198590 800 6.2 2.5 3.0
199097 500 9.9 8.0 5.0
199702 436 7.7 3.4 3.8
197802 924 8.1 5.6 7.2
Source: Chinese Statistical Abstract, various issues.
1980s was diferent from that in the 1990s, so that even with lower growth
greater reduction in poverty could take place in the 1980s. This trend has
recently been reversed. Jha et al. (2006) report that the rural headcount
ratio, which is usually higher than the urban headcount ratio, was 20.6
percent in 2004 compared to higher than 26 percent in 1999. Hence there
exists scope for poverty reduction through higher growth in India.
Furthermore, unemployment has been rising in the post-reform period.
In the organized sector employment barely changed between 1991 and
2001; from 1997 it actually fell. Total employment (organized and unor-
ganized) is growing at about 1 percent per annum half the projected
growth rate of the labor force. Consequently, the unemployment rate for
males in the rural sector rose from 5.6 percent in 199394 to 9 percent in
2004. In addition India also has substantial underemployment.
If India is to reduce poverty rapidly, its GDP must grow at 8 percent
or more on a sustained basis. Kelkar (2004) argues that growth could
accelerate essentially because of a broad series of fnancial sector reforms,
increased globalization and widening and deepening of product and fnan-
cial markets, and benefcial structural changes particularly on the supply
side. In addition to its surplus labor India is set to reap an important
demographic dividend as the proportion of population in the working age
group (1564 age bracket) is expected to climb from 60.9 percent in 2000 to
over 66 percent by 2030. The labor force is less nutritionally deprived and
increasingly literate. Economic theory and international experience indi-
cate that this could lead to sharp rises in labor productivity and an upward
shift in the trend rate of growth. However, this labor force must be produc-
tively employed for these productivity gains to be realized (Jha, 2005).
Low interest rates (to enhance investment) and a slightly undervalued
exchange rate with low volatility (to boost exports) are critical to sustain-
ing high growth rates. An appropriate monetary policy must address these
requirements.
Table 12.2 GDP and per capita GDP growth in India
Period GDP growth (%)
Aggregate
(average annual)
Per capita
(average annual)
197282 3.5 1.2
198292 5.2 3.0
19922002 6.0 3.9
Source: Kelkar (2004).
12.3 EVALUATING THE RATIONALE FOR
INFLATION TARGETING
The time inconsistency literature argues that a discretionary policy setting
leads to higher infation without any gains in output (Barro and Gordon,
1983; Kydland and Prescott, 1977). The high welfare cost of infation
raises the appeal of rules-based policy. Under a rules-based regime central
banks set explicit values for intermediate targets, which they can control,
and which are strongly related to the ultimate goals of monetary policy
(viz. stabilization of output and infation), which cannot be directly
controlled.
In recent times emerging market economies, including India, have
experimented with three nominal targets: exchange rate, money supply
growth and infation.
4
The relative advantages/disadvantages of exchange
rate and money growth targeting are portrayed in Table 12.3.
Bernanke and Mishkin (1997) argue that IT, in contrast to exchange
rate targeting but like monetary targeting, enables monetary policy to
focus on domestic considerations and to respond to shocks to the domestic
economy.
5
IT, like exchange targeting and unlike monetary targeting, has
the advantage that people easily understand it. Since the central bank has
a numerical infation target, the chances of slipping into a time inconsist-
ency trap are reduced. Svensson (1999), Bernanke and Mishkin (1997)
and White (2004) argue that IT is decision making under discretion
with central banks following a targeting rule which sets interest rates to
reduce the deviation between conditional infation forecast (the interme-
diate target of monetary policy) and the infation target to zero over the
horizon.
In an emerging market economy such as India the problems of mon-
etary management, in general and infation control, in particular, get
compounded by low policy credibility (Calvo and Mishkin, 2003). This
can lead to sudden stops of capital infows, making such countries prone
to fnancial crises. This increases the appeal of rules-based monetary
regimes (like IT). Taylor (2002) argues that rules-based policies enhance
the anticipation efects of monetary policy. Given less developed fnancial
markets such anticipatory efects are likely to be lower. Yet monetary
policy could still have considerable efects through movements of wages
and property prices. With an IT regime shocks from the monetary regime
may be lower.
However, this rationale for IT is incomplete. For instance, Kirsanova et
al. (2006) show, in the context of the UK, that an IT regime in itself cannot
be considered to be an optimal monetary policy framework and optimal
policy response must include terms of trade or exchange rate terms an
Table 12.3 Advantages and disadvantages of the nominal anchors of
exchange rate targeting and monetary targeting
Anchor: Exchange rate targeting
Advantages 1. This fxes the infation rate for internationally traded goods and
thus directly contributes to keeping infation under control. It
is especially useful for sharply reducing infation in emerging
market economies
2. If the exchange rate peg is credible, it anchors infation
expectations to the infation rate in the anchor country to whose
currency it is pegged
3. An exchange rate provides an automatic rule for the conduct of
monetary policy that avoids the time-inconsistency problem
4. An exchange rate is simple and direct and, therefore, is well
understood by the public
Disadvantages 1. An exchange rate target leads to loss of independent monetary
policy (Obstfeld and Rogof, 1996). Hence the ability of the
monetary authorities to respond to shocks is compromised
2. The exchange rate peg may persuade large-scale foreign
borrowing. In the case of emerging market economies such
loans are invariably denominated in foreign currency. Large
accumulation of such loans may lead to a crisis. In most
developed countries a devaluation may have little direct efect
on the balance sheets (since debts are denominated in home
currency) but not so in emerging market economies since debts
are denominated in foreign currency
3. Bernanke and Mishkin (1997) argue that exchange rate pegs can
lead to fnancial fragility
4. Although exchange rate targeting may be initially successful in
bringing infation down a successful speculative attack can lead to
a resurgence of infation
Anchor: Monetary targeting
Advantages 1. An advantage over exchange rate targeting is that monetary
targeting enables a central bank to adjust its monetary policy to
cope with domestic considerations
2. A monetary target is easily understood by the public but not as
well as an exchange rate target
3. Monetary targets have the advantage of being able to promote
almost immediate accountability for monetary policy
Disadvantages 1. Typically the link between money growth and infation is subject
to long and uncertain lags
2. The demand for money may not be stable, there may be
instability of velocity and the money supply may not be
controllable (Jha and Rath, 2003). This is especially true of broad
monetary targets such as M2 or M3 and less so of narrow money
argument likely to hold with greater certainty for developing and tran-
sition countries such as India.
Even if IT leads to price stability it may not guarantee fnancial sta-
bility. The RBI (2004) notes that the 1990s a decade of relative price
stability witnessed a number of episodes of fnancial instability indi-
cating that price stability is not suf cient for fnancial stability. Large
movements in capital fows and exchange rates afect the conduct of
monetary policy continually. Bernanke and Gertler (2001), Bernanke
(2003), Bean (2003) and Filrado (2004) argue that even if price stability
does not imply fnancial stability in the short run, price stability does
not endanger fnancial stability and price stability, and fnancial stability
would reinforce each other in the long run. In an economy with relative
price stability the interest rate should not remain passive (as it would in
an IT regime) in the presence of a sudden capital outfow. Thus the RBI
(2004) notes [while] there is very little disagreement over the fact that
price stability should remain a key objective of monetary policy, reserva-
tions persist about adopting it as the sole objective of monetary policy
(p. 56).
6
Empirical evidence suggests that in emerging market economies central
bank interest rates react more strongly to changes in the exchange rate
rather than changes in the infation rate or output gap (Mohanty and
Klau, 2004). Hence, at this time, the standard tool to target infation
short-term interest rate is unlikely to be particularly useful.
7
12.3.1 The Mechanics of Infation Targeting
The modus operandi of a typical IT regime is as follows. The central
bank revises its infation and output forecast at a frequency determined
by monetary policy committee meetings using updated information. If
the conditional infation forecast is higher than the target, the central
bank raises the interest rate to minimize such deviation by the end of
the targeting horizon, and vice versa. Households and frms then decide
upon their consumption and investment plans. Blinder (1998) and Taylor
(1993, 2002) argue that this is close to what many policy makers do in
practice.
It has become common to compare ex post the actual setting of policy
rates by central banks with what would have been predicted by the Taylor
rule which suggests that (short-term) interest rates (the federal funds rate
in the USA or the bank rate in India) should be changed in response to
deviation of infation from a target and an output gap the so-called reac-
tion function of central banks (Clarida et al., 1998; Mohanty and Klau,
2004; Svensson, 1999). In applying his rule to the USA for the 198792
period Taylor shows that it described the actual performance of policy
well. IT is not applied mechanically and focuses on containing infation as
a medium-term goal (Bernanke and Mishkin, 1997).
8
The choice of the price index to be used by an IT regime is critical.
Typically the consumer price index (CPI) or, preferably, a measure of core
infation that ignores excessively volatile prices, for example, food and
energy is used.
9
Whether IT should respond to asset prices is debatable.
Bean (2003) analyses an objective function minimizing output gaps and
deviation from infation targets and argues that a middle solution between
completely ignoring asset prices and including asset prices in the price
index number, if these afect infationary expectations, be used for IT.
Central banks operate in an environment of considerable uncertainty
about the functioning of the economy and global capital fows. Hence
the conduct of monetary policy must be informed by examining a
number of indicators and cannot rely on just one intermediate target.
Most central banks, including those of developed countries, practice
liquidity management involving estimating market liquidity, auton-
omous of policy action, and initiate liquidity operations to steer mon-
etary conditions and are thus able to switch between quantitative targets
and interest rate targets in response to macroeconomic circumstances.
Most central banks try to build in automatic stabilizers in the liquidity
management framework, for example, by setting reserve requirements
on an average basis to allow the fnancial system the leverage to adjust
to temporary/seasonal liquidity shocks on its own account without
central bank action and exercising an explicit preference for encasing
short-term interest rates in a corridor around some optimal rate rather
than at a point target.
12.4 THE PERFORMANCE OF INFLATION
TARGETING
There is considerable debate about whether IT improves performance
in regard to infation and output. Ball and Sheridan (2003) argue that
the adoption of IT does not lead to a systematic improvement in the
growth-infation trade-of; Hu (2004) argues otherwise. Fraga et al.
(2003) focus on emerging market economies, including India, and show
that economies working with IT have higher volatilities of output, infa-
tion, interest rates and exchange rates than developed countries using IT
(Table 12.4).
Preparing for a switch to an IT regime requires considerable back-
ground work. Financial markets should be suf ciently developed
and global capital markets should have adequate confdence in these
markets, thus enabling the adoption of a suf ciently fexible exchange
rate regime and the central bank should have a high degree of indepen-
dence and be able to use short-term interest rates as the main operating
instruments. Such conditions may not be satisfed in many transition
countries.
Table 12.4 Volatility and average of selected variables for 1997:12002:2
(quarterly data)
Countries Volatility of basic variables Average
Infation Exchange
rate*
GDP
growth**
Interest
rate
GDP
growth
Infation
Developed Economies
Australia 2.05 0.13 1.96 0.58 4.78 5.89
Canada 0.83 0.04 1.30 1.14 3.57 1.96
Iceland 2.45 0.15 3.13 3.02 4.17 4.05
New Zealand 1.21 0.16 3.61 1.47 3.09 1.65
Norway 0.77 0.10 2.25 1.46 2.66 2.44
Sweden 1.11 0.12 2.41 0.44 2.58 1.24
Switzerland 0.54 0.08 1.14 0.92 1.79 0.85
United Kingdom 0.92 0.06 0.79 1.13 2.61 2.46
Average 1.24 0.11 2.07 1.27 3.16 2.57
Median 1.02 0.11 2.11 1.13 2.88 2.20
Emerging Market Economies
Brazil 2.09 0.31 2.06 7.06 1.81 5.89
Chile 1.30 0.17 3.25 3.11 3.88
Colombia 5.43 0.25 3.38 10.02 0.81 12.51
Czech Republic 3.46 0.09 2.73 5.81 1.18 5.31
Hungary 4.09 0.16 1.13 11.21
Israel 3.18 0.10 3.36 3.34 2.98 4.35
Mexico 5.98 0.07 3.17 7.26 4.05 11.72
Peru 3.04 0.11 3.45 5.50 2.11 3.89
Poland 4.13 0.11 2.40 4.14 3.85 8.40
South Africa 2.13 0.26 1.11 3.65 2.26 6.51
South Korea 2.36 0.14 6.38 5.52 4.31 3.73
Thailand 3.25 0.14 6.13 6.72 0.08 2.88
Average 3.37 0.15 3.40 5.47 2.41 6.69
Median 3.22 0.14 3.25 5.52 2.26 5.60
Note: * refers to the coef cient of variation (standard deviation/mean); ** growth rate
measured comparing the current quarter to the same quarter of the previous year.
Source: International Financial Statistics, IMF (quarterly data) (2003).
12.5 RECENT INDIAN EXPERIENCE WITH
NOMINAL TARGETING
The RBI has never pursued a pure nominal targeting regime, opting for
a combination of rules-based and discretionary measures with the former
changing over time. In the 1980s and early 1990s the RBI opted for a
nominal exchange rate peg externally, and monetary control internally.
However, both policy mechanisms have faltered. An infexibly pegged
exchange rate has proved to be unsustainable in the presence of strong
capital fows
10
whereas the instability of the money demand function as
well as its supply (Jha and Rath, 2003) indicates that monetary targeting,
by itself, is no longer a feasible option.
Further, the RBI faces a persistent fscal overhang and has to support
high fscal defcits. If fscal policy is imprudent and the central bank does
not help fnance the defcit, the end result would still be infationary as the
public debt/GDP ratio would turn unsustainable in the medium term and
the price level could at least partially be determined by the fscal theory of
the price level.
11
Fiscal defcits are infationary and put pressure on real interest rates
and crowd out private investment (Engen and Hubbard, 2004). There is
a vicious cycle between infation and budget defcits high defcits cause
higher infation, which raise interest rates, thus increasing the defcit by
raising debt service payments. Further, higher infation reduces the real
value of tax collections.
The literature has emphasized frameworks based on the clear mandates
of central bank independence and fscal responsibility legislation. Fiscal
rules restrict government spending which checks excessive build-up of
defcits and public debt, imparting stability to the economy. Concurrently
fscal rules may restrict the governments ability to take countercyclical
policy measures and hence contribute to increased business cycle volatil-
ity. Fiscal policy rules are likely to be efective if accompanied by strong
commitments and increased transparency (Bayoumi and Eichengreen,
1995) hence the widespread consensus for central bank independence
backed by fscal discipline to contain infation and stabilize infationary
expectations.
Although price stability, output growth, reduction of exchange rate
vola tility and fnancial stability are monetary policy goals for the RBI,
none of these are under its direct control. The RBI sets intermediate targets
which it can control and which have a stable relationship with the ultimate
goals of monetary policy. A narrow target such as base money may be
fully within RBI control but may have only weak links with the ultimate
objectives of monetary policy. A broad target such as nominal income
may be closely related to the ultimate objectives of monetary policy but
not be amenable to RBI control. Both money supply and money demand
have become unstable since the initiation of fnancial sector reforms.
Other nominal targets have similar problems. Hence a purely rules-based
monetary policy regime seems unhelpful.
12.6 RECENT DEVELOPMENTS IN MONETARY
POLICY DESIGN IN INDIA
With the progressive widening of fscal defcits from the 1960s, the burden
of fnancing was borne by the RBI and the banking system. The support
of the banking system to the governments borrowing program involved
progressive increases in the statutory liquidity ratio to 38.5 percent by the
early 1990s. Although interest rates on government securities were steadily
raised to enhance their attractiveness, it became increasingly dif cult to get
voluntary subscriptions even at high interest rates. The cash reserve ratio
was increased from 3 percent in the early 1970s to almost 25 percent (if
incremental reserve requirements are taken account of) by the early 1990s.
Nevertheless, liquidity growth remained excessively high during the 1970s
and 1980s and spilled over onto infation. With expansionary fscal policy
there are limits to the efectiveness of monetary policy in containing infa-
tion. The combined defcit of central and state governments has been close
to 10 percent of GDP for more than 15 years and the share of net bank
credit to the government in fnancing the fscal defcit has hovered around
10 percent of GDP for much of the past decade.
The 1985 Chakravarty Committee on Monetary Policy recommended
that price stability emerge as the dominant objective of monetary policy
along with commitment to fscal discipline (RBI, 2002, p. 67). Price stabil-
ity was seen to be critical to sustain the process of reforms begun in 1991
(RBI, 1993). In the latter half of the 1990s, as the economy slowed down,
monetary policy pursued an accommodative stance with an explicit prefer-
ence for a softer interest rate regime with a constant vigil on infation.
The RBI formally adopted a multiple indicator approach in April 1998.
These are to (1) maintain a stable infation environment; (2) maintain
appropriate liquidity conditions to support higher economic growth;
(3) ensure orderly conditions in the exchange market to avoid excessive
volatility in the exchange rate; and (4) maintain stable interest rates (RBI,
2002).
Besides broad money, which remains an information variable, other
macroeconomic indicators including interest rates, rates of return on
money, capital and government securities markets along with data on
currency, credit extended by banks and fnancial institutions, fscal posi-
tion, trade, capital fows, infation rate, exchange rate, refnancing and
transactions in foreign exchange available on a high frequency basis are
juxtaposed with output data for drawing policy perspectives when formu-
lating monetary policy.
This check list approach has been criticized as watering down the
concept of a nominal anchor. However, it is dif cult to fnd a variable
that would encapsulate the large number of factors that need to go into
monetary policy making at this stage of transition from a relatively
autarkic administered economy to a relatively open market-oriented one.
The RBI uses a mix of policy instruments including changes in reserve
requirements, and standing facilities and open market operations which
afect the quantum of marginal liquidity and changes in policy rates, for
example, the bank rate and repo/reverse repo rates, which impact the price
of liquidity with short-term interest rates signaling the stance of monetary
policy.
Shifts in monetary policy transmission channel necessitated policy
impulses which would travel through both quantity and rate channels and
episodes of volatility in foreign exchange markets emphasized the need
for swift policy reactions balancing the domestic and external sources of
monetization in order to maintain orderly conditions in fnancial markets.
Even within the set of indirect instruments the preference is for market-
based instruments.
Another serious challenge comes from the capital account, especially
the high volatility of capital fows vis--vis trade fows.
12
Since external
borrowings are denominated in foreign currency, large devaluations are
infationary and cause serious currency mismatches with adverse efects
on the balance sheets of borrowers. The need for reserves as self-insurance
emanates from the volatile nature of capital fows (including sharp rever-
sals) and refects weakness in the existing international fnancial architec-
ture. Indias ratio of net foreign assets to reserve money grew from 11.9
percent in 1990 to 44.5 percent in 1996, 65.8 percent in 2000 and 117.3
percent in 2003.
12.7 INDIAS UNPREPAREDNESS FOR INFLATION
TARGETING
That transition economies such as India may not be ready for IT is the
considered view not just of the RBI but also International Monetary Fund
(IMF) economists. Masson et al. (1997) argue that economic structures in
developing countries (including India) are incapable of supporting an IT
regime in the short to medium runs, because such countries do not satisfy
a number of prerequisites for the successful implementation of IT. The
authors consider these to be:
1. Independence of the central bank: this refers both to operational
conditions and the policy space within which the RBI can operate.
There are limits to the efectiveness of monetary policy in containing
infation when faced with expansionary fscal policy. Domestic and
fnancial markets should have enough depth to absorb the place-
ment of public and private debt instruments; and the accumulation of
public debt should be sustainable. In the Indian case, while there is
some evidence to suggest that the latter condition is satisfed (Jha and
Sharma, 2004), the frst is defnitely not (Sharma, 2004). If both
condit ions are not satisfed the independence of monetary from fscal
policy is compromised and the interest rate transmission channel of
policy is weak and incompletely evolved.
In addition, the central government can, even with fnancial liber-
alization, apply subtle pressure on the RBI to alter monetary policy.
I give two instances of these. In the latter half of 2004, when infation
topped 8 percent and real interest rates had become negative, the RBI
wanted to raise the bank rate to lower infation but could not, under
government pressure. Similarly in early 2005 the Governor of the RBI
publicly voiced concern over volatile FII infows and suggested a fscal
approach to capping them. However, the Finance Minister almost
immediately rebufed him.
2. Refraining from using other nominal anchors: successful adoption
of IT requires that other nominal variables such as the exchange rate
should not be targeted. However, India needs to maintain a stable
and competitive exchange rate to encourage exports. Even in devel-
oped economies, which have explicitly opted for it, IT is associated
with high exchange rate volatility. In view of their vulnerability to
exchange rate crises, developing countries such as India are wary of
excessive volatility.
3. Predominance of demand as opposed to supply shocks: IT implicitly
assumes that monetary policy has to respond primarily to demand
shocks. Balakrishna (1991) has underscored the role of supply shocks
in determining infation in India.
13
These make infation dependent on
monetary as well as non-monetary factors. If infation rises because
of a demand shock, the pursuit of IT will stabilize both infation and
output. However, if infation rises because of an adverse supply shock,
the pursuit of IT will exacerbate the recessionary efect on output by
reducing demand (McKibbin and Singh, 2003).
4. Practical dif culties in the implementation of IT: the high frequency
data requirements including those of a fully dependable infation rate
for targeting purposes are not yet available (RBI, 2004).
12.8 CHECKING FOR VIABILITY OF INFLATION
TARGETING IN INDIA
A prerequisite for the RBI to pursue IT is that there should be a stable and
signifcant relationship between the measure of infation to be controlled
and short-term interest rates. I test for this using monthly data over the
period April 1992 to March 1998 from the RBIs Handbook of Statistics on
the Indian Economy (2000). The variables used are as follows:
IIP: Index of industrial production (1980815100); REER: Index of
real efective exchange rate (36-country), 19855100; Narmon: Narrow
money; Cmrate: Call money rate; Xrate: Exchange rate of Indian rupee
vis-a-vis US dollar (monthly averages); CPI: Consumer price index for
industrial workers (1982 5100); WPITR20: Trimmed WPI (Mohanty
et al., 2000); WPI: Wholesale price index (199394 5100); WPIADM:
Wholesale administered price index (ibid.). Monthly dummies were added
and logs were taken of all variables except the Cmrate. Augmented Dickey
Fuller tests (not reported here to conserve space) indicated that all series
are I(1).
14
The bivariate relationships between the three candidate infation meas-
ures and the monthly economic indicators, the P values from bivariate
Granger causality tests are reported in Table 12.5. Each entry gives the P
values for the null hypothesis that the indicator does not cause the infa-
tion measure, that is, the probability of obtaining a sample, which is even
less likely to conform to the null hypothesis of no Granger-causality than
the sample at hand. These Granger causality results are reported up to
eight lags.
The WPITR20 measure of infation assumes that the WPI is the head-
line measure of infation and defnes the trimmed mean infation index as:
WPITR
a
5
1
a1 2 2a
a
100
bb
a
n21
i 5k1l
w
i
p
i
(12.1)
where WPITR
a
is the trimmed WPI computed by ordering the component
price change data p
i
and associated weights w
i
and removing the compo-
nents on each tail of the distribution by a percent. The numbers of com-
ponents trimmed from the left and right tails of the distribution are k and
l respectively. When a 5 0 the trimmed mean would equal the weighted
mean whereas in the case of a 5 50 it would equal the weighted median.
The root mean square error (RMSE) for any level of trimming is defned
by:
RMSE
a
5
a
n
i 51
(p
t
a
2 p
t
)
2
/n (12.2)
Table 12.5 P Values from Bivariate Granger Causality Tests
CPI IIP Exrate Narmon REER Cmrate
Lags
1 0.22 0.67 0.00* 0.64 0.35
2 0.4 0.72 0.00* 0.99 0.43
3 0.69 0.87 0.00* 0.61 0.93
4 0.1 0.5 0.00* 0.46 0.8
5 0.01* 0.25 0.00* 0.36 0.55
6 0.00* 0.13 0.00* 0.26 0.58
7 0.00* 0.12 0.00* 0.12 0.69
8 0.00* 0.16 0.00* 0.03* 0.82
WPITR20
Lags
1 0.06 0.00* 0.07 0.01* 0.14
2 0.01* 0.00* 0.01* 0.00* 0.09
3 0.00* 0.00* 0.00* 0.00* 0.00*
4 0.00* 0.00* 0.00* 0.00* 0.04*
5 0.00* 0.00* 0.00* 0.00* 0.19
6 0.00* 0.00* 0.00* 0.00* 0.2
7 0.00* 0.00* 0.00* 0.00* 0.14
8 0.00* 0.00* 0.00* 0.00* 0.26
WPIADM
Lags
1 0.00* 0.33 0.08 0.75 0.45
2 0.00* 0.09 0.01* 0.56 0.23
3 0.00* 0.03* 0.00* 0.44 0.18
4 0.00* 0.00* 0.00* 0.2 0.97
5 0.00* 0.00* 0.00* 0.01* 0.44
6 0.00* 0.00* 0.00* 0.00* 0.41
7 0.00* 0.00* 0.00* 0.00* 0.4
8 0.00* 0.00* 0.00* 0.00* 0.12
Note: * 5 Signifcant at 5 percent level; CPI: Consumer price index; WPITR: Trimmed
whoesale price index; WPIADM: Price index for the administered goods; IIP: Index of indus-
trial production; Narmon: Narrow money; Exrate: Exchange rate Rs/$; REER: Real efective
exchange rate; Cmrate: Call money rate.
where p
a
t
is the trimmed WPI with a trimming ratio of a percent from
each of the tails of the price distribution at time t, p
t
is the 36-month
centered moving average change in WPI at time t, and n is the number
of samples. Mohanty et al. (2000) conclude that this RMSE is mini-
mized for a 5 20. Data on this variable are available in Mohanty et al.
(2000).
The results of the Granger causality test indicate a weak relation
between Cmrate and the measures of infation. In fact only WPITR20
seems to have a causal relation with Cmrate. On the other hand, the links
between the measures of infation and IIP, Narmon, Xrate and REER are
much stronger. Hence the causality tests do not provide support for using
interest rates as instruments in a policy of IT.
A drawback of the crude Granger causality testing is that it provides no
information about whether the sign of the (dynamic) bivariate relationship
is theoretically correct. Further, there may be omitted variables. A VAR
on the variables lCPI, lIIP, lNarmon, lREER and Cmrate indicates that
the conclusions of Table 12.5 are broadly correct.
15
Figure 12.1 shows that the 95 percent confdence band for the
0.02
0 10
step
model 1: Cmrate > LCPI
20 30
0.01
0
0.01
95% CI for OIRF
OIRF
Note: The band indicates the 95 percent confdence interval for the orthogonalized impulse
response function.
Figure 12.1 Orthogonalized impulse response function (OIRF) of call
money rate on log CPI
orthogonalized impulse response function of Cmrate on lcpi is very wide,
hence adding to our agnosticism about the ef cacy of IT in India.
The error correction model for lCPI is not signifcantly responsive to
any of the error correction terms. Hence it appears that IT may be dif cult
to pursue in the Indian context.
12.9 CONCLUSIONS: OPTIONS FOR INFLATION
TARGETING IN INDIA
This chapter has argued that the primary objective of Indian monetary
policy, at least in the medium term, has to be the attainment of higher
economic growth. Moreover, since India has high infation aversion, this
objective does not confict with short-term stabilization.
The design of monetary policy in India is circumscribed by the fact
that the liberalization of fnancial markets is far from complete so that
the interest rate transmission channel is incomplete. Further, the banking
system has strong monopoly elements. Moreover, as the fnancial sector
liberalizes some major government-owned mutual fund operations have
had to be bailed out. The exacerbation of such contingent liabilities along
with already high fscal defcit aggravates monetary policy dif culties in
the Indian context.
Against this background this chapter has argued that the multi- objective
formulation pursued by the RBI has merit and that such monetary policy
should be pursued to maintain stable interest and infation rates and
a slightly undervalued currency in order to engineer higher export-led
growth.
This policy has, however, led to substantial capital infows with attend-
ant build-up of reserves and necessitated considerable sterilization opera-
tions. This has now emerged as a signifcant problem with its continuance
at the current pace seemingly unsustainable if, for no other reason, than
the fact that such reserves attract low yields. Currently two policy pack-
ages to address this issue have been discussed. The frst is geared towards
fscal correction and monetary expansion. A second policy measure is
weighted towards real exchange rate appreciation (more in line with IT)
and would involve relatively larger current account defcits. Real appreci-
ation could be secured by nominal appreciation or higher infation. Both
policies would lead to low infation rates and reduced infows of foreign
capital and, therefore, lower accumulation of reserves at given rates of
sterilization. Policy packages that use import liberalization would, like
real appreciation, permit higher absorption via higher current account
defcits but without penalizing exports. The optimal package for India is
a judicious combination of these two broad sets of policies with greater
emphasis on fscal consolidation and import liberalization, rather than
real exchange rate appreciation. These are essential elements of an appro-
priate monetary policy regime for India.
Since rapid export growth is important, it makes sense to err on the
side of undervaluation of the exchange rate, enabling India to capture a
larger share of world markets. Growing exports, in turn, raise the incen-
tive to invest. The propensity to save also rises in response to the increased
proftability of export-oriented investment. Moreover, an undervalued
exchange rate is likely to boost saving by raising the share of profts in
national income. This argument does not imply that unlimited real depre-
ciation is feasible or desirable, just that there should be a bias towards
mild undervaluation because it can play a supportive role to complemen-
tary outward-oriented trade policies in generating a virtuous circle of
higher saving, investment and growth. Along this path import demand
would grow concomitantly and getting a current account surplus is not
inevitable.
Clearly India has been conducting some form of real exchange rate tar-
geting leading to a sharp rise in foreign exchange reserves. Lal et al. (2003)
indicate that this has come at high economic costs a claim that has been
successfully disputed.
16
IT would require India to pursue a clean foat
reducing the need for large reserves. But the price to be paid is the pos-
sibility of a highly unstable or inappropriate exchange rate. Indias policy
makers were wise to reject this regime and opt for managed foating plus
selective controls on capital fows. Reserves are now at a very comfortable
level and continue to rise rapidly. The question of whether and how to
absorb foreign infows is far more pertinent now than it was in the 1990s.
Clearly sterilization has outlived its usefulness. Some sterilized reserve
accumulation can continue to maintain the present ratio of reserves
to GDP. Further increases in the ratio should be avoided except as a
purely short-term response to manifestly short-term infows. The policies
espoused here have the advantage that, in addition to promoting balance
of payments adjustment, they are desirable independently of the balance of
payments and of the temporary or permanent character of the infows.
Naturally, due to political constraints, these policies can only be pursued
at a moderate pace. If there is continued acceleration of infows, despite
the adoption of the suggested strategy, the government should consider
tightening capital infow controls
17
so that the strategy is not derailed. A
corollary would be that capital account convertibility is eschewed.
It is not being suggested that India should resist an exchange rate appre-
ciation indefnitely. Once Indian GDP has grown in excess of 8 percent for
more than two decades, so that real incomes have gone up substantially
and unemployment and poverty have dropped sharply, India could con-
template changes in the monetary policy regime, essentially in the direc-
tion of relaxing the undervaluation of the exchange rate. However, even
at that point, given the importance of variables other than infation in
an optimal monetary policy framework, there would be no argument for
moving to IT.
NOTES
1. I am grateful to Gerald Epstein and three anonymous referees for helpful comments
and Anurag Sharma for research assistance. The usual disclaimer applies.
2. At the very least there is no evidence that economic growth hurts poverty alleviation
(Winters et al., 2002).
3. Results from the 19992000 National Sample Survey (Central Statistical Organisation,
2000) indicate a larger drop in poverty; however, this surveys methodology is not com-
parable with those of the earlier surveys, correcting for which reveals a modest drop in
poverty.
4. Nominal income, another intermediate target, is both hard to target and poorly related
to the ultimate objectives of monetary policy.
5. Another alleged advantage of an IT regime is that deviations from infation targets are
routinely allowed in response to supply shocks by excluding some combination of food
and energy prices, indirect tax changes, terms of trade shocks and the direct efects of
interest rate changes on the index.
6. See also Epstein (2000).
7. For an application to India, see Section 12.8.
8. Seyfried and Bremmer (2003) discover that the Reserve Bank of Australia pays par-
ticular attention to infationary pressures, as measured by the GDP gap. They fnd a
relatively high degree of persistence and low speed of adjustment in the interest rate
indicating that the central bank is interested in interest smoothing in addition to IT. See
also Lomax (2005).
9. India with one wholesale price index and four CPIs still does not have a single price
index with widespread acceptability for IT. Measures of core infation are not com-
puted of cially (Mohanty et al., 2000).
10. Joshi and Sanyal (2004) indicate the RBI has been targeting the index of real efective
exchange rate (REER) of the Indian rupee with regard to the currencies of fve coun-
tries, USA, Japan, UK, Germany and France, at the 199394 level. Patel and Srivastava
(1997) note that such targeting has had more than a transitory efect. However this
benign relationship may break as reforms lead to greater capital mobility.
11. In Latin America Jacome and Vazquez (2005) fnd no causal relationship between
central bank independence and infation, although the association between the two is
strong.
12. The Indian rupee is convertible on the current account but capital account converti-
bility is not permitted.
13. Callen and Chang (1999) review the literature on infation in India. The threshold of
infation over which price stability should take precedence over the growth objective
has been estimated to be between 4 and 6.5 percent (Samantaraya and Prasad, 2001).
14. Since the Mohanty et al. (2000) data set stops at 1998 we do not extend our analysis
beyond that date.
15. Detailed results are reported in the full version of the paper available on the PERI
website, accessed 6 May 2007 at http://www.peri.umass.edu.
16. Lal et al. (2003) argue that Indias growth rate in the 1990s could have been up to
2.7 percent per annum higher if foreign exchange infows during the decade had been
fully absorbed. However, Joshi and Sanyal (2004) argue that if net foreign infows had
been absorbed domestic spending (and not foreign exchange reserves) would have
risen. Reserves as a proportion of GDP rose over the 1990s by an average of about 1.2
percent per annum. If the entire increase in reserves had been absorbed into investment
each year, the ratio of investment to GDP averaged over the decade would have been
1.2 percent higher than it actually was. Given fxed incremental capital output ratio
(ICOR) (of 2.8 in the 1990s) the increase in Indias growth rate of GDP would have
been only 1.2/2.8 5 0.4 percent per annum (approx.) higher over the decade. This sacri-
fce would be even lower if (1) the ICOR would have risen (in line with the assumption
of diminishing returns to capital); (2) some of the reserve accumulation spilled over
onto higher consumption, thus reducing the growth rate; and (3) furthermore, the level
of foreign exchange reserves in India was inadequate in 1991 and accumulating foreign
reserves was necessary.
17. Through a Chilean-type tax, for example.
REFERENCES
Balakrishna, P. (1991), Pricing and Infation in India, Delhi: Oxford University
Press.
Ball, L. and N. Sheridan (2003), Does infation targeting matter?, National
Bureau for Economic Research, working paper no. 9577.
Barro, R. and D. Gordon (1983), A positive theory of monetary policy in a
natural rate model, Journal of Political Economy, 91 (4), 589610.
Bayoumi, T. and B. Eichengreen (1995), Restraining yourself: the implications of
fscal rules for economic stabilization, IMF Staf Papers, 42, 3248.
Bean, C. (2003), Asset prices, fnancial imbalances and monetary policy: are infa-
tion targets enough?, Bank for International Settlement working paper no.
140.
Bernanke, B. (2003), Constrained discretion and monetary policy, BIS Review,
5, 18.
Bernanke, B. and M. Gertler (2001), Should central banks respond to movements
in asset prices?, American Economic Review, 91 (2), 2537.
Bernanke, B. and F. Mishkin (1997), Infation targeting: a new framework for
monetary policy?, Journal of Economic Perspectives, 11 (2), 97116.
Blinder, A. (1998), Central Bank in Theory and Practice, Cambridge, MA: MIT
Press.
Callen, T. and D. Chang (1999), Modeling and forecasting infation in India,
International Monetary Fund working paper WP/99/119, Washington, DC.
Central Statistical Organisation (2000), National Sample Survey, New Delhi:
Government of India.
Clarida, R., J. Gali and M. Gertler (1998), Monetary policy rules in practice: some
international evidence, European Economic Review, 42 (4), 103367.
Dollar, D. and A. Kraay (2001), Growth is good for the poor, World Bank policy
research paper no. 2587, pp. 150.
Engen, E. and R. Hubbard (2004), Federal governments and interest rates,
National Bureau for Economic Research working paper no. 1068.
Epstein, G. (2000), Myth, mendacity and mischief in the theory and practice of
central banking, accessed 6 May 2007 at www.umass.edu/peri.
Filrado, A. (2004), Monetary price and asset price bubbles: calibrating the mon-
etary policy tradeofs, Bank for International Settlement working paper no.
155.
Fraga, A., I. Goldfajin and A. Minella (2003), Infation targeting in emerging
market economies, National Bureau for Economic Research, working paper
no. 10019.
Hu, Y. (2004), Empirical investigations of infation targeting, Mimeo, Institute of
International Economics, Washington, DC, February.
Jacome, L. and F. Vazquez (2005), Any link between central bank independence
and infation? Evidence from Latin America and the Caribbean, International
Monetary Fund working paper no. WP/05/75.
Jha, R. (2004), Reducing poverty and inequality in India: Has liberaliz-
ation helped?, in A. Cornia (ed.), Inequality, Growth and Poverty in an Era
of Liberalization and Globalization, Oxford: Oxford University Press, pp.
287326.
Jha, R. (2005), The political economy of recent economic growth in India, in R.
Jha (ed.), Economic Growth, Economic Performance and Welfare in South Asia,
Houndmills: Palgrave Macmillan, pp. 2851.
Jha, R. and D. Rath (2003), On the endogeneity of the money multiplier in India,
in R. Jha (ed.), Indian Economic Reforms, Basingstoke: Palgrave Macmillan, pp.
5172.
Jha, R. and A. Sharma (2004), Structural breaks, unit roots, and cointegration:
a further test of the sustainability of the Indian fscal defcit, Public Finance
Review, 32 (2), 196219.
Jha, R., R. Gaiha and A. Sharma (2006), Mean consumption, poverty and
inequality in India in the sixtieth round of the national sample survey, ASARC
working paper no. 2006/11, Australian National University.
Joshi, V. and S. Sanyal (2004), Foreign infows and macroeconomic policy in
India, Mimeo, Oxford University, pp. 154.
Kelkar, V. (2004), India on the growth turnpike, Narayanan Oration, presented
at Australia South Asia Research Centre, Australian National University.
Kydland, F. and E. Prescott (1977), Rules rather than discretion: the inconsist-
ency of optimal plans, Journal of Political Economy, 87 (2), 47392.
Lal, D., S. Bery and D. Pant (2003), The real exchange rate, fscal defcits and
capital fows India: 19812000, Economic and Political Weekly, 38 (47),
496576.
Lomax, R. (2005), Infation targeting in practice models, forecasts and hunches,
speech by the Deputy Governor of the Bank of England to the 59th Atlantic
Economic Conference, 12 March.
Masson, P., M. Savastano and S. Sharma (1997), The scope for infation target-
ing in developing countries, International Monetary Fund working paper no.
97/130.
McKibbin, W. and K. Singh (2003), Issues in the choice of a monetary regime
for India, in R. Jha (ed.), Indian Economic Reforms, Basingstoke: Palgrave
Macmillan, pp. 1150.
Mohanty, M. and M. Klau (2004), Monetary policy rules in emerging economies:
issues and evidence, Bank for International Settlement working paper no.
149.
Mohanty, D., D. Rath and M. Ramaiah (2000), Measures of core infation for
India, Economic and Political Weekly 35 (5), 27382.
Obstfeld, M. and K. Rogof (1996), Foundations of International Macroeconomics,
Patel, U. and P. Srivastava (1997), Some implication of real exchange rate target-
ing in India, Mimeo, Indian Council for Research on International Economic
Relations, New Delhi, March.
RBI (Reserve Bank of India) (1993), Annual Report 199192, Mumbai: RBI.
RBI (2000) Handbook of Statistics on the Indian Economy, Mumbai: Reserve Bank
of India.
RBI (2002), Reserve Bank of India Bulletin, Mumbai: RBI.
RBI (2004), Report on Currency and Finance, Mumbai: RBI.
Samantaraya, A. and A. Prasad (2001), Growth and infation in India: detecting
the threshold level, Asian Economic Review, 43 (3), 41420.
Seyfried, W. and D. Bremmer (2003), Infation targeting as a framework for
monetary policy: a cross-country analysis, Australian Economic Review, 36(3),
2919.
Sharma, A. (2004), Fiscal defcits, fnancial crises and adjustment in a semi-open
economy, unpublished PhD dissertation, Australian National University.
Svensson, L. (1999), Infation targeting as a monetary policy rule, Journal of
Monetary Economics, 43 (3), 60754.
Taylor, J. (1993), Discretion versus policy rules in practice, CarnegieRochester
Conference Series on Public Policy, 39, 195214.
Taylor, J. (2002), The monetary transmission mechanism and the evaluation of
monetary policy rules, in N. Loayza and K. Schmidt-Hebbel (eds), Monetary
Policy: Rules and Transmission Mechanisms, Santiago: Central Bank of Chile.
White, W. (2004), Making macro prudential concerns operational, speech
delivered at Symposium on Financial Stability at De Nederlandsche Bank,
Amsterdam, March.
Winters, A., N. McCulloch and A. McKay (2002), Trade liberalisation and
poverty: the empirical evidence, Department of Economics, University of
Sussex discussion paper no. 88, Brighton.
APPENDIX 12.A1 CAPITAL CONTROLS: OR WHY
DID INDIA ESCAPE THE EAST
ASIAN CRISIS?
It is instructive to compare India and the East Asian countries in 1996
(that is, just before the East Asian crisis of 1997). The frst six columns
of Table 12.A1 indicate that, in most respects, Indias fundamentals
(fscal balance, infation, current account balance, non-performing assets,
debt-exports ratio and debt-service ratio) were worse or no better than the
crisis-countries. All these countries were on loose dollar peg and India was
only marginally diferent from the rest in this regard except for the fact
that India did not allow its real exchange rate to appreciate and was able
to maintain its real exchange targeting posture.
The critical diference between India and the crisis-countries can be seen
in the last two columns of Table 12.A1. India managed to keep short-
term debt under control, both in relation to total debt and in relation to
foreign exchange reserves and thus avoided an unstable debt structure, an
outcome that was the direct result of controls on debt-creating short-term
infows.
Table 12.A1 Various countries: indicators of crisis-vulnerability, 1996
FB/
GDP
(%)
P/P
(% p.a.)
CAB/
XGS
(%)
NPA
(%)
NCEDT/
XGS
(%)
TDS/
XGS
(%)
SDT/
EDT
(%)
SDT/
RES
(%)
India 9.0 9.0 11.7 17.3 103.6 21.2 5.3 27.1
Indonesia 1.0 8.0 13.0 8.8 180.5 36.6 25.0 166.7
Korea 0.0 4.9 14.6 4.1 82.0 9.4 49.4 192.7
Malaysia 0.7 3.5 6.4 3.9 40.4 9.0 27.9 39.7
Philippines 0.3 8.4 9.9 n.a. 80.1 13.4 19.9 67.9
Thailand 0.7 5.8 19.5 7.7 110.9 12.6 41.5 97.4
Note: FB/GDP: Fiscal balance as a proportion of GDP; P/P: Rate of consumer price
infation; CAB/XGS: Current account balance as a proportion of exports of goods and serv-
ices; NPA: Non-performing assets of commercial banks as a proportion of total advances;
NCEDT/XGS: Non-concessional external debt as a proportion of exports of goods and
services; TDS/XGS: Debt service as a proportion of exports of goods and services; SDT/
EDT: Short-term external debt as a proportion of total external debt; SDT/RES: Short-term
external debt as a proportion of foreign exchange reserves.
Sources: FB/GDP, NPA: Bank of International Settlements Annual Reports (199798 and
19992000) and Government of India, Economic Survey (19992000).
CAB/XGS, NCEDT/XGS, TDS/XGS, SDT/EDT, SDT/RES: World Bank, Global
Development Finance (1999).
P/P: IMF International Financial Statistics (2000).
India was able to resist the pressure to adopt capital account convert-
ibility essentially because of three reasons: frst, the ideology of laissez-
faire is still not dominant in India, and second, foreign banks, which are
normally a strong pressure group in favor of capital account convertibil-
ity, had a very small presence in the country. Finally, India was too big to
be bullied into adopting capital account convertibility by Wall Street, the
IMF and the US Treasury (Joshi and Sanyal, 2004).
271
13. Towards an alternative monetary
policy in the Philippines
Joseph Anthony Lim
1
13.1 INTRODUCTION
The modern macroeconomic history of the Philippines had been marked by
periodic balance of payment crises with massive devaluations that resulted
in high infation. These crises were followed by drastic monetary and fscal
recessionary programs implemented by the International Monetary Fund
(IMF), as a precondition for the release of emergency funds.
Monetary targeting was implemented in the Philippines from the 1980s
to the 1990s under IMF sponsorship. It is generally accepted as contribut-
ing to the depth of recessions in the Philippines in the last 25 years. Since it
was replaced by infation targeting only in 2002, it is dif cult not to include
in this chapter a discussion of the experience with monetary targeting.
The next section describes the monetary targeting experience of the
Philippines. Section 13.3 discusses the shift from the monetary targeting
regime to the infation targeting regime starting 2002. Section 13.4 places
monetary policy in the context of the complex macro situation and devel-
opment needs of the Philippines. The last section gives detailed recommen-
dations for an alternative monetary policy.
13.2 MONETARIST TARGETING AND POLICY
IN THE PHILIPPINES: CRITICIZING THE
DEMAND-SIDE CURE FOR INFLATION AND
CURRENT ACCOUNT DEFICITS
The economic crises which occurred from the late 1940s to the present
had been connected with balance of payment and foreign exchange crises.
These have led to some very sharp recessions especially the economic col-
lapse in 198485 that destroyed any chance of the Philippines becoming
an East Asian success story.
Carrying a strong belief that current account defcits and infation
are due to excessive aggregate demand caused by monetary expansion,
monetarisms response to balance of payment crises involved damaging
pro-cyclical monetary and fscal austerity that deepened the crisis and
recession. Starting in the early 1960s the IMF had become the standard
funder of last resort during crises. In the early 1980s quarterly monetary
targeting became the norm. These monetary targets became very tight every
time balance of payments deteriorated and infation increased. Monetary
targets were based on targets on the monetary base and achieved through:
(1) high required reserve ratio; (2) high policy rates of the central bank;
and (3) open market sale of central bank bills and government securities in
order to reduce the monetary base. It was the use of the third instrument
that was most damaging as it directly reduced liquidity and credit in the
fnancial sector.
13.2.1 Supply-Side Causes of Infation
Figure 13.1 gives a picture of infation rate based on the consumer price
index (CPI) and GDP growth rate. The graph shows that high infation
usually happens simultaneously with lower growth or recessions, rather
than during periods of high growth and high aggregate demand.
This is particularly true for the periods 198485 (economic collapse),
199091 (another recession) and 1998 (Asian crisis period). This is because
these periods were periods of signifcant currency devaluations resulting
from balance of payment crises. This brings about stagfation that explains
the high infation and recession.
In fact, the role of currency devaluations and oil price shocks in explain-
ing periods of high infation in the Philippines is very clearly illustrated in
Figure 13.2.
It can be seen that the devaluation in 1970 brought high infation in
197071. The frst oil price shock in 197374 brought about high infa-
tion in 1973 and, especially in 1974. Again the second oil price shock and
worldwide infation in 197981 brought about the high infation during the
same years. The massive devaluations in 1983 and 1984 led to high infa-
tion in the economic collapse period of 198485. The moderate devalua-
tion in 199091, plus the oil price shock due to the frst Gulf War crisis
brought about the infation in 1990 and 1991. The signifcant devaluation
during the Asian crisis in 1998 brought a slight uptick in infation, but
nowhere near the high infation that occurred in previous devaluations.
Thus, by just using two types of supply-side shocks currency depre-
ciation and oil price shock one can explain practically all the above-10
percent infation in modern Philippine history. High infation periods are
not triggered by high domestic demand but by supply-side shocks.
Towards an alternative monetary policy in the Philippines 273
If one uses monthly or quarterly data on infation, one will also see
that agricultural price shocks make their impact on price infation due to
weather disturbances leading to food shortages.
13.2.2 Monetary Contraction and Impact on GDP and Unemployment
Using monetary contraction to fght infation aggravates recessions and
unemployment, since the original supply shocks (devaluation, oil price
shocks or food shortages) already have recessionary impact. It is import-
ant to point out that the serious stagfationary periods 198485, 1991 and
1998 were all periods when the Philippines undertook severe monetary and
credit contraction.
Figure 13.3 shows the relationship of lending rate and GDP growth
rate. Lending rates rise during crises because of the higher infation, but
credit and monetary contraction goes beyond this since the authorities
purposely contract the monetary base and increase required reserves and
the central bank policy rate. As expected, one sees a strong inverse rela-
tionship between the lending rate and GDP growth rate.
10
0
10
20
30
40
50
1
9
5
0
1
9
5
3
1
9
5
6
1
9
5
9
1
9
6
2
1
9
6
5
1
9
6
8
1
9
7
1
1
9
7
4
1
9
7
7
1
9
8
0
1
9
8
3
1
9
8
6
1
9
8
9
1
9
9
2
1
9
9
5
1
9
9
8
2
0
0
1
2
0
0
4
P
e
r
c
e
n
t

(
%
)
GDP per capita growth
CPI inflation rate
Source: National Statistics Coordination Board; International Financial Statistics (IMF).
Figure 13.1 GDP per capita growth versus CPI infation rate
The lending rates in Figure 13.3 actually underestimate the costs of
borrowing since during periods of monetary contraction, many frms are
credit-rationed. Figure 13.4 shows the relationship between growth of real
money and GDP growth rate. Declines in real money are associated with
declines in GDP growth in 1960, 1964, 1974, 198485, 1991 and 1998.
Figure 13.5 shows that the recessions of 198485, 1991 and 1998 created
signifcant upticks in the unemployment rate.
13.2.3 Monetary Contraction and External Defcits
Much of the demand-suppressing monetary austerity programs were
implemented not only to reduce infation but to reduce aggregate demand
to ensure that current account defcits were also reduced during balance
of payment crises. And it is often overdone to produce current account
surpluses.
Figure 13.6 shows a graph of the current account defcit (as percentage
of GDP) and GDP per capita growth rate over the years. Concentrating
10
0
10
20
30
40
50
60
70
80
90
1
9
4
9
1
9
5
2
1
9
5
5
1
9
5
8
1
9
6
1
1
9
6
4
1
9
6
7
1
9
7
0
1
9
7
3
1
9
7
6
1
9
7
9
1
9
8
2
1
9
8
5
1
9
8
8
1
9
9
1
1
9
9
4
1
9
9
7
2
0
0
0
2
0
0
3
P
e
r
c
e
n
t

(
%
)
CPI inflation rate
Growth rate of exchange rate
Figure 13.2 CPI infation versus growth of exchange rate
on three recession periods 198485, 1991 and 1998 it is clear that every
crisis was preceded by a year or two of very signifcant current account
defcits. This is because import demand grows during boom years as
developing countries are dependent on imported raw materials and capital
goods. It is also clear that after every recession, the current account defcit
improved, and in the case of the crisis in the mid 1980s and the Asian
crisis the current account turned positive after the recession. It should
be emphasized that the improvement in the current account balance was
caused partly by the massive devaluation and partly by the recession which
had been aggravated by the monetary contraction dictated by monetarist
policy. Monetarist theory and the IMF interpret large trade defcits as
overspending and require monetary and fscal tightness. Thus, although
monetary targeting is supposed to target only one variable infation it
is also afected by foreign exchange depletion and capital outfows during
crises.
Finally the last possible reason for monetary tightening during balance
of payment crises is to stem the fight of capital and attract them back into
the country, and at the same time stave of the damaging exchange rate
10
5
0
5
10
15
20
25
30
35
1
9
7
6
1
9
7
8
1
9
8
0
1
9
8
2
1
9
8
4
1
9
8
6
1
9
8
8
1
9
9
0
1
9
9
2
1
9
9
4
1
9
9
6
1
9
9
8
2
0
0
0
2
0
0
2
2
0
0
4
2
0
0
6
P
e
r
c
e
n
t

(
%
)
Lending rate
Figure 13.3 GDP per capita growth versus lending rate
collapse. This was mainly the excuse in the 198485 debt crisis and the
1998 Asian crisis. However, in both cases it was not the high interest rates
that stopped the currency depreciation and balance of payments outfow
but mainly the devaluation-cum-recession which improved the current
account by improving the trade accounts. Figure 13.7 tracks the current
account and balance of payment balances. It can be seen in the graph
that the balance of payments improvement from 1984 to 1987 following
the 198485 collapse was mainly due to very signifcant improvements in
the current account. In the 1998 Asian crisis even the improvement of the
balance of payments in 1998 was less than the current account improve-
ments, indicating net capital outfows in a period of high interest rates. But
in subsequent years the positive balances in the current account exceeded
that of the balance of payments, which meant that there was continuing
net capital outfow in the balance of payment account from 1999 to 2002
(most likely due to the lackluster performance of the Philippine economy
30
20
10
0
10
20
30
40
50
1
9
5
2
1
9
5
5
1
9
5
8
1
9
6
1
1
9
6
4
1
9
6
7
1
9
7
0
1
9
7
3
1
9
7
6
1
9
7
9
1
9
8
2
1
9
8
5
1
9
8
8
1
9
9
1
1
9
9
4
1
9
9
7
2
0
0
0
2
0
0
3
2
0
0
6
P
e
r
c
e
n
t

(
%
)
Real money supply
growth rate
Figure 13.4 GDP growth rate versus real money supply growth
and the political troubles of the Estrada and Arroyo governments). In
the crises of 198485 and 1998 the current account went into signifcant
positive territory because of the joint efects of devaluation and output
and demand contraction. This proves that monetarist aggregate demand
suppression is successful in stopping foreign exchange outfows and
current account defcits not by stemming capital outfows and encourag-
ing return of foreign exchange due to high interest rates but, as previous
graphs prove, by creating sharp recessions and severe unemployment,
which reduce current account defcits. Infation rates are also lowered
because recessions kill consumer and investment demand.
An important lesson from the discussion above is that the roles of the
current account, the exchange rate regime and issues concerning how to
stem capital fight should be incorporated in any alternative to the current
monetary policies.
From the 1980s to 2000 the sole purpose of monetary policy was
to fght overspending that had caused current account defcits and
high infation. It ensured that monetary policy could not be used for
8
6
4
2
0
2
4
6
8
10
12
14
1
9
8
1
1
9
8
2
1
9
8
3
1
9
8
4
1
9
8
5
1
9
8
6
1
9
8
7
1
9
8
8
1
9
8
9
1
9
9
0
1
9
9
1
1
9
9
2
1
9
9
3
1
9
9
4
1
9
9
5
1
9
9
6
1
9
9
7
1
9
9
8
1
9
9
9
2
0
0
0
2
0
0
1
2
0
0
2
2
0
0
3
2
0
0
4
P
e
r
c
e
n
t

(
%
)

GDP Growth rate
Unemployment rate
Source: National Statistics Coordination Board; Labor Force Survey, National Statistics
Of ce; International Financial Statistics (IMF).
Figure 13.5 Unemployment rate versus GDP growth rate
countercyclical and growth purposes. The adoption of fnancial lib-
eralization policies starting in the 1980s also ensured that monetary
policy could not be used to fnance targeted or prioritized sectors of the
economy. This complemented the prescription of the multilateral agen-
cies for the country to abandon industrial policy and undertake trade
liberalization.
At the same time the policy is not matched by a confrmation of the
theory behind the demand theory of infation and overspending. Figure
13.8 graphs the infation rate on the y-axis and unemployment rate on the
x-axis for the years from 1959 to 2004 to fnd out any trace of a Phillips
curve. One can see there is nothing that looks like it in the graph in any
sub-period. This is because infation is caused by supply shocks and not
demand-led. (The graph looks like it is producing natural rates of infa-
tion rather than natural rates of unemployment.)
10
8
6
4
2
0
2
4
6
1
9
7
7
1
9
7
9
1
9
8
1
1
9
8
3
1
9
8
5
1
9
8
7
1
9
8
9
1
9
9
1
1
9
9
3
1
9
9
5
1
9
9
7
1
9
9
9
2
0
0
1
2
0
0
3
2
0
0
5
P
e
r
c
e
n
t

(
%
)
Current account balance, % of GDP
Figure 13.6 Current account balance as percentage of GDP versus GDP
per capita growth
13.3 FROM MONETARY TARGETING TO
INFLATION TARGETING
13.3.1 A Switch in Regime: From Monetary Targeting to Infation
Targeting
In the Philippines research work in the central bank itself pointed to the
overkill created by monetarist policy in trying to tame infation. This is
mainly because the quantity theory of money equation assumes a unitary
coef cient of money to prices with output holding steady, despite massive
monetary contraction. The use of a single estimated infation equation
assumes all explanatory variables (exchange rate movements, output
growth, and so on) to be exogenous and that only the monetary variable
will decrease infation. A study by a staf member of the central bank
12
10
8
6
4
2
0
2
4
6
8
1
9
7
7
1
9
7
9
1
9
8
1
1
9
8
3
1
9
8
5
1
9
8
7
1
9
8
9
1
9
9
1
1
9
9
3
1
9
9
5
1
9
9
7
1
9
9
9
2
0
0
1
2
0
0
3
2
0
0
5
P
e
r
c
e
n
t

(
%
)
Current account balance, % of GDP
BOP, % of GDP
Figure 13.7 Current account and balance of payments, percentage of GDP
(Dakila, 2001) shows that with a simultaneous equation system where
exchange rate movements and output movements are endogenized, the
degree of monetary tightening required to achieve a stipulated reduction
of infation is lessened. This is because monetary tightening cum recession-
ary policies tends to eventually cause some appreciation of the currency
and a fall in output growth (both of which tend to be defationary).
When single-digit infation rates were achieved in the mid 1990s under
the Ramos administration, the central bank relaxed monetary targeting to
what was called the modifed monetary targeting policy. This allowed the
monetary targets to be exceeded as long as infation targets were being met.
This was the start of a hybrid system of infation and monetary targeting.
The of cial reason given was that, because of fnancial liberalization, the
link between quantitative monetary targets and infation had weakened
due to structural breaks in the income velocity of money and volatilities
and instabilities in the money multiplier. Thus the high liquidity and large
monetary expansion in the mid 1990s failed to have any impact on infa-
tion as it remained below double-digit rates and continued to decline until
the Asian crisis (see Guinigundo, 2005).
The laxer monetary policy in the mid 1990s of course was shattered by
the contagion efects of massive depreciation pressures during the Asian
crisis. The major ways used to stave of the depreciation (and indirectly
1
9
19
29
39
49
59
8.00 8.50 9.00 9.50 10.00 10.50 11.00 11.50 12.00
Unemployment rate
I
n
f
l
a
t
i
o
n

r
a
t
e
Source: Labor Force Survey, National Statistics Of ce; International Financial Statistics
(IMF).
Figure 13.8 Infation versus unemployment rate
avoid infationary pressures) were constant and periodic raising of the
reverse repurchase rate and the raising of liquidity reserve ratios.
2
However, the massive currency depreciation (around 40 percent) during
the Asian crisis had a very low pass-through to infation (see Figure 13.2).
It did not even lead to (and only approached) double-digit infation. This
easily allowed infation rates to fall after the crisis. Infation rates reached
a very low 3 percent in 2002 and 2003, replicating the infation rates in the
developed world.
By January 2000, monetary authorities were already studying the
alleged successes of developed and developing countries that had adopted
infation targeting (from a prior regime of exchange rate pegging or
monetary targeting). The decision was to make the big switch to infa-
tion targeting starting January 2002. The switch entailed the following
el ements: (1) continuation of the announcement of infation targets based
on a band over a two-year time period; (2) adoption of a more passive
monetary quantitative policy, and employing more the repurchase and
reverse repurchase rates (policy rates) as the monetary instrument; (3)
increased sophistication in infation rate estimation and in using single and
multi-equation models to forecast infation and setting of infation targets;
(4) the use of forward-looking models with monetary instruments reacting
to and aiming to infuence infationary expectations rather than actual
infation; (5) creation of an advisory committee to recommend monetary
policies based on the new infation targeting regime; and (6) the issuance
of a quarterly infation report explaining the central banks policies and
achievement or non-achievement of the infation targets.
But the most important elements are: (7) the stipulation of escape
clauses that exempts the central bank from achieving the infation targets.
This means that if the infation target is not achieved, the central bank
can opt to not do anything if the reasons are: (a) sudden changes in prices
of agricultural commodities and products; (b) natural calamities or cata-
strophic events; (c) volatility in oil prices; and (d) sudden changes in gov-
ernment policies, such as tax structures.
Another important element is: (8) the setting up of a core infation
rate, as opposed to the overall CPI or headline infation rate. The core
takes out oil and agricultural products whose prices are easily afected by
external shocks and weather disturbances. This is an important element
since even if overall headline infation rate is not achieved, as long as the
core infation rate is within the infation target, the central bank can also
opt not to do anything.
A central bank report (Guingundo, 2005) mentions two important
things: (1) the current infation targeting method allows for ample room
for judgment and discretion of policy makers; and (2) in explaining why
overshooting of the infation target in the fourth quarter of 2004 did not
lead to increases in policy rates by the central bank, the publication gives
the following explanation: (a) the infationary pressures caused by supply-
side shocks (oil price increase) were not susceptible to monetary action,
since the latter would work on the demand side; (b) the central bank fore-
casts indicated that the pressures would subside in 2006; and (c) there were
downside risks to the overall strength of economic activity. The last point
is extremely important as it proves that current central bank policy does
put weight on the strength or weakness of economic activity.
13.3.2 A More Benign Policy?
So far infation targeting in the Philippines from 2002 to 2006 has not led
to very drastic monetary tightening unlike monetary targeting.
The improvement in infation targeting over monetary targeting can be
due to a number of factors. First, the nature of the policy tools makes infa-
tion targeting more benign than monetary targeting. To illustrate, main-
taining a monetary base target will entail credit tightening since it does
not allow money supply and domestic credit to grow with the economy
and with infation. On the other hand, maintaining the policy rate of the
central bank (that is, keeping the reverse repurchase rate constant within
reasonable limits) does not entail credit tightening and allows money
supply and domestic credit to grow moderately. Thus to make monetary
targeting and infation targeting equally restrictive will entail increasing
policy rates drastically, which is much more obvious and prone to public
criticism compared to reducing the monetary base target (which had not
been transparent and kept in IMF memorandums). The Philippines, in its
infation targeting history (2002 to present), has not increased the policy
rate drastically and so infation targeting policy in the Philippines has not
mimicked monetary targeting by drastically cutting money supply and
credit.
More importantly, the adoption of infation targeting was not done as a
precondition to IMF conditionality but was a conscious switch decided by
the central bank. Thus, using policy rates as the key instrument allows the
central bank to implicitly inject growth objectives in deciding what rates to
maintain, even if on paper the main goal of infation targeting is infation
reduction. Second, the escape clauses and the use of core infation allow
the Philippines to use similar instruments as the US Fed and to decide on
policy rate changes based on the strength of the economy and not only on
infation targets. The policy has been quite lax since world and domestic
infation has not been very high between 2002 and 2006.
Figure 13.9 shows that since 2003, the core infation has always been
below the headline infation. Infation targets of the central bank were
exceeded in periods when oil prices were rising in the world market (2004
05) when infation rates went up beyond 6 percent and 7 percent. But
policy rates were raised only infrequently (the last one in October 2005)
since infation targeting was implemented in 2002. The justifcations were:
(1) core infation was below the headline infation; and (2) overshooting of
the infation target was due to the oil price shock a supply rather than
a demand factor. Between 2002 and 2006, the central bank raised policy
rates by only 75 basis points (due to the higher infation led by rising oil
prices in 2005 and 2006) compared to the more than 400 basis points done
by the US Fed in the same period. Of course, the Philippine policy rate
started at a higher level of 9 percent compared to the US Fed rate of 1
percent.
13.3.3 Alternate Tightening and Loosening in 2007
2007 saw an initial monetary tightening but ultimate loosening of mon-
etary policy. In the frst half of the year the government mopped up liquid-
ity by allowing banks and their customers to park their money in special
deposit accounts with the central bank, receiving interest rates equivalent
to the repurchase rate. The of cial reason is too much liquidity due to
massive infows of remittances of overseas workers. What is left unsaid,
however, is that the real danger of the high liquidity is that, coupled with
little credit going to private businesses, the high liquidity could possibly
lead to the creation of bubbles in the fast rising equities and real property
markets.
But when infation was seen to be hovering around 3 percent and below
0
2
4
6
8
10
Q
u
a
r
t
e
r
l
y

a
v
e
r
a
g
e

i
n

p
e
r
c
e
n
t
(
2
0
0
0

=

1
0
0
)
12
1995 2006 1998 1999 2000 2001 2002 1997 1996 2005 2004 2003
Core inflation
Headline inflation
Source: Bangko Sentral ng Pilipinas (BSP).
Figure 13.9 Headline and core infation rates: frst quarter 1995 to frst
quarter 2006
(way below the infation target of 4 percent to 5 percent for 2007), the
central bank reduced their policy rates by as much as 200 points in the frst
11 months of 2007. An additional strong reason to do this is the alarm-
ing appreciation of the peso due to the massive infow of remittances and
portfolio infows.
In 2008 and 2009 when the global fnancial crisis erupted, the peso started
to depreciate due to strong capital outfows. In 2008, when food and fuel
infation brought CPI infation rate in the Philippines to the teens (see
Figure 13.9), the Bangko Sentral ng Pilipinas (BSP) increased policy rates
by 100 basis points in the period from June 2008 to August 2008. When
food and fuel infation waned in the fourth quarter of 2008 to the present
period (March 2009), the BSP reduced policy rates by 125 basis points
between December 2008 to March 2009. The current overnight deposit
rate and overnight lending rates of 4.75 percent and 6.75 percent respec-
tively, are the lowest since infation targeting started in January 2002.
13.4 THE NEED FOR A HOLISTIC VIEW AND AN
ALTERNATIVE MONETARY POLICY
Even with the laxer monetary policy, this has not led to adequate credit
expansion for investment needs, better employment prospects and a more
stable external fnancial sector. It is clear that the fnancial, monetary,
fscal, external and real sectors are integrally linked and afect one another
critically, and one cannot expect monetary policy alone to solve the macro
problems.
13.4.1 The Recent Fiscal Crisis
From the mainstream point of view, the biggest challenge to the current
Philippine macroeconomy after the Asian crisis is the large fscal defcits
and public debt burden in the period 2002 to 2006. This problem is resur-
facing in the global recession years of 2009 and beyond as tax collection
is expected to fall with the economic slowdown and government spending
will have to increase to pump prime the economy.
Fiscal surplus had been achieved by the Ramos administration before
the Asian crisis. The Asian crisis, however, brought back fscal defcits.
Table 13.1 shows fscal defcits worsening after the crisis, with the national
government defcit reaching more than 5 percent of GDP in 2002. Public
sector borrowing requirements reached more than 6 percent in the same
year as the defcit of government corporations, especially that of the
National Power Corporation, became larger.
285
T
a
b
l
e

1
3
.
1

F
i
s
c
a
l

a
n
d

p
u
b
l
i
c

f

n
a
n
c
e

f

g
u
r
e
s

f
o
r

t
h
e

P
h
i
l
i
p
p
i
n
e
s
I
t
e
m
1
9
9
8
1
9
9
9
2
0
0
0
2
0
0
1
2
0
0
2
2
0
0
3
2
0
0
4
2
0
0
5
2
0
0
6
N
a
t
i
o
n
a
l

G
o
v
t

R
e
v
e
n
u
e
s

(
P

B
i
l
l
i
o
n
)
4
6
2
.
5
4
7
8
.
5
5
1
4
.
8
5
6
7
.
5
5
7
8
.
4
6
3
9
.
7
7
0
6
.
7
8
1
6
.
2
9
7
9
.
6
A
s

%

o
f

G
D
P
:
N
a
t
i
o
n
a
l

G
o
v
e
r
n
m
e
n
t

(
N
G
)

R
e
v
e
n
u
e
s

1
7
.
4
1
6
.
1
1
5
.
3
1
5
.
6
1
4
.
6
1
4
.
8
1
4
.
5
1
5
.
1
1
6
.
3

o
f

w
h
i
c
h
:

T
a
x

1
5
.
6
1
4
.
5
1
3
.
7
1
3
.
6
1
2
.
8
1
2
.
8
1
2
.
5
1
3
.
0
1
4
.
3
N
a
t
i
o
n
a
l

G
o
v
e
r
n
m
e
n
t

E
x
p
e
n
d
i
t
u
r
e
s

1
9
.
3
1
9
.
9
1
9
.
3
1
9
.
7
1
9
.
9
1
9
.
5
1
8
.
4
1
7
.
8
1
7
.
4
N
a
t
i
o
n
a
l

G
o
v
e
r
n
m
e
n
t

S
u
r
p
l
u
s
/
D
e
f

c
i
t

(
1
.
9
3
.
8
4
.
0
4
.
0
5
.
3
4
.
6
3
.
8
2
.
7
1
.
1
P
u
b
l
i
c

S
e
c
t
o
r

B
o
r
r
o
w
i
n
g

R
e
q
u
i
r
e
m
e
n
t
s
4
.
2
4
.
6
5
.
2
5
.
2
6
.
8
6
.
4
5
.
8
3
.
4
n
.
a
.
A
s

%

o
f

N
G

R
e
v
e
n
u
e
s
:
N
G

D
e
b
t

S
e
r
v
i
c
e

P
a
y
m
e
n
t
s
3
5
.
6
4
2
.
9
4
4
.
3
4
8
.
4
6
1
.
9
7
3
.
5
8
5
.
1
8
3
.
2
8
7
.
2
I
n
t
e
r
e
s
t
2
1
.
6
2
2
.
2
2
7
.
4
3
0
.
8
3
2
.
1
3
5
.
4
3
6
.
9
3
6
.
7
3
1
.
7
P
r
i
n
c
i
p
a
l
1
4
.
0
2
0
.
7
1
6
.
9
1
7
.
6
2
9
.
8
3
8
.
1
4
8
.
2
4
6
.
5
5
5
.
6
A
s

%

o
f

G
D
P
:
A
s

o
f

J
u
l
y

2
0
0
6
T
o
t
a
l

N
a
t
i
o
n
a
l

G
o
v
e
r
n
m
e
n
t

D
e
b
t
5
6
.
1
5
9
.
6
6
4
.
6
6
5
.
7
7
1
.
0
7
8
.
2
7
9
.
0
7
2
.
3
6
9
.
6
D
o
m
e
s
t
i
c
3
1
.
9
3
2
.
9
3
1
.
8
3
4
.
4
3
7
.
1
3
9
.
7
4
1
.
5
4
0
.
2
3
8
.
2
F
o
r
e
i
g
n
2
4
.
2
2
6
.
8
3
2
.
7
3
1
.
3
3
3
.
9
3
8
.
5
3
7
.
5
3
2
.
0
3
1
.
4
A
s

%

o
f

G
D
P
:
T
o
t
a
l

P
u
b
l
i
c

S
e
c
t
o
r

D
e
b
t
9
4
.
6
1
0
1
.
5
1
0
8
.
0
1
0
6
.
0
1
1
0
.
2
1
1
8
.
2
1
0
9
.
8
9
3
.
4
n
.
a
.
D
o
m
e
s
t
i
c
3
5
.
2
3
2
.
8
3
2
.
1
3
2
.
7
3
4
.
4
3
5
.
7
3
5
.
4
3
3
.
0
n
.
a
.
F
o
r
e
i
g
n
5
9
.
5
6
8
.
7
7
5
.
9
7
3
.
3
7
5
.
9
8
2
.
5
7
4
.
4
6
0
.
4
n
.
a
.
S
o
u
r
c
e
:

B
a
n
g
k
o

S
e
n
t
r
a
l

n
g

P
i
l
i
p
i
n
a
s

(
B
S
P
)
;

B
u
r
e
a
u

o
f

T
r
e
a
s
u
r
y
,

N
a
t
i
o
n
a
l

E
c
o
n
o
m
i
c

a
n
d

D
e
v
e
l
o
p
m
e
n
t

A
u
t
h
o
r
i
t
y
.
The gravity of the fscal problem is due to the fact that the post-Asian
crisis economic recovery from 1999 to 2005 had failed to improve tax
efort. The tax efort peaked at more than 17 percent of GDP in 1996 and
1997 and consistently fell after that and bottomed at 12.5 percent in 2004
(Table 13.1). The falling tax efort despite signifcant GDP growth forced
the government to undertake substantial tax reforms to respond to down-
grades by the rating agencies of the sovereign debt. The tax efort improved
to around 14.3 percent in 2006 mainly due to the expanded coverage and
higher value-added taxation (raised from 10 percent to 12 percent).
The falling tax efort and rising public debt burden after the Asian crisis
brought about a serious situation wherein public debt service principal
and interest debt payments made up 85 percent of government revenues
in 2005 and 87 percent in 2006 (Table 13.1).
3
The improving fscal defcits
from 2003 to 2006 were brought about only because total non-debt expen-
ditures of the government (as percentages of GDP) especially social and
economic services were cut drastically as interest payments increased.
Table 13.1 shows total national government expenditures falling to 17.4
percent of GDP in 2006 from more than 19 percent in 1998. Figure 13.10
shows the gravity of the situation as economic and social services as
percentages of GDP continuously fell in recent years while the share of
interest payments went up.
0
1
2
3
4
5
6
7
1
9
9
6
1
9
9
7
1
9
9
8
1
9
9
9
2
0
0
0
2
0
0
1
2
0
0
2
2
0
0
3
2
0
0
4
2
0
0
5
2
0
0
6
P
e
r
c
e
n
t
Social services
Defense
Net lending
Interest payments
Economic services
General public
services
Source: Department of Budget and Management (DBM).
Figure 13.10 National government expenditures, percentage of GDP
Despite improvements in the tax efort in 2006, the high public debt
burden is expected to continue for some time as Table 13.1 indicates.
Public investment (and other economic and social spending) has yet to
recover from cutbacks, while private investments remain low despite high
economic growth. In the frst quarter of 2007, despite a whopping 6.9
percent GDP growth, the tax efort fell to 12.2 percent (from 13 percent in
the same period of 2006). The inability of revenue generation to keep pace
with economic growth does not bode well for public investment and the
infrastructure needs of the country, which have been neglected in recent
years due to fscal tightness. This is most relevant in 2009 when pump-
priming is vital to prevent the economy from being adversely afected by
the global economic recession, which already has reduced exports by 40
percent to 50 percent in the frst two months of 2009.
During the period of high issuance of government securities to fnance
the defcits in 2002 to 2006, there was no signifcant rise in the treasury bill
rates since the fnancial institutions teaming with liquidity preferred
government securities to private loan assets (after having been burnt in
the Asian crisis). This enabled the Treasury to issue treasury bonds at even
lower rates.
13.4.2 More Open Capital Accounts and More Volatile Exchange Rate
Movements
Figure 13.2 shows that since the 1980s the exchange rate has been very
volatile. Obviously the exposure to fnancial and capital account liberal-
ization has created a Pandoras box of dangerous short-term capital fows
and volatile movements.
Even the years following the Asian crisis from 1999 to 2005 have
been periods of volatile short-term portfolio fows with international
capital shunning the country due to political instability and fscal prob-
lems. Starting 2005, short-term capital came into the country due to
the increases in VAT and expectations of the narrowing of the fscal
defcits. But they intermittently fowed out also due to political instability,
increases in the US interest rates in 2005 and early 2006, and ultimately the
sub-prime fasco in the US that brought stock markets tumbling down for
high-yielding emerging markets like the Philippines.
In late 2006 and throughout 2007 the strong appreciation of the peso
due to remittances of overseas workers, short-term capital infows and a
generally weak dollar internationally is causing major concerns among
exporters, overseas workers and domestic manufacturing competing with
imports. On the other hand, other domestically oriented sectors are happy
as infationary pressures are stemmed and economic growth is stimulated.
Trade defcits may be expected to deteriorate (although this has not hap-
pened very signifcantly yet). But the latest picture of the Philippines has
changed from one of high current account defcits during periods of high
growth (see Figure 13.6) to consistently positive current account balances
(despite trade defcits) because of the massive infux of foreign exchange
earnings of the overseas workers.
13.4.3 Low Financial Confdence and Credit to the Private Sector
The post-Asian crisis period has been marked by low fnancial confdence
due to the trauma of non-performing assets suddenly increasing and due
to stringent fnancial supervision to achieve higher capital adequacy ratios
and loan loss provisions. Figure 13.11 shows M2 (money plus quasi-
money) and domestic credit as percentages of GDP.
It is clear that the Philippines has not achieved substantial fnancial deep-
ening as M2 and domestic credit fell drastically from its 1997 peak. The lack
of fnancial deepening is partly caused by the various recession and monetary
0.000
10.000
20.000
30.000
40.000
50.000
60.000
70.000
80.000
90.000
1
9
4
8
1
9
5
1
1
9
5
4
1
9
5
7
1
9
6
0
1
9
6
3
1
9
6
6
1
9
6
9
1
9
7
2
1
9
7
5
1
9
7
8
1
9
8
1
1
9
8
4
1
9
8
7
1
9
9
0
1
9
9
3
1
9
9
6
1
9
9
9
2
0
0
2
2
0
0
5
P
e
r
c
e
n
t
Domestic credit, % of GDP
M2, % of GDP
Source: International Financial Statistics (IMF).
Figure 13.11 M2 and domestic credit, percentage of GDP
tightening periods. Another big reason is the decline in fnancial confdence
due to the fnancial crises. This is very clear in the decline of domestic credit
(as percent of GDP) from the mid 1980s until 1992 due to the fnancial and
economic collapse of 198485. It is happening again in the post-Asian crisis
period as both domestic credit and M2 as percentages of GDP declined
from 1998 until 2004. But the picture of the Philippines in 2000 to 2008
changed from one of high current account defcits during periods of high
growth (see Figure 13.13) to positive current account balances (despite
trade defcits) because of the massive infux of foreign exchange earnings of
the overseas workers. The positive current account balance is being threat-
ened in 2009 and beyond as the global economic recession has signifcantly
reduced exports and is threatening the remittances of the overseas Filipinos
workers who are at risk of losing their jobs.
The Asian crisis increased fnancial regulations on banks. The Basel
international standard for minimum capital adequacy (net worth to risk
asset) ratios of banks is currently at 8 percent. The Philippines has been
more stringent and imposed a minimum capital adequacy ratio of 10
percent starting in 2001. The actual current average capital adequacy ratio
for Philippine banks ran to a high 15 percent to 16 percent in 2005. This
situation, together with the perennial political crisis, has so far discour-
aged banks from aggressively lending to the private sector. The banking
system is awash with liquidity with a strong appetite for government
securities whether peso or dollar denominated rather than private
lending. This will constrain investments and employment generation in an
economy with still underdeveloped long-term capital markets.
13.4.4 Persistently High Unemployment and Low Investment Rates
Figure 13.4 shows persistently high unemployment in the latest economic
recovery period of 200004 despite positive economic growth. Figure
13.12 gives more detail on the employment picture. It shows that a major
trend in the employment picture is the downward employment absorption
capacity of agriculture and the low and stagnant employment absorption
capacity of industry. The only sector adequately absorbing the growing
labor force is the service sector.
It must be pointed out that the industrial and agricultural sectors are
the main tradable sectors. With increased trade liberalization, globaliz-
ation and competition among countries, these sectors are now exhibiting
increasing output-employment ratios as output increases are not matched
by equivalent employment increases. This means labor shedding and
labor-cost cutting in areas whose products are facing stif competition from
imports. Thus, services, which is largely a non-tradable sector, becomes
the biggest absorber of employment. But this is not enough to absorb the
expelled labor from the tradable sector, and the new labor force entrants
(see Lim and Bautista, 2006). Of course the result is that unemployment
remains persistently high, hovering at 10 percent and 11 percent.
This situation leads us to explore whether a more employment-sensitive
monetary policy can help alleviate unemployment and underemployment
through an integrated scheme of credit allocation to labor-intensive and
employment-generating activities.
Related to this is the fact that investment rates have fallen since the
Asian crisis (as is true for most of the East Asian economies). Investments
as a share of GDP fell from above 24 percent before the Asian crisis in
1997 to 21 percent in 2000 and to 14.8 percent in 2006. The steady and
continuous fall of the investment rate, partly due to the collapse in public
and infrastructure investments due to the fscal bind, does not bode well
for future supply capacity and productivity improvements in the economy,
even as the GDP growth rate has improved considerably in 2004 to 2007.
This is another area where monetary policy may play a more active role.
A lax and accommodating monetary policy is even more urgent in the
global recession period of 2009 and beyond when exports are plunging and
0
5
10
15
20
25
30
35
40
45
1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004
P
e
r
c
e
n
t
Services
Industry
Unemployed
Agriculture
Source: Labor Force Survey, National Statistics Of ce.
Figure 13.12 Employment by sector and unemployed, as percentage of
labor force
consumption and investments are being threatened by possible declines in
remittances from overseas Filipinos and losses in domestic business con-
fdence because of losses in the stock market, depreciating currency and
adverse global conditions.
13.4.5 External Shocks and High Oil Prices
The volatile world oil price movements starting in the second half of 2004
has increased the average infation rate in the Philippines from a low of 3
percent in 2002 and 2003 to more than double in 2005 and 2006. The con-
tinuing threat of higher oil prices may lead to tighter monetary policies at
home and abroad.
Thus, a policy to respond to increasing world oil prices and return to
higher world infation and interest rate regimes is now an urgent task.
If this supply-led infation leads to a policy of fghting infation through
demand-suppressing methods of increasing interest rates, recessionary
pressures may again become a major monetary policy.
13.5 COMPONENTS OF AN ALTERNATIVE
MONETARY AND FINANCIAL POLICIES
13.5.1 An Undervalued Currency and Tax-Based Capital Controls on
Infows
The frst component of the recommended alternative scheme is a reason-
ably pegged exchange rate regime, targeted at an undervalued level during
good and healthy periods of global trade, and at a slightly overvalued
level during bad times of global trade contraction. (This recommendation
is consistent with the policy prescriptions of Frenkel and Taylor, Chapter
2, this volume.)
The peso has depreciated signifcantly in 2008 and 2009 due to the
massive capital outfows from the stock market and the signifcant fall in
exports during the current global economic recession. A weak currency is
not healthy during collapsing global trade and shrinking export markets
for this brings about stagfationary tendencies on the domestic economy
without any positive efects from the export market. This puts the central
bank in a dilemma as a lax monetary policy may aggravate the peso depre-
ciation. This is true because of a fexible exchange rate regime coupled with
complete capital account liberalization.
A healthy economy requires an exchange rate that does not appreciate
too much during good times when the economy and world economy is
strong. It also requires an exchange rate that does not go on a freefall col-
lapse during the bad times. This requires capital controls and a reversal on
the capital account liberalization that most emerging markets had under-
taken in the 1990s. It would therefore be practical to propose a tax-based
(or market-based) capital controls on infows like Chile and Malaysia,
or quantitative restrictions on entry and exit of hot money to and from
the short-term equity and bond markets of the country, like China and
India. Because of the dif culty in undertaking this (note the recent unsuc-
cessful Thai attempt to tax hot money), it is suggested that a regional
ASEAN13 efort to undertake a common policy on capital controls be
explored.
13.5.2 Incorporate Output and Employment Targets to the Current
Infation Targeting Regime
If one envisions a transition away from the infation targeting regime,
it would be wise to recommend that output and employment goals be
explicitly stated as part of the objectives of monetary policy. After all,
not adversely afecting economic activity has explicitly been cited by the
central bank as one of the reasons why policy rates were not increased
drastically despite infationary pressures from world oil prices in 2005 and
2006.
13.5.3 A More Active Role in Stimulating the Economy
The above monetary policy can have some benefcial impact, especially
on the current problems of high fscal defcits, lack of fnancial conf-
dence and unemployment. The more accommodating monetary policy
may be complementary to the moves to improve and increase fnancial
loans in the post-Asian crisis period, and to ofset the natural conserva-
tive tendencies in credit expansion due to higher capital adequacy ratios
and loan-loss provisions. It is crucial in the current economic slowdown
brought about by the global economic recession. Furthermore, this will
create a better atmosphere for involving credit allocation in employment
generating activities to be discussed in the next section. Finally, since oil
prices have returned to normal levels and world interest rates are low in
the current period of global economic recession, the more accommodating
policy may allow some room for monetizing the fscal defcits inasmuch
as the pass-through of monetary increases to infation has been accepted
as weak and unstable. Fiscal expansionary policies especially in social,
economic and infrastructure spending are vital in returning the system
to quality and employment-generating growth. And due to the high debt-
to-GDP and debt service ratios of the Philippines, it is dif cult to support
fscal expansion with higher public debts. Monetizing the fscal defcit will
be dif cult to implement practically since the current authorities and the
IMF have succeeded in institutionalizing the non-monetization of fscal
defcits. Strong political will from the national government and monetary
authorities will again be required.
13.5.4 Credit Allocation to Stimulate Investments in High Value-Added
and Employment Generating Sectors
Inasmuch as unemployment is a major problem in the post-Asian crisis
period and investment rates are low particularly in sectors that may
have positive externalities for the economy and inasmuch as the fnan-
cial sector is reluctant to use market mechanisms to lend to the private
sector, it is worth considering whether targeted credit programs can help
in alleviating the unemployment and low investment problem. However,
the Philippines since the 1980s has moved away from subsidized and
targeted credit schemes as the fnancial liberalization school has pre-
dicted distortionary and adverse efects from such policies, and because
the bad experiences with the Marcos administration has convinced
many economists and technocrats that subsidized and targeted credit to
potential cronies is detrimental. Relying more on the fnancial markets
and the private sector, they fgure, is more benefcial than government
interventions. Thus, by September 2002, the central bank rediscount
window has been liberalized to allow a generalized and uniform access
to the facility by all sectors of the economy at market rates. The use of
the facility has been reoriented for money supply management (comple-
menting open market operations) instead of selective credit allocation
(such as to exports and small-scale industries) and development fnanc-
ing (Guinigundo, 2005).
There are some targeted credit schemes outside the scope of the central
bank administered by the Land Bank of the Philippines (LBP) and
Development Bank of the Philippines (DBP) targeted at agricultural
cooperatives, farmers groups and small-scale industries with funds from
the Department of Agriculture (DA), Department of Agrarian Reform
(DAR), other agencies and multilateral organizations. The Department of
Trade and Industry (DTI) also has some credit lines targeted to small- and
medium sized enterprises (SMEs). The current system has become rather
schizophrenic as the formal system and the big government agencies such
as the National Economic Development Authority (NEDA) and the
central bank promote fnancial liberalization and reduction of targeted
credit in the formal sector. But the Arroyo government currently promotes
microfnance lending by government-supported agencies as one of its key
anti-poverty strategies.
Although there are some striking successes by some rural banks, coopera-
tive banks and other microfnance units in providing credit to develop local
economies and some economic sectors, this is not being mainstreamed.
The targeting of SMEs and micro-enterprises, largely in the informal
sector, for microfnance has positive aspects inasmuch as these entities
have high employment generating potentials. However, the provision of
microfnance has oftentimes been politicized and lacks a holistic approach
of providing the SMEs and micro-enterprises access to markets and link-
ages in the formal sector, access to technology and good management
practices, skills development and technical assistance for product develop-
ment. This, unfortunately, is consistent with the overall trade and industrial
policy that the government has been following for several decades, which
prohibits picking winners and frowns at promoting priority economic
sectors (closely subscribing to the policies of the Washington Consensus
view and World Trade Organization (WTO) rules). Thus credit allocation
in this setting becomes directionless and limited to providing small liveli-
hood programs for some targeted poor areas rather than to permanent,
productive and growing industries and employment for a vibrant economy
to beneft a wider pool of poor and low-income families.
It is therefore recommended that a targeted credit allocation program
be set up by the central bank and related institutions (such as key
state banks) to give prioritized credit to key sectors that satisfy the
following:
1. Exhibit strong potential for successful take-of, viability and high
repayment but may require lumpy investments and/or sufer the frst
mover
4
problem.
2. Have strong employment generating efects, and/or strong interlink-
ages with the other sectors of the economy, which will lead to multi-
plier efects in the economy.
3. Have technology or knowledge spillover efects, as well as economies
of scale.
4. Are part of an integrated set of industries that sufer coordination
failure problems.
Market failures, endogenous growth and strategic trade theories are
now mainstream economic theories that justify the above interventions. It
is only the strong resistance of institutions that support the Washington
Consensus that prevent developing countries from applying the policy
implications of the new economic theories.
13.5.5 Heterodox Policies of Income Policies, Price Controls and Price
Stabilization
Since most of the infationary pressures are from supply-side shocks, other
means may be more productive than an immediate monetary demand-
reduction response.
In many of the weather and natural calamity shocks the government has,
for practical reasons and with some success, gone into temporary price con-
trols and constant monitoring of basic foodstufs and imposing heavy pen-
alties on hoarding. Importation of agricultural products is also undertaken
during periods of agricultural shortages. These policies should be continued
and enhanced. Improvements are especially needed in the area of equitable,
transparent and ef cient distribution of imported foodstufs during periods
of agricultural shortages. The inef cient distribution system and lack of
transparency in many of the sales of the National Food Authority (NFA)
during periods of food and agricultural shortages call for an overhaul in the
system and more participation of the private sector in the distribution of
temporarily imported foodstufs during periods of shortages.
The oil price shocks had led many in the Philippines to question the
deregulated structure of the oil industry. There is a perception that there is
an asymmetry of quick price increases during times when world prices are
rising, but slow and lagging price decreases when world prices are falling.
The question of a domestic oil cartel has arisen (led by Shell, Caltex and a
previously government-owned oil company, Petron), as many new players
are fnding it hard to compete in a setting of volatile world prices. Regulatory
boards still regulate electricity charges and transportation fares, while the
Department of Energy has clout to stop unreasonable price increases in oil
and gasoline products. Thus, the importance of regulation in public utilities
and oil/gasoline products become crucial in fostering competition and stop-
ping cartel-like pricing. The Philippines also lacks anti-trust legislation that
will remove monopoly and predatory pricing in key economic sectors.
If world oil prices increase signifcantly again, it is recommended that
an oil price stabilization fund (used moderately successfully in the 1970s
by the Marcos government) be explored to tackle the situation. Equally
important is the need to develop, in the medium and long run, alternative
fuel sources and to re-enact laws giving special tax incentives for frms
providing these alternative fuel sources.
The Philippines has an incomes policy that deliberately keep wages low
and lagging behind price increases even as their monitoring and imple-
mentation are grossly inadequate. With labor productivity increasing,
this has led to lower infation but has also led to a long-run decline in real
wages, which may have contributed to worsening income distribution and
poverty. A more balanced approach that this chapter recommends involves
the following: (1) depending more on tripartite agreements derived by gov-
ernment, labor and employers to impose agreed-upon ceilings on price and
wage increases; (2) enact laws and rules that punish monopoly and cartel-
like pricing (especially for the oil and telecommunications industries); and
(3) do more monitoring and temporary regulation of key products, such as
food, oil, telecommunications and rent-control housing, especially during
periods of price instabilities.
The above set of alternative monetary policies points to a need for a
change in mindset from a dichotomy between the fnancial and real sectors
to a viewpoint wherein the fnancial sector is integrally linked and sup-
portive of the real sector. The strategy is to coordinate and link the current
targeted credit programs in the anti-poverty strategy campaign with the
initiatives of the economic departments and agencies in developing and
promoting key priority economic sectors that have high value-added, high
technology spillover, multiplier efects and employment generating poten-
tials. Successes in this arena will hopefully spill over to the formal and big
business sector of the economy.
13.5.6 Testing the New Alternative
Inasmuch as the central bank already has a long-term and a short-term
macroeconometric model, it is suggested that the new alternative be tested
employing similar types of macro model. The new econometric model
poses some challenges: (1) How to model a more discretionary monetary
policy based on multiple objectives of both output/employment gener-
ation and price stability (various simulation scenarios and quadratic loss
functions might be used). (2) How to defne and model overheating and
excessive aggregate demand in the model to identify when some mon-
etary contraction or aggregate demand reduction may be benefcial to the
economy and to identify when monetary contraction will be unreasonably
recessionary. (3) How to model the incomes policies and temporary price
controls. (4) How to model increased credit allocation and incorporating
its impact on the growth of key economic sectors with high value-added
and employment generation. No doubt diferent scenario simulations are
needed (with low, medium and high scenario assumptions) to make the
models more useful to policy makers.
13.5.7 A Dif cult Endeavor
The above alternative set of policies runs counter to the current macro policies
of the Philippine government and go beyond policies traditionally reserved
for the central bank. Thus the adoption of the alternative monetary policy is
a formidable endeavor that requires a change, not only in the infation target-
ing regime, but also in the current macro framework of the country.
To achieve this, one does not need to drastically change the mandate
and autonomy of the central bank. What is needed is perhaps a gradual
but sure and progressive change in its attitude. The frst hurdle of shifting
away from overemphasizing infation over growth has been achieved. The
next step is just to make this more explicit. The hardest part is convincing
not only the central bank but the entire government, business and fnancial
sectors of the country that high and sustained growth requires an active
industrial policy supported by monetary and credit policies. The last
hurdle is the most dif cult one as it entails a rejection of the Washington
Consensus and the adoption of a new and fresh perspective. This is
what heterodox economists, including those involved in this volume, are
working and striving for.
The current global fnancial crisis triggered by a deregulated fnancial
system existing side by side infation targeting regimes that jacked up
global interest rates, which ultimately led to the subprime fasco may
give many economists and policy makers cause to question the old mon-
etary paradigm and pave the way for a framework that provides stronger
and healthier links between the fnancial/monetary sector and the real and
productive sector of the economy.
NOTES
1. The author is grateful to Francis Dakila, Digna Paraso and Zeno Abenoja of the Bangko
Sentral ng Pilipinas for their support and help. He would also like to thank Gerald
Epstein and Erin Yeldan for their positive comments and suggestions in improving this
chapter. All mistakes are the authors.
2. These are cash and short-term government securities which a bank is allowed to keep as
required reserves.
3. The increase in 2006 was, however, due more to conscious prepayment of foreign debt (as
the peso became strong) and domestic debt (as interest rates started to fall).
4. Haussman and Rodrik (2002) refer to the market failure problem of information spill-
over or frst mover problem as the problem wherein the frst mover bears all the risks.
If they succeed, others will imitate them and reduce their market share. If they fail, they
will bear all the losses.
REFERENCES
Dakila, F. (2001), Evaluation of alternative monetary policy rules: a simulation-
based study for the Philippines, PhD thesis, University of the Philippines,
Diliman.
Frenkel, R. and L. Taylor (2006), Real exchange rate, monetary policy and
employment, DESA working paper no. 19.
Guinigundo, D. (2005), Infation targeting: the Philippine experience, Bangko
Sentral ng Pilipinas, Metro Manila.
Haussman, R. and O. Rodrik (2002), Economic development as self-discovery,
National Bureau for Economic Research working paper no. 8592.
Lim, J. and C. Bautista (2006), External liberalization, growth and distribution in
the Philippines, in L. Taylor (ed.), External Liberalization in Asia, Post-Socialist
Europe, and Brazil, New York: Oxford University Press, pp. 267310.
Lim, J. and M. Montes (2002), Structural adjustment after structural adjustment,
but why still no development in the Philippines?, The Asian Economic Papers, 1
(3) (Summer), 90119.
299
14. Monetary policy in Vietnam:
alternatives to infation targeting
Le Anh Tu Packard
1
14.1 INTRODUCTION
One of the most important challenges facing policy makers is to determine
how monetary policy should be conducted in order to meet their countrys
national development goals. In recent years a growing number of central
banks have convinced each other that the siren song of infation targeting
is worth pursuing,
2
even though a strong theoretical case that this mon-
etary rule possesses superior welfare properties has yet to be established.
Infation targeting calls for the explicit acknowledgment that low and
stable infation is the overriding goal of monetary policy, which implies
that a low infation target should have supremacy over other development
objectives.
3
For Vietnam, the quest for a pro-development monetary policy has
become more urgent because the country is entering a new developmental
phase that will be shaped by the terms of its accession to the World Trade
Organization (WTO) and commercial treaties with its trading partners.
Mindful of both the opportunities and risks that come with this phase,
the Vietnamese government has been looking into macroeconomic and
monetary policy guidelines to manage this period of unprecedented expo-
sure to the world economy. In keeping with recent fashion among central
banks, the State Bank of Vietnam (SBV) expressed interest in exploring
the feasibility of infation targeting. However, the general consensus is that
at present Vietnam does not meet the necessary conditions to implement
an infation targeting regime because the central bank lacks adequate tools
to carry out an efective infation targeting monetary policy. Additionally,
other of ces of economic management in the Ministries of Finance,
Planning and Investment, Industry and Trade do not view infation tar-
geting as a matter of urgency, nor are they willing to cede to the SBV that
degree of power. There is also the larger question of whether an infation
targeting regime is compatible with Vietnams development priorities, and
whether it would help improve economic performance over the long run.
The context for discussing monetary policy in Vietnam is as follows:
upon launching sweeping reforms during the 1990s, the country has gen-
erally followed the East Asian developmental state model. In the view
of its political leaders, monetary policy should serve as a tool to meet
the countrys socioeconomic development goals, which are rapid and
sustainable growth, modernization, industrialization and poverty reduc-
tion. According to the 1998 Law on the State Bank of Vietnam, the task
of the central bank is to stabilize the value of the currency, secure the
safety of the banking system and facilitate socioeconomic development in
keeping with the nations socialist orientation (Kovsted et al., 2002, Thao
2004). SBV senior of cials consider this a mandate to control infation
and promote economic growth (Nghia, 2005). The implied assumption
shared by policy makers and the public is that the nominal exchange rate
and the domestic price level are closely linked. This link became highly
visible during the Doi Moi (Renovation) reforms when Vietnam was grap-
pling with hyperinfation and large depreciation of the parallel market
exchange rate. Unlike many other developing countries, Vietnam has a
strong domestic constituency for low infation because of its earlier trau-
matic experience with hyperinfation, which heightened public sensitivity
to price movements. That said, a strong preference for low infation and
willingness to carry out policies to support a low infation environment
does not mean that the central bank has a clear mandate to adopt formal
infation targeting.
This chapter aims to contribute to the search for the right mix of macro-
economic and monetary policies that can best serve Vietnam in the coming
period of greater openness and intensifed competition in both domestic
and global markets. It examines the factors that should guide monetary
policy, taking into account the current state of Vietnams transition to a
more market-oriented economy and the challenges posed by dollarization,
fnancial repression, informal and underdeveloped fnancial markets, and
rapid international economic integration. Not surprisingly, it fnds that
critical gaps in knowledge, institutional arrangements, tools and rules are
impeding the efectiveness of monetary policy.
Sharing the view that the main task of the central bank should be to
maintain macroeconomic prices that are conducive to rapid and sustain-
able economic growth, an alternative to infation targeting is proposed.
To support Vietnams transition to a more market-oriented economy,
the central bank should instead target a real exchange rate (RER) that is
stable and competitive. It is argued that this key relative price has a more
powerful infuence on the allocation of labor and capital, and on the com-
position of domestic output, than administrative levers typically employed
by centrally planned economies. Maintained over an extended period, a
Monetary policy in Vietnam: alternatives to infation targeting 301
stable and competitive RER promotes an ef cient allocation of resources
and employment-creating growth, reinforces macroeconomic and fnan-
cial stability, and encourages fnancial market development.
The chapter argues that a stable and competitive RER is a superior
intermediate target for several reasons. First, this target clearly is consist-
ent with the Law on the State Bank of Vietnam,
4
which states that the
SBVs task is to stabilize the value of the currency. Since Vietnam now
has multiple important trading partners, the value of the domestic cur-
rency should be stabilized against a trade-weighted basket of currencies
that take into account diferences in their respective rates of infation.
Second, it improves the transparency of monetary policy and strength-
ens confdence in the central banks ability to conduct monetary policy
efectively. In other words, the central bank is assigned a task that is
realistic and therefore doable. Third, a stable and competitive RER can
contribute substantially to economic growth and employment creation
if it is supported by complementary fscal, monetary and industrial poli-
cies. Fourth, it can have positive medium- to long-term impacts on struc-
tural change and development through a variety of channels: resource
allocation, changes in production techniques and growth of capital
stock including stock of human capital (Frenkel and Taylor, 2005).
5

Fifth, compared to a strict focus on infation targeting which tends to
slow economic growth and lower employment growth (Epstein, 2003), a
RER target is more likely to be a more efective stabilizing force and can
do a better job in dampening output volatility during periods of global
turbulence.
A stable and competitive RERs long-term positive impact on resource
allocation and the composition of output takes place through its infu-
ence, both direct and indirect, on key macroeconomic prices such as the
domestic interest rate, the relative price of traded to non-traded goods,
the relative cost of capital and labor, and the import-export price ratio.
The RER, used in conjunction with appropriate commercial and indus-
trial policies, can serve as a development tool in coordination with other
monetary policy instruments to strengthen the economys overall com-
petitiveness, increase aggregate productivity, maintain external balance,
contain infation and stabilize asset markets (Frenkel and Taylor, 2005).
International evidence from cross-country empirical research provides
support for this view: instability (variability) of the RER is found to be
negatively related to growth (Corbo and Rojas, 1995; Montiel, 2003),
and overvaluation of the RER (in other words, an uncompetitive RER)
has been linked with slower growth (Montiel, 2003; Razin and Collins,
1997).
The chapter is organized as follows: Section 14.2 provides a brief history
of Vietnams banking system. Section 14.3 analyses the macro economy
from the perspective of identifying transmission mechanisms. Section
14.4 examines the issues surrounding the framework for monetary policy.
Finally, Section 14.5 describes the merits of a stable and competitive RER
as a superior alternative to infation targeting, and ofers some concluding
comments.
14.2 VIETNAMS BANKING SYSTEM: BRIEF
HISTORY
From 1976 to 1989, like other centrally planned economies, Vietnams
single-tier banking system was owned and controlled by the state. The
SBV provided nearly all domestic banking services through a vast
branch network. Bank lending was state directed, and credit ration-
ing was imposed because fnancial resources were scarce. During this
period, SBV of ces served as the interface between state planning, the
national budget and state entities including some 12 000 state-owned
enterprises (SOEs).
6
Under central planning, the SBV was not required
to carry out many traditional functions of commercial banking, such as
credit analysis or risk management, and its main task was to ensure that
fnancial resources were allocated to economic units in accordance with
the plan.
The quantity of currency outside banks was very high (the ratio of cur-
rency outside banks to nominal GDP reached 9.2 percent in 1986) as the
government attempted to monetize sharply rising fscal defcits as revenue
growth failed to keep pace with rising expenditures (World Bank, 1991).
7

SOEs in Vietnam lacked fscal discipline as they operated under the soft
budget constraint that was common among socialist countries. To cir-
cumvent credit rationing, they engaged in unauthorized credit creation
through various means such as abuse of the check payment system and
use of supplier credits
8
as a substitute for borrowing in credit markets.
These practices had infationary consequences, created fnancial problems
for the SBV and contributed to deterioration in the consolidated balance
sheets of SOEs.
Before money and capital markets were established during the 1990s,
household liquid and semi-liquid assets mainly consisted of the domestic
currency, gold, hard currency notes, and easily tradable commodities,
such as rice. Remittances from overseas Vietnamese
9
contributed to the
dollarization of the economy. Continued eforts by households and other
economic agents to protect themselves from infation by reducing their
domestic currency holdings (causing the ratio of currency outside the
banking system to GDP to decline from 9.2 percent in 1986 to 6.6 percent
in 1988) only worsened the infationary spiral.
The 198789 macroeconomic crisis and hyperinfation provided the
impetus for the comprehensive and coordinated Doi Moi reforms. In 1988
the Prime Minister signed Decree No. 53/ND which ended the monobank
system and created a two-tier system consisting of the SBV as the central
bank and four state-owned commercial banks (SOCBs). In addition, the
government ended the state monopolies on fnancing foreign trade and
on providing long-term fnance. The intent was to increase management
autonomy and responsibility, and to introduce the pressure of competition
in order to improve bank performance (ibid.).
In 1990 the government promulgated two banking ordinances for a
two-tier banking system. These ordinances transformed the SBV into
a central bank with oversight over the domestic banking system and
provided the legal framework for commercial banks and other fnancial
institutions. The government liberalized entry into the banking system and
lifted rules on sectoral specialization of the SOCBs. Commercial banks
were given responsibility for the operation and control of their fnances
and implementation of universal banking activities. The decision to raise
the interest rate for household deposits in the formal banking system
increased confdence in the domestic currency and encouraged households
to deposit their dong assets in bank accounts (Figure 14.1). In 1989 RERs
on households rose sharply, and encouraged a steady rise in the value of
household deposits from VND 207 billion in March 1989 to VND 1348
billion by January 1990 (Figure 14.2).
As regards the current conduct of monetary policy, the broad division
of labor currently is as follows: the governments task is to prepare a plan
4
2
0
2
4
6
8
10
12
14
A
p
r
-
8
9
M
a
y
-
8
9
J
u
n
-
8
9
J
u
l
-
8
9
A
u
g
-
8
9
S
e
p
-
8
9
O
c
t
-
8
9
N
o
v
-
8
9
D
e
c
-
8
9
J
a
n
-
9
0
F
e
b
-
9
0
M
a
r
-
9
0
A
p
r
-
9
0
M
a
y
-
9
0
Nominal household demand deposit rate: monthly
Nominal household 3-month deposit rate: monthly
Inflation rate: monthly
Real household demand deposit rate: monthly
Sources: State Bank of Vietnam, World Bank (1991, p. 86, 1992).
Figure 14.1 Infation and interest rates, Vietnam 198990
for monetary policy that includes an infation forecast, while the National
Assemblys tasks are to set an annual target for the infation rate that is
consistent with the state budget and economic growth objectives, and to
supervise the implementation of monetary policy. As it currently stands,
Vietnams monetary policy strategy is but a component of the broader
fve-year socioeconomic development strategy that is formulated by the
government and the ruling Communist Party. Within this framework,
the SBVs role is to come up with a concrete action plan for the banking
sector, which includes setting targets for the amount of liquidity needed
by the economy. Specifcally, the SBV announces annual targets for total
liquidity (M2 growth) and credit growth. However, the goal of keeping
within credit growth targets does not appear to receive high priority, as
actual credit growth in both 2004 and 2005 were well above the targeted
growth rate of 25 percent.
The SBV is also tasked with stabilizing the exchange rate,
10
an appro-
priately vague mandate that ought to direct the authorities to maintain a
trade-weighted infation-adjusted exchange rate that promotes sustainable
(non-infationary) growth over the medium to long term. Like China and
Singapore, the Vietnamese currency is of cially pegged to a basket of cur-
rencies, and like China, the authorities do not disclose the currency compo-
sition in the basket nor the basket weights. From January 2000 to December
2006, Vietnams real efective exchange rate has depreciated measurably,
especially in comparison to China and its South East Asian neighbors
(Figure 14.3). This depreciation corresponds to a moderate acceleration of
Vietnams GDP growth rate from 6.8 percent in 2000 to 8.4 percent in 2005,
and a somewhat more rapid acceleration of the infation rate.
0
200
400
600
800
1000
1200
1400
1600
0
2
4
6
8
10
12
14
Apr-89 Jun-89 Aug-89 Oct-89 Dec-89
Real household 3-month
deposit rate: monthly
Household monthly
bank deposits
Sources: SBV, World Bank (1992).
Figure 14.2 Household deposits, Vietnam
14.3 DESCRIPTION OF THE MACRO ECONOMY
14.3.1 GDP and Macro Aggregates: Mechanisms of Adjustment
To carry out monetary policy efectively, policy makers in Vietnam need
to have a much better grasp of the actual mechanisms of transmission
and adjustment than they do at present. For example, the transmission
of monetary policy via the interest rate channel is unclear because credit
market segmentation, fnancial repression and credit rationing add addit-
ional layers of murkiness to the process. Through its short-term policy rate
and commercial bank reserve requirement, the SBV is able to infuence
the commercial bank lending rate and activity levels of enterprises that
borrow from the formal fnancial sector. However, its infuence over credit
growth in the informal fnancial sector and informal lending rates is not at
all clear.
11
Furthermore, the picture is obscured by the countrys ongoing struc-
tural transformation that has led to a gradual fattening of the investment
spending curve, as investment spending becomes more sensitive to interest
rate changes. This is illustrated in Figure 14.4, where the expected link
between gross capital formation by enterprises and real lending rates has
not emerged until after 1994. Even so, investment spending in Vietnam
continues to be less sensitive to interest rate movements compared to other
countries with more developed fnancial sectors. This is because retained
earnings continue to be the main source of fnancing for business capital
spending.
To elaborate further on the problem of incomplete information, both
70.00
80.00
90.00
100.00
110.00
120.00
130.00
J
a
n
-
2
0
0
0
M
a
y
-
2
0
0
0
S
e
p
-
2
0
0
0
J
a
n
-
2
0
0
1
M
a
y
-
2
0
0
1
S
e
p
-
2
0
0
1
J
a
n
-
2
0
0
2
M
a
y
-
2
0
0
2
S
e
p
-
2
0
0
2
J
a
n
-
2
0
0
3
M
a
y
-
2
0
0
3
S
e
p
-
2
0
0
3
J
a
n
-
2
0
0
4
M
a
y
-
2
0
0
4
S
e
p
-
2
0
0
4
J
a
n
-
2
0
0
5
M
a
y
-
2
0
0
5
S
e
p
-
2
0
0
5
J
a
n
-
2
0
0
6
M
a
y
-
2
0
0
6
S
e
p
-
2
0
0
6
J
a
n

2
0
0
0

=

1
0
0
Vietnam China Thailand
Singapore Indonesia Malaysia
Figure 14.3 Real efective exchange rate
aggregate money supply and important elements of the money demand
function are unknown (Hauskrecht and Nguyen, 2004) because the
economy is partially dollarized and there is no reliable data about the
quantity of US dollars and stock of gold outside the banking system that
are used as a medium of exchange and a store of value.
12
In particular, in
Vietnam the aggregate true stock of money is hard to estimate because it
includes M2 (recorded by the SBV), foreign deposits held in banks, as well
as two signifcant unobserved variables, private sector foreign currency
holdings and gold in circulation. It is also likely that the domestic and
foreign currency will have diferent velocities (ibid.) with diferent trajec-
tories, thus complicating the formulation based on the simple formula-
tions that rest on the quantity theory of money.
Figures 14.5A and 14.5B present the velocity time path for currency
outside banks (V1) and for total liquidity M2 (V2), which includes cur-
rency outside banks, domestic currency deposits and foreign currency
deposits. Two mutually ofsetting infuences on velocity deserve mention.
Ongoing structural reform of the fnancial sector and improvements in the
payments system increases velocity. At the same time, in a multi-currency
economy a large-scale portfolio switch to the domestic currency can lower
velocity. Figures 14.5A and 14.5B illustrates this: from 1991 to 1994
households and frms switched to the domestic currency and reduced their
non-bank foreign currency and gold holdings as they trusted more the
governments ability to control infation. This brought about a decline in
velocity and also led to greater monetary deepening.
5.0
0.0
5.0
10.0
15.0
P
e
r
c
e
n
t

(
%
)
20.0
25.0
30.0
1
9
9
2
1
9
9
3
1
9
9
4
1
9
9
5
1
9
9
6
1
9
9
7
1
9
9
8
1
9
9
9
2
0
0
0
2
0
0
1
2
0
0
2
2
0
0
3
2
0
0
4
2
0
0
5
Lending rate, working
capital (short-term), real
Inflation rate (CPI) Gross capital
formation by
enterprises,
% of GDP
Figure 14.4 Infation rate, lending rate and gross capital formation by
enterprises
Accurate tracking of domestic credit growth is also critical to the efec-
tive conduct of monetary policy. This is yet another problem for the central
bank, because key variables that afect fnancial sector development and
domestic credit growth in Vietnam are dif cult to estimate. These include
the magnitude of inter-frm credit as percent of aggregate credit creation
and the quality of their accounts receivable (which may pose a signifcant
risk to the banking system). Finally complicating the task is the murky
link between bank credit growth, the infation rate and actual borrowing
by business enterprises (see Figure 14.4) due to the coexistence of formal
and informal fnancial markets, and the role of inter-frm credit.
As regards exchange rates, the impact of the central banks exchange
Figure 14.5A Velocity time paths, 19862002
Velocity 2: nominal GDP/M2
Velocity 1: nominal GDP/currency outside banks
Inflation rate
1
0
1
2
3
4
5
6
7
8
9
0
2
4
6
8
10
12
14
16
1986 1988 1990 1992 1994 1996 1998 2000 2002
P
e
r
c
e
n
t

p
e
r

a
n
n
u
m

(
%
)
V
e
l
o
c
i
t
y

r
a
t
e
Velocity 2: nominal GDP/M2
Velocity 1: nominal GDP/currency outside banks
Inflation rate
100.0
0.0
100.0
200.0
300.0
400.0
500.0
600.0
700.0
800.0
900.0
0.00
2.00
4.00
6.00
8.00
10.00
12.00
14.00
16.00
1
9
8
6
1
9
8
7
1
9
8
8
1
9
8
9
1
9
9
0
1
9
9
1
1
9
9
2
1
9
9
3
1
9
9
4
1
9
9
5
1
9
9
6
1
9
9
7
1
9
9
8
1
9
9
9
2
0
0
0
2
0
0
1
2
0
0
2
2
0
0
3
2
0
0
4
2
0
0
5
P
e
r
c
e
n
t

p
e
r

a
n
n
u
m

(
%
)
V
e
l
o
c
i
t
y

r
a
t
e
Figure 14.5B Velocity time paths, 19862005
rate policy on the real economy is dif cult to determine because it is not
possible to predict how Vietnams informally pegged exchange rate regime
afects the growth of monetary aggregates. The SBV does not provide
information on its interventions in the foreign exchange market, and there
is no explicit sterilization policy. Yet, thus far, SBV actions to manage the
informal peg do not appear to have negative consequences. The M2 growth
rate has not been overly volatile and the infation rate has been low.
14.4 MACROECONOMICS AND INSTITUTIONAL
FRAMEWORKS OF CENTRAL BANKING
14.4.1 Issues Surrounding Scope for Infation Targeting
Although there is interest in infation targeting on the part of the SBV,
and a steady stream of international expert advice is provided on infa-
tion targeting for the SBV senior management, the consensus view is
that at present the conditions to support a formal infation targeting
monetary framework are not met. The reasons are evident (see Sections
14.1 and 14.2) when we consider the four main conditions outlined by
the International Monetary Fund (IMF) that are deemed necessary to
support such a framework (Carare et al., 2002):
The central bank has a clear mandate to make infation targeting the
primary objective of monetary policy and is publicly accountable for
meeting this objective.
The infation target will not be subordinated to other objectives and
monetary policy will not be dominated by fscal priorities.
The fnancial system is developed and stable enough to implement
the infation targeting framework.
The central bank has adequate policy instruments to be able to infu-
ence infation.
At present, the SBV has limited scope to implement monetary policy
using market-based indirect instruments to infuence infation,
13
although
this has long been its declared objective, because fnancial markets are thin
and not well developed (the government securities market is segmented
and illiquid). In addition, as explained above, the government is only at
the early stage of building the necessary foundations (including timely
access to a high frequency databank of key economic and fnancial vari-
ables needed for policy analysis) for developing a reasonable understand-
ing of the links between the stance of policy and infation.
Nevertheless, there are many additional reasons why a rigid infation
targeting framework is not appropriate for Vietnam, even if the conditions
for infation targeting are met. First, it gives primacy to the wrong target
(infation), forcing policy makers to operate in a framework that implicitly
accords higher priority to infation than to other more pressing develop-
ment objectives. For example, it obliges the central bank to automatically
adopt a tightening stance whenever the infation indicator rises above its
target range, or risk being branded as incompetent for failing to stick to
the infation target. A rigid infation targeting framework also sets false
standards for judging the quality of monetary policy, distracting policy
makers from more serious and arduous eforts to understand the actual
workings of their economy.
Second, it is not so easy to determine what should be the right rate
of infation to target, and the SBV may fnd it much too tempting to
simply follow the lead of other central banks even when that may not suit
Vietnams particular circumstances.
Third, it sends the wrong message about what is needed to ensure
good policy making. The implicit underlying justifcation for infation
targeting is that policy makers cannot be trusted to make sound policy
decisions. The assumption is that they tend to give in to short-sighted
political demands that can harm national social welfare over the long run.
Therefore, to protect against this, policy makers must be bound to tight
rules and explicit objectives, and they must be held publicly accountable
to meeting these objectives. Although there are valid points in these argu-
ments, tying their hands with rigid rules and wrong targets could very
well push the economy onto a diferent path that departs sharply from the
countrys development goals.
14.4.2 Quality of Monetary policy
Surprisingly, the government managed to achieve favorable macro-
economic results in spite of having to operate somewhat in the dark
(given the critical gaps in information described in some detail above)
with the crudest of monetary tools to infuence aggregate demand. One
explanation for this success is that these constraints did not prevent the
government from pursuing the right fscal and monetary policies (see
Section 14.2). The economic administration seems to have maintained
macroeconomic stability and succeeded in keeping infation under
control over a prolonged period, from 1990 to 2005. The gains from
achieving this credibility can be seen in the progress made in monetary
deepening, as the ratio of M2 to nominal GDP more than doubled from
its nadir in 1993. The re-intermediation of foreign currency previously
held outside the banking system also indicates greater confdence in the
banking system.
Thus far, policy making credibility has remained strong despite
poor information and weak monetary instruments because key fscal
and monetary policies have been well managed. Both the stock and fow
of government debt (including debt denominated in foreign currency)
have stayed at prudent levels, the outlook for fscal balance remains
healthy and the expected trajectory for the current account defcit does
not give cause for concern. However, the monetary authorities do not
have adequate tools at present to protect the economy from exogenous
shocks, which means that Vietnams vulnerability to external shocks
will increase as its economy becomes more integrated with the global
economy.
Moreover, the risk of policy error is likely to increase if the govern-
ment fails to address the issue of critical gaps in information. As noted in
Section 14.2, the coexistence of formal and informal fnancial markets and
the role of inter-frm credit in liquidity creation have made the relationship
between bank credit growth and actual borrowing by business enterprises
blurry and hard to predict. Without better information, the central bank
runs the risk of misinterpreting data and may respond inappropriately,
with dire consequences. For example, the SBV may attribute a too high
rate of credit growth to excessive monetary or fscal easing, when in fact
these high numbers may actually be the result of credit reallocation due to
a secular rise in the formalization of credit and decline in informal sector
lending.
14.5 INVESTIGATING ALTERNATIVES TO
INFLATION TARGETING
14.5.1 The Real Exchange Rate is a Better Target
An alternative policy target should be consistent with and support wide-
ranging development priorities. In other words, an important criteria for
the monetary target is that it should play a positive development role and
actively support the economys structural transformation. To this end, we
need to specify the critical components of this economic transformation in
order to sharpen monetary policys role, and to ensure that it becomes a
coherent part of the nations development strategy.
Taking into account these considerations, a stable and competitive
real exchange rate (SCRER) is regarded to serve as a better intermediate
target because it helps to advance Vietnams national priorities, which
are rapid and sustainable growth, modernization, industrialization and
poverty reduction. In contrast to infation targeting regimes that have
been found to push many economies into a lower employment growth
trajectory, SCRER targeting would contribute to growth and employment
creation through its impact on resource allocation, rewarding frms that
adopt forward-looking production technologies and encouraging them to
develop promising new businesses.
To lay the foundations for sustainable growth, strategies that give
priority to developing the medium to large enterprise (MLE) sector may
be more efective than strategies that advocate concentrating resources
on developing the SME sector. This is because mid-size enterprises have
better capacity for learning and for technological innovation, and can
create more jobs faster. Development of this sector will accelerate for-
malization of the economy (enabling policy makers to better monitor
economic activity), promote human capital development, technological
development and development of management skills, and strengthen the
competitiveness of domestic frms. Given that the long-term survival and
growth of enterprises depends on their being able to maintain a healthy
proft rate, the stability of employment growth in the formal sector is
closely linked to an environment that is conducive to MLE growth. An
important aspect of this environment is that monetary policies send con-
sistent signals to af rm the basic stability of key macroeconomic relative
prices including the RER. This is needed so that enterprises will be conf-
dent enough to proceed with their investment plans in order to develop in
areas that are most likely to be proftable.
A SCRER targeting framework for monetary policy is key to promot-
ing rapid expansion of the MLE sector. This is necessary to ensure that
employment in the formal sector (which is dependent on MLE growth)
will increase at a rate that can absorb Vietnams rapidly growing labor
force (about 1.2 million new entrants to the workforce every year). The
economic well-being of Vietnamese workers depends on this, because
wage rates in the formal sector are signifcantly higher than informal
sector wage rates. A SCRER target also helps the government to reduce
its reliance on administrative levers to bring about desired changes in the
economy. Of cials have less justifcation to yield to pressure from frms
in import-substitution sectors for special protections. If frms in sectors
such as paper, steel and cement are unable to survive and prosper in a
favorable price environment created by a SCRER, the government should
conclude that they are unlikely to achieve long-term commercial viability.
Consequently, the economy would achieve better resource allocation if
these frms were to close down their operations.
NOTES
1. Earlier versions of this chapter were presented to the May 2005 CEDES/Amherst
Research Conference in Buenos Aires and the July 2005 Da Nang Symposium on
Continuing Renovation of the Economy and Society. Financial support for this project
has been provided by the Ford Foundation, UNDESA and the Rockefeller Brothers
Foundation. My gratitude and thanks go to two anonymous referees, Gerald Epstein,
Erin Yeldan, Jaime Ros, Lance Taylor, Per Berglund, Dang Nhu Van and Phillipe
Scholtes for their insightful comments and valuable ideas. I am responsible for all
remaining errors and omissions.
2. Bernanke and Mishkin (1997).
3. However, Mishkin (2000), an advocate of infation targeting, acknowledges that price
stability is a means to an end, a healthy economy, and should not be treated as an end
in itself and that central bankers should not be obsessed with infation control.
4. Law on the State Bank of Vietnam is dated 12 December 1997.
5. Frenkel and Taylor (2005) emphasize that the RER must be kept at a stable and com-
petitive level for a relatively long period if the positive efects are to take place. The
reason is that responses to the new (competitive) set of relative prices take time because
they involve restructuring frms and sectoral labor market behavior. This takes place
over time via changes in the pattern of output among frms and sectors, and adjust-
ments in technology and organization of production.
6. In 1989 the SOE sector was made up of about 12 000 enterprises, of which 3100 were
in industry; while the remaining were in trade, construction, agriculture and services.
Most SOEs were provincial or district enterprises that were managed by the Industrial
Bureaus of the provincial or district Peoples Committees (World Bank, 1991). The
reform of state enterprises, a key component of the Doi Moi reforms, subjected the SOEs
to a hard budget constraint. By 1992 the number of SOEs fell by nearly half to 6545
enterprises, and their labor force was cut from 2.7 million to 1.7 million (IMF, 1998).
7. The expenditures included the costs of maintaining a large military force, direct subsi-
dies to SOEs and indirect subsidies associated with price controls.
8. This is done by delaying or failing to repay credit extended by their suppliers which gen-
erally were other SOEs. The Vietnamese term employed by SOE managers to describe
this practice is chiem von nhau (conquering each others working capital).
9. This usually took place through informal channels due to unfavorable regulations gov-
erning formal money transfers. Recipients were forced to take the money in Vietnamese
currency at an exchange rate which efectively gave them half or sometimes only a third
of the amount they could get in the open market (Beresford and Dang Phong, 2000).
10. SBV intervention in the foreign exchange market to support the targeted rate of
exchange takes the form of buying and selling foreign currency or engaging in foreign
exchange swaps.
11. As in other countries where credit market segmentation play an important role, frms
in Vietnam that have access to the formal banking system become key actors in the
process of credit creation. They act as fnancial intermediaries to credit-constrained
frms by providing the latter with trade credit. In other words, institutional factors help
to turn inter-frm credit into an imperfect substitute for bank credit.
12. Both function as a quasi second legal tender or parallel currency in the economy
(Hauskrecht and Nguyen, 2004). The government can track the quantity of currency
outside banks and the quantity of dong and dollar deposits. However, the quantity of
gold and hard currency held by households and other economic agents that are used as
a medium of exchange and store of wealth is not known.
13. As regards indirect monetary policy instruments, during the mid 1990s the SBV intro-
duced and has been using required reserves, refnancing and discount lending facilities,
open market operations and foreign exchange interventions. The refnancing rate and
the discount rate together defne the upper and lower band for the open market opera-
tions rate although on occasion this rate does move outside the band (Camen, 2006).
REFERENCES
Beresford, Melanie and Dang Phong (2000), Economic Transition in Vietnam:
Trade and Aid in the Demise of a Centrally Planned Economy, Cheltenham, UK
and Northampton, MA, USA: Edward Elgar.
Bernanke, B. and F. Mishkin (1997), Infation targeting: a new framework for
monetary policy?, Journal of Economic Perspectives, 11 (2).
Carare, A., A. Schaechter, M. Stone and M. Zelmer (2002), Establishing initial
conditions in support of infation targeting, International Monetary Fund
working paper WP/02/102.
Camen, U. (2006), Monetary policy in Vietnam: the case of a transition country,
Bank for International Settlement papers no. 31, accessed at www.bis.org/publ/
bppf/bispap31t.pdf.
Epstein, Gerald (2003), Alternatives to infation targeting monetary policy for
stable and egalitarian growth: a brief research summary, working paper no. 62.
employment, DESA working paper no. 19, United Nations, February; extended
version accessed in Alternatives to Infation Targeting, 2, from the Political
Economy Research Institute, accessed at www.un.org/esa/desa/papers/2006/
wp/9_2006.pdf.
Hauskrecht, A. and T.H. Nguyen (2004), Dollarization in Viet Nam, Indiana
University, Kelley School of Business, Department of Business Economics
and Public Policy working paper, accessed at http://ideas.repec.org/p/iuk/
wpaper/2004-25.html.
IMF (International Monetary Fund) (1998), Vietnam: selected issues and statisti-
cal annex, IMF staf country report no. 98/30.
IMF (1999), Vietnam: selected issues, IMF staf country report no. 99/55.
IMF (2003), statement by Sean Nolan, IMF Division Chief, Asia and Pacifc
Department, at the Consultative Group Meeting for Vietnam, accessed at www.
imf.org/external/np/dm/2003/120203.htm.
IMF (2004), staf report for the 2004 Article IV consultation.
Kovsted, J., J. Rand, F. Tarp, N.D. Tai, N.V. Huong and T.M. Thao (2002),
Financial sector reforms in Vietnam: selected issues and problems, CIEM/
NIAS discussion paper no. 0301, Hanoi.
Mishkin, Frederic S. (2000), Infation targeting in emerging market countries,
American Economic Review, 90 (2) (May), 1059.
Montiel, Peter (2003), Macroeconomics in Emerging Markets, Cambridge:
Cambridge University Press.
Monetary Authority of Singapore (2001), Singapores exchange rate policy,
accessed 8 August 2006 at www.mas.gov.sg.
Nghia, L.X. (2005), Orientations of the banking sector development strategy
towards international integration, paper presented at the 18 January seminar
on Preparation for the 20062010 Five Year Social Economic Development
Plan in the Banking Sector, Hanoi.
Razin, Ofair and Susan M. Collins (1997), Real exchange rate misalignments and
growth, National Bureau for Economic Research working paper no. w6174.
Thao, N.D. (2004), Strategy for the development of banking services in Vietnam
in the context of international economic integration, Project VIE/02/009
Banking Service Sector, Hanoi, August.
Thanh, V.T., D.H. Minh, D.X. Trong, H.V. Thanh and P.C. Quang (2001),
Exchange Rate in Vietnam: Arrangement, Information Content and Policy
Options, Hanoi: Central Institute for Economic Management, February.
Tho, T.V., N.N. Duc, N.V. Chinh and N. Quan (2000), Kinh Te Vietnam. 1955
2000. Tinh Toan Moi, Phan Tich Moi [Vietnam Economy. 19552000. New
Estimates, New Analysis], Hanoi: Statistical Publishing House, December.
UNIDO (2004), Vietnam Agenda 21. National Conference and Forum on
Sustainable Development in Vietnam, Document prepared by UNIDO Country
Of ce in Vietnam for the Forum on Sustainable Industry and Business, Hanoi,
December.
World Bank (1991), Transforming a state owned fnancial system: a fnancial
sector study of Vietnam, report no. 9223-VN, Washington, DC, April.
World Bank (1992), Vietnam: restructuring public fnance and public enterprises,
report no. 10134-VN, Washington, DC, April.
315
accountability 3, 4
aggregate demand 16, 28, 42, 118, 124,
187, 19091, 199
see also efective demand
anti-infationary policy 6, 75, 87
anti-unemployment policy 735
Argentina 23, 131, 179202
Augmented DickeyFuller Test 260
balance of payments 58, 272, 276, 279
black market see parallel markets
Brazil 22, 131, 13955, 158
Brazilian Central Bank 13943, 145,
1478, 1512
call money rate 26061
capital adequacy ratio 289
capital controls 224, 41, 44, 55, 190,
193, 1978, 233, 2434, 248, 264,
26970, 2912
capital fight 139, 142, 243
capital fows, international 40, 204,
21518, 220, 223
capital management techniques 1824,
225, 2335, 2434
central banks 324
credibility of 15960, 166, 172, 173
and development 5
net worth of 1945
central bank independence 4, 8, 87, 187
Chakravarty Committee 257
Chile 24, 158, 190, 244
China 24, 112, 249
class 7187
class confict 71, 75, 93
cointegration 163, 166
Colombia 158
Comit de Poltica Monetria 141
computable general equilibrium
model see general equilibrium
model
conditional least squares 119
Conselho Monetrio Nacional 140
contagion efects 139, 151
convertibility 1813, 185, 187, 190
corto 160, 1679, 172
crawling band 159
crawling peg 45
credit allocation techniques 2021
credit targeting 24
currency crisis 139, 142, 1478
current account 139, 145, 152, 180,
271, 2749, 2889
Czech Republic 77, 78, 80
defation 93, 97101
demand management 72, 86, 12930
devaluation 17982, 187, 2723, 2757
see also exchange rate, real exchange
rate
development channel 16
dirty foat 140, 153, 155
disinfation 142, 145, 154, 171
Doi Moi reforms 300, 303
East Asia 248, 26970
economic growth 132, 17981, 191,
249, 257, 263
see also infation and economic
growth, SCRER
efective demand 29, 323, 36
see also aggregate demand 28
emerging market economies 158
employment 245
targeting 18, 24, 22747
equilibrium, macroeconomic 347, 47
expectations 1812, 1845, 192, 1967
exchange rate 28
appreciation 139, 142, 145, 148, 152,
153, 155
depreciation 139, 142, 144, 145, 148,
151, 153, 155
Index
devaluation 17982, 187
and economic development 28, 3031
and economic policy 28, 37
and external balance 28
and fnance 29
and infation 29
and monetary policy 159, 16771
over-valuation 301
pass-through 8, 139, 1714
pegged 291
and resource allocation 28, 29
targeting 13942, 1467, 150, 155,
17999, 252, 31011
volatility 180, 187, 1912, 197
see also real exchange rate, nominal
exchange rate
expansionary policy 723, 75
expansionary fscal contraction 203
export sector 108
see also tradable goods
external balance 29
external fragility 220, 223
Federal Reserve 41
fnancial crisis 146
fnancial instability 5
fscal balance 23, 18081, 1879, 190,
2847, 2923
fscal dominance 3, 205, 214
fscal overhang 256
fscal policy 160, 19091, 1969
fexible exchange rate 16061
see also exchange rate
France 778, 80
GEAR (Growth, Employment and
Redistribution) 2289
gender 93112
gender bias 1089
and central bank policy 11112
see also infation targeting
gender discrimination 95
gender equality 110
gender equity 11012
general equilibrium model 24, 21123
Granger Causality 26062
Greenspan, Alan 3
Haiti 1212
heterodox economics 72
Hodrick-Prescott flter 97, 99, 102
hyperinfation 117
IMF (International Monetary Fund)
5, 8, 9, 119
and conditionality 5
see also infation targeting
impulse response function 164, 238,
2623
income distribution 4, 8
incomes policies 21, 1312, 160
India 24870
infation
aversion 7887
confict theories of 75
core 254, 2813
and economic growth 9, 56,
11636
expected 48
and inequality 132
and poverty 734, 76
and unemployment 93112
infation preferences 7187
infation rate targets 9, 18
infation targeting 4, 412, 45, 55
alternatives to 1724, 2917
and asset price stability 9
and budget defcit 14950
and central bank accountability 8
and economic growth 1011, 56,
11636
and escape clauses 2812
and exchange rate stability 16
and foreign exchange reserves 15
and gender bias 1089, 111
and IMF 5, 44
and industrial policy 3031
and infation rates 8
in open economies 44
and price stability 3
and primary fscal surplus 150
and real exchange rates 4467
and sacrifce ratios 9
in small economies 55
and trade balance 1117
and unemployment 1112, 15, 24,
56, 7187, 93112
infation targeting countries, list 6
infation targeting, socially responsible
alternatives 1724
Index 317
infation variability 158
ILO (International Labour
Organization) 4
India 23, 95107
interest rate 38, 45, 57, 6061, 11630,
15960, 167, 170, 172, 1835, 187,
1928
see also real interest rate
interest rate parity 1923
Ireland 95, 122, 135
Israel 95, 131, 135
Instituto Brasileiro de Geografa e
Estatstica 141
International Social Survey Program
7792
J-curve 34
Jacobian matrix 54, 64
jobless growth 21011
Kaldor 35
Keynesian models 49, 50, 60
labor intensity channel 16
labor market 29, 323
Lebac 18290
Lerner Symmetry Theorem 30
liberalization 4
logistic regression 81, 836
Lula 1445, 147
macroeconomic channel 16
macroeconomic instability 5
Malaysia 24, 244
Mexico 18, 55, 95, 131, 135, 15878
Minsky, Hyman 39
monetarism 117
monetary base 179, 18590, 192, 194,
195
monetary policy 412, 45, 934, 97,
102, 105, 11011, 132, 182, 187,
190, 1923, 1979
pro-development 299
money supply growth 94, 97,
1056
and womens employment
10810
monetary targeting 159, 190, 2512,
256, 2712, 275, 27982
modifed 280
macroeconomics
new classical 71, 73
new consensus 49, 55
New Keynesian 72
see also post-Keynesian economics
and structuralist
macroeconomics
NAIRU (non-accelerating infation
rate of unemployment) 72,
117
narrow money 252, 26061
Nicaragua 122
nominal anchor 258
nominal exchange rates 3739, 41, 42,
45, 46, 182, 192, 193, 1957
and middle income countries 39
and targeting 159
see also exchange rate, real exchange
rate
nonlinearity 119
non-tradables see non-traded goods
non-traded goods 28, 3036
North American Free Trade
Agreement (NAFTA) 165
open market operations 45, 57
outlier 119, 12021
output gap 1667
output-infation trade-of 158
output variability 158
parallel market 300
pass-through 163, 1656, 1714
see also exchange rate
Peru 131, 158
Philippines 234, 27198
Phillips Curve 71, 117, 166, 278
political economy 712, 756
and gender 93112
policy space 7
post-Keynesian economics 49, 72
potential output 1525
poverty 734
alleviation of 249
price rigidities 58
price stability 9, 11, 14
primary surplus 141,150
primary surplus target 203, 210, 219,
2234
productivity 29, 313, 357
prudential regulations 1989
rational expectations 712, 86
real exchange rate 2842, 3768,
94, 97, 1059, 179202, 2634,
26970, 300
appreciation of 32, 39, 55, 171
and capital fows 4041
and central bank intervention 39,
41
depreciation of 31, 166, 168, 171,
180, 189, 198
and employment 36, 65
fxed 55
and output 1635
overvaluation 301
and targeting 44, 45, 173, 2512, 264
and wage share 53
see also exchange rate, nominal
exchange rate
real interest rate 94, 97, 1026, 109,
1413, 1458, 150, 152, 155, 206,
208, 212, 220, 232
see also interest rate
relative infation aversion see infation
aversion
remittances 2834, 287, 289, 291, 302
Reserve Bank of India 248
reserve requirements 45, 52
Ricardo-Viner Model 30
root mean square error 261
Rwanda 122
sacrifce ratio 8, 97
SCRER (stable and competitive real
exchange rate) 16, 17, 18081,
187, 18992, 1969, 31011
seigniorage 195
SELIC (Sistema Especial de
Liquidao e Custodia) interest
rate 141, 143, 147
Singapore 95, 100108
stagfation 105
state-owned enterprises 302
supply shock 10, 146
South Africa 24, 55
South Korea 24, 122
State Bank of Vietnam 299301
sterilization 18890, 194, 1968
structuralist macroeconomics 1617,
445
surplus labor 250
Sweden 131
Taylor Rule 49, 57
Tinbergen Rule 194
tradables see traded goods
traded goods 28, 3036, 96, 1089
transparency 3, 8, 18
trilemma 8, 1718, 39, 41, 45, 47, 52,
140, 181, 1924
Turkey 223, 20326
twin targeting 20326
uncovered interest parity condition
193
under-employment 232
unemployment, male 96
unemployment, female 96
unemployment preferences 7187
VAR (vector auto-regression) models
23638, 262
Vietnam 23, 299314
volatility 8, 9, 15
Zimbabwe 122

Beyond Inflation Targeting

Caricato da

Informazioni sul documento

Descrizione originale:

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

Beyond Inflation Targeting

Caricato da

Copyright:

Formati disponibili

Beyond Infation Targeting

The determinant of the Jacobian is:

Potrebbero piacerti anche