Sei sulla pagina 1di 288

Wages, Employment,

Distribution and Growth


International Perspectives

Edited by

Eckhard Hein, Arne Heise and Achim Truger


Wages, Employment, Distribution and Growth

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
This page intentionally left blank

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Wages, Employment,
Distribution and Growth
International Perspectives

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Edited by

Eckhard Hein
Arne Heise
and
Achim Truger

palgraye rn a emit lan

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Selection and editorial matter © Eckhard Hein, Arne Heise and
Achim Truger 2006
Individual chapters © contributors 2006
All rights reserved. No reproduction, copy or transmission of this
publication may be made without written permission.
No paragraph of this publication may be reproduced, copied or transmitted

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
save with written permission or in accordance with the provisions of the
Copyright, Designs and Patents Act 1988, or under the terms of any licence
permitting limited copying issued by the Copyright Licensing Agency, 90
Tottenham Court Road, London W1T 4LP.
Any person who does any unauthorized act in relation to this publication
may be liable to criminal prosecution and civil claims for damages.
The authors have asserted their rights to be identified
as the authors of this work in accordance with the Copyright,
Designs and Patents Act 1988.
First published 2006 by
PALGRAVE MACMILLAN
Houndmills, Basingstoke, Hampshire RG21 6XS and
175 Fifth Avenue, New York, N.Y. 10010
Companies and representatives throughout the world.
PALGRAVE MACMILLAN is the global academic imprint of the Palgrave
Macmillan division of St. Martin's Press, LLC and of Palgrave Macmillan Ltd.
Macmillan® is a registered trademark in the United States, United Kingdom
and other countries. Palgrave is a registered trademark in the European
Union and other countries.
ISBN-13: 978-1-4039-4962-2
ISBN-10:1-4039-4962-X
This book is printed on paper suitable for recycling and made from fully
managed and sustained forest sources.
A catalogue record for this book is available from the British Library.
Library of Congress Cataloging in Publication Data
Wages, employment, distribution, and growth / edited by Eckhard Hein,
Arne Heise, and Achim Truger.
p. cm.
Includes bibliographical references and index.
ISBN 1-4039-4962-X (cloth)
1. Labor supply. 2. Wages. 3. Income distribution. 4. Economic
development. I. Hein, Eckhard, 1963- II. Heise, Arne, 1960- III. Truger, Achim.
HD5706.W24 2006
331.1—dc22 2005051392
10 9 8 7 6 5 4 3 2 1
15 14 13 12 11 10 09 08 07 06
Printed and bound in Great Britain by
Antony Rowe Ltd, Chippenham and Eastbourne

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Contents

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
List of Tables and Figures
Notes on the Contributors

Introduction
Eckhard Hein, Arne Heise and Achim Truger
1 Labour-Market Flexibility and Economic Expansion
Amit Bhaduri
2 The Causes of High Unemployment: Labour-Market
Sclerosis v. Macroeconomic Policy
Thomas I. Palley
3 Is Capital Stock a Determinant of Unemployment?
Philip Arestis, Michelle Baddeley and Malcolm Sawyer
4 Deflation Risks in Germany and the EMU: The Role
Wages and Wage Bargaining
Eckhard Hein, Thorsten Schulten and Achim Truger
5 The Influence of Unemployment, Productivity and
Institutions on Real Wage Trends: The Case of Italy
1970-2000
Enrico Sergio Levrero and Antonella Stirati
6 Unequal Fortunes, Unstable Households: Has Rising
Inequality Contributed to Economic Troubles for
Households in the USA?
Heather Boushey and Christian E. Weller
7 The Effects of Economic Liberalization on Income
Distribution: A Panel-Data Analysis
Gerardo Angeles-Castro
8 Pensions and Distribution in an Ageing Society: A
Non-Conventional View
Sergio Cesaratto
9 Do Profits Affect Investment and Employment? An
Empirical Test Based on the Bhaduri-Marglin Model
Ozlem Onaran and Engelbert Stockhammer

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
vi Contents

10 Class Conflict and the Cambridge Theory of


Income Distribution 223
Thomas I. Palley
11 The Dynamics of Profit- and Wage-led Expansion: A Note 247
Amit Bhaduri

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Index 255

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
List of Tables and Figures

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Tables
2.1 Country macroeconomic and labour-market
institution data 24
2.2 Time-series unemployment rate regression, pooled
annual data, 20 OECD countries, 1983-94 30
2.3 Decomposition of the causes of changing unemployment
rates into factors due to changing labour-market institution
(DMICRO) and macroeconomic slowdown (DMACRO) 37
2.4 Measures of country income distributions 41
3.1 Cointegrating vectors 60
3.2 Model specification statistics 61
3.3 Key to variables and statistics/diagnostics as in
Tables 3.1 and 3.2 62
4.1 Wage trends and extent to which the scope for
distribution is exploited in the EMU 81
4.2 Wage trends and extent to which the scope for
distribution is exploited in Germany 82
5.1 Determinants of rates of change in industrial wages 98
5.2 Determinants of rates of change in real earnings in
the business sector 100
5.3 Rates of change in hourly productivity, hourly
labour costs, real exchange rates in manufacturing in
the main industrialized countries 107
6.1 Levels and changes of profit shares and labour shares
of national income 119
6.2 Savings and consumer debt, business cycle averages 124
6.3 Selected use of non-financial corporate resources,
business cycle averages 125
6.4 Selected macroeconomic measures, business cycle
averages, 1948-2003 126
6.5 Sources and uses of household finances 134
6.6 Regression results for determinants of household debt,
1980-2003 136
6.7 Regression results for determinants of mortgage
debt, 1980-2003 138

vn

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
viii List of Tables and Figures

6.8 Regression results for determinants of credit-card


debt, 1980-2003 139
6.9 Regression results for determinants of debt
composition, 1980-2003 140
6.10 Regression results for economic distress measures 144

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
7.1 Estimation methods 159
7.2 Scenarios 163
7.3 Scenarios using changes in trade volume (CTRAGDP) 166
7.4 Exports by sector to total exports 168
7.5 Level of employment (UNEMP), employment by sector
(EMPSEC) and government size (GOVEXPEN) 169
A7.1 Characteristics of data-sets on income inequality 175
A7.2 List of countries (93) 175
8.1 Evolution of the world population: alternative UN
scenarios by regions (billions) 184
8.2 Impact of migration flows: alternative UN scenarios
(thousands) 188
8.3 Evolution of labour supply 2000-50, OECD projections 190
8.4 Evolution of participation rates in OECD countries:
scenarios by the OECD 192
8.5 Changes in old-age pension spending 2000-50,
OECD estimates 197
9.1 Summary of the model 209
9.2 Hypotheses 211
9.3 Summary of impulse responses for the UK, the
USA and France 214
9.4 Summary of impulse responses for South Korea and Turkey 216
11.1 Classification of regimes (a > 0, p > 0 or p < 0) 251

Figures
2.1 The economic policy menu 44
4.1 Unit-labour-cost growth and inflation rate (consumer
prices) in Germany, 1961-2003 (per cent) 74
4.2 Unit-labour-cost growth and inflation rate (consumer
prices) in EMU, 1961-2003 (per cent) 75
4.3 Labour income shares in Germany and the EMU,
1960-2003 (per cent of GDP at current factor costs) 76
4.4 Unit-labour-cost growth in Germany and the EMU,
1991-2003 (per cent) 78
4.5 Inflation rate (consumer prices) in Germany and
the EMU, 1991-2003 (per cent) 78

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
List of Tables and Figures ix

4.6 Remuneration per employee in Germany and the EMU,


1991-2003 (annual increase in per cent) 79
5.1 Annual rates of change of real hourly wages of
production workers in trade and industry, 1956-2000 95
5.2 Real gross and net wages in industry (1972 = 100),

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
1972-2000 96
5.3 Unemployment and rate of growth of real wages in
industry, 1960-99 (moving averages) 99
5.4 Unemployment rates and annual rates of change in
industrial and total employees, 1971-2000 105
10.1 The national income tree 225
10.2 The Kaldor (1956) model 227
10.3 The Kaldor-Pasinetti-Kalecki model 230
10.4 Ambiguous effect of an exogenous increase in the
degree of monopoly power in the Kaldor-Pasinetti-
Kalecki model 235
10.5 Expansionary effect of a redistribution of the wage
bill to workers in the Kaldor-Pasinetti-Kalecki model 236
A10.1 Profit-led dynamics with IS steeper than MM 242
A10.2 Profit-led dynamics with IS flatter than MM 242
A10.3 Wage-led dynamics with MM flatter than IS 243
A10.4 Wage-led dynamics with MM steeper than IS 243

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Notes on the Contributors

Gerardo Angeles-Castro is a PhD student sponsored by Conacyt

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
(Mexico), and Teaching Assistant at the Economics Department,
University of Kent, UK.
Philip Arestis is Professor and University Director of Research, Cambridge
Centre for Economic and Public Policy, University of Cambridge, UK.
Michelle Baddeley is Deputy Director, Cambridge Centre for Economic
and Public Policy, University of Cambridge, UK.
Amit Bhaduri is Professor of Economics, University of Pavia, Italy, and
Visiting Professor, Council for Social Development, New Delhi, India.
Heather Boushey is a Research Economist at the Center for Economic
and Policy Research (CEPR), and Research Affiliate with the National
Poverty Center at the Gerald R. Ford School of Public Policy, Washington,
DC, USA.
Sergio Cesaratto is Full Professor of Economic Policy and of
Development Economics at the University of Siena, Italy.
Eckhard Hein is a Senior Researcher at the Macroeconomic Policy
Institute (IMK) in the Hans Boeckler Foundation, Duesseldorf, and
Visiting Professor at the University of Hamburg and at Carl von
Ossietzky University, Oldenburg, Germany.
Arne Heise is Professor of Economics at the University of Hamburg,
Germany.
Enrico Sergio Levrero is Assistant Professor, University of Roma Tre, Italy.
Ozlem Onaran is Associate Professor at Istanbul Technical University,
and Visiting Faculty Member at Wirtschaftsuniversitat Wien, Austria.
Thomas I. Palley is an independent economic consultant who works in
Washington, DC, USA. He can be reached at www.thomaspalley.com.
Malcolm Sawyer is Professor of Economics, University of Leeds, UK.
Thorsten Schulten is a Senior Researcher at the Institute for Economic
and Social Research (WSI) in the Hans Boeckler Foundation, Duesseldorf,
Germany.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Notes on the Contributors xi

Antonella Stirati is Associate Professor, University of Roma Tre, Italy.


Engelbert Stockhammer is Assistant Professor at Wirtschaftsuniversitat
Wien, Austria.
Achim Truger is a Senior Researcher at the Macroeconomic Policy

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Institute (IMK) in the Hans Boeckler Foundation, Duesseldorf, Germany.
Christian Weller is a Senior Economist at the Center for American
Progress and Research Associate at the Economic Policy Institute,
Washington, DC, USA.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
This page intentionally left blank

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Introduction
Eckhard Hein, Arne Heise and Achim Truger

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
The subjects covered in this book are among the most controversial in
the history of economics as an academic discipline: the relation between
wages and employment, the effects of wages on distribution, and the
relation between distribution and growth. However, taking a look at the
state of mainstream economics today, all of these debates seem to have
been resolved by and large. In New Classical as well as in mainstream
New Keynesian Economics there is a clear cut inverse relation between
real wages and employment, at least in the long run. This is also true for
the 'new consensus' models in macroeconomics.1 Although the New
Classical and the New Keynesian schools of thought differ with respect
to the determinants of short-run economic activity and also with respect
to the effectiveness of macroeconomic policies, in the long run it is the
real wage rate which determines employment. Therefore, the main-
stream in both schools of thought focuses on structural reforms in the
labour market and in the welfare state when it comes to fighting persist-
ent unemployment. To improve labour-market flexibility and to make
the social benefit systems more 'employment-friendly' is regarded as the
key to raising employment. This usually includes the reduction of
employment protection legislation, of benefit replacement rates and
durations, and of the tax wedge as well as the decentralization of
wage-setting in order to adjust real wages to workplace productivity.
Macroeconomic policies are assumed to be ineffective in determining
real variables in the long run and should therefore supply a 'stable envi-
ronment', which means that monetary as well as fiscal policies should
aim at assuring price stability.
Income distribution and, in particular, functional income distribution
is hardly discussed at all in mainstream economics today. Atkinson
(2000) speculates that this may be due to the wide acceptance of the

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
2 Introduction

hypothesis of constant income shares in the long run, which, however,


cannot be observed empirically (Atkinson 1997, 2000). When new
growth theories deal with distribution issues, it is personal distribution
of income and wealth which can affect accumulation and hence long-
run productivity growth through different channels. 2 Under the condi-

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
tions of imperfect capital markets, unequal distribution of income and
wealth may impede accumulation of human capital and R&D invest-
ments which determine productivity growth in these models. Unequal
distribution may also cause social unrest and state intervention which
disrupt the transformation of savings into productive real investment.
However, effective demand for goods and effects that functional income
distribution may have on effective demand and capital investment do
not play a role in these models. The long run growth process is purely
supply-side determined.
The papers collected in this volume take a different perspective. They
challenge the view that unemployment is exclusively determined by
structural characteristics of the labour market and the social benefit sys-
tem. Macroeconomic policies and investment in capital stock are
included into the analysis and it is shown that they have a major role to
play when it comes to the short and long-run determination of unem-
ployment. Wage-setting in the labour market has no direct impact on
employment but nominal wages set in this market rather affect the price
level. Following the recommendations of mainstream economics with
respect to labour-market reforms, decentralization of wage bargaining
and wage moderation in an environment of low growth and serious
effective demand problems may therefore contribute to deflationary
risks and may hence worsen the economic situation instead of improving
it. It is also shown that unemployment and 'structural reforms' aiming
at more open and more flexible markets at the national and the inter-
national level may have major feedback effects on functional and per-
sonal income distribution, causing real wage growth to lag behind
productivity growth, falling labour income shares and more unequal
personal income distribution. These developments may then also con-
tribute to slow growth and rising unemployment. This is shown in
applying growth models which rely on the principle of effective demand
also in the long run and which incorporate the effects of distribution
struggle into these models.
In an introductory chapter Amit Bhaduri analyses why the neo-
liberal view on labour-market flexibility and restrictive fiscal and mone-
tary policies as a precondition for economic expansion has become the
dominant view in academics and in economic policy-making today

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Eckhard Hein, Arne Heise and Achim Truger 3

superseding the Keynesian style of demand management. According to


Bhaduri, the latter failed for several internal and external reasons:
Keynesian macroeconomic policies were set in the context of a closed
economy and did not seem to apply to increasingly open and globalized
economies. Full employment orientation lost ground as soon as the

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
competition between capitalist and socialist economic systems had been
resolved; and internal contradictions of the success of the 'golden age' of
welfare capitalism, that is full employment and increasing inflationary
pressure, were not sufficiently taken into account in this approach.
However, neo-liberal policies characterized by methodological individu-
alism and a lack of macroeconomic reasoning suffers from some serious
fallacies of composition: cutting wages may improve the profit margin
of a single firm, but the total volume of sales in an economy is likely
to suffer and overall employment may deteriorate. In particular, neo-
liberal policies do not pay attention to the two-way relationship
between the growth rates in real wages and in productivity, in which
productivity growth allows for increasing wages through cost and price
cutting of firms and rising real wage demands by workers induces firms
to increase productivity in order to maintain profit margins. This virtu-
ous circle propelling long-term growth in capitalism may be undermined
by neo-liberal policies.
In an empirical chapter Thomas I. Palley challenges the conventional
wisdom that high European unemployment is the result of rigid and
inflexible job markets. The chapter accounts for both micro- and macro-
economic factors and considers cross-country economic spillovers.
Palley's estimations show that macroeconomic factors dominate in
explaining unemployment. These factors are robust to changes in empir-
ical specification. Labour-market institutions do matter for unemploy-
ment, but not in the way conventionally spoken about. Unemployment
benefits and union density have no effect. The level of wage bargaining
coordination and the extent of union wage coverage both matter, and if
properly paired can raise incomes without causing unemployment.
Lower tax burdens can also lower unemployment, but a far more cost
effective fiscal approach is to increase spending on active labour-market
policies. The bottom line is that high unemployment in Western Europe
has been the result of self-inflicted dysfunctional macroeconomic policy.
European policy makers adopted a course of disinflation, high real inter-
est rates, and slower growth that raised unemployment. Moreover, they
all adopted this course at the same time, thereby generating a wave of
trade based spillovers that generated a continent wide macroeconomic
slowdown and further raised unemployment.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
4 Introduction

Philip Arestis, Michelle Baddeley and Malcolm Sawyer examine the


proposition that investment and capital stock can contribute substan-
tially to the determination of the rate of unemployment and real activ-
ity, and thereby to the determination of the NAIRU (Non-Accelerating
Inflation Rate of Unemployment) in a range of EMU countries. In their

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
model, the effect of demand on unemployment and wages is empha-
sized. The mechanism works through changes in capacity utilization,
which themselves depend on economic activity and the stock of capital.
It is argued here that negative demand shocks affect employment and
investment adversely. When shocks reverse, unemployment may not
fall to previous levels, due to insufficient capital equipment. Their
empirical results show that the general rise in the unemployment rate in
the countries examined over the past thirty years was to a large and
significant extent due to insufficient investment, leading to a lower
(than otherwise) capital stock.
Based on a post-Keynesian approach concerning the relationship
between wages, prices and employment, the chapter by Eckhard Hein,
Thorsten Schulten and Achim Truger studies the extent to which unit
labour cost trends have been responsible for disinflation and deflation-
ary tendencies in Germany and the European Monetary Union (EMU).
According to their view, excessive nominal wage moderation, particu-
larly in Germany, has not only caused deflationary risks but also
contributed to falling labour income shares. Weakened trade union bar-
gaining power in a period of sustained mass unemployment and the
integration of trade unions into a new 'competitive corporatism' aiming
at wage moderation in order to improve international price competi-
tiveness are seen as major reasons for this development. It is concluded
that the excessive wage restraint in Germany not only exacerbates
stagnation and deflationary tendencies in Germany but might also
mean deflationary risks for the other EMU countries.
The influence of unemployment, productivity growth and institu-
tions on real wage trends in Italy is analysed in the chapter by Sergio
Levrero and Antonella Stirati which is based on a Classical view on the
determination of wages. In this view wage trends are affected by a set of
historical and current circumstances that can be broadly classified as
labour-market conditions, the degree of organization of the bargaining
parties involved, and broader economic and political/institutional
factors affecting bargaining power. The empirical analysis shows that
the primary factor affecting real wage trends in Italy in the 1970-2000
period appears to have been unemployment. Broader economic and
political/institutional factors have had an accelerating influence in

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Eckhard Hein, Arne Heise and Achim Truger 5

particular periods. It is argued that the unprecedented vulnerability of


real wages to the exogenous increase in prices after the lira devaluation
in 1992 was caused by institutional changes, namely the new wage-
setting procedures agreed upon in the same year, combined with the
unions' inability to ensure the development of firm-level bargaining.

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
The chapter by Heather Boushey and Christian Weller deals with
rising inequality of household income in the USA and the macro-
economic consequences. They find that the distribution of national
income between capital and labour has become more unequal as has
the distribution within labour. At the same time that inequality rose,
consumer debt and household economic distress grew, too. The evi-
dence on a positive link between rising inequality and innovation is
not supported by the data. The data either suggest no connection or a
potential negative link. In comparison, the link between growing
inequality and aggregate demand is somewhat ambiguous. One way
to clarify this ambiguity is the possibility that debt has increased and
that it has increased more among low-income households, particularly
in the form of credit cards and non-bank credit. The authors find that
credit-card debt is especially sensitive to changes in inequality and
that rising inequality may thus have contributed to rising personal
bankruptcies.
In a panel-data analysis Gerardo Angeles-Castro examines the effects
of economic liberalization on income distribution for a set of develop-
ing countries. His results are in keeping with the theoretical foundations
and expectations that have supported the global liberalization process
only insofar that low inflation, fiscal discipline, larger employment and
domestic efficiency can benefit income distribution. On the other hand,
the results undermine other aspects of these theoretical foundations and
expectations, since the benefits of trade on income distribution are
weak, foreign direct investment (FDI) worsens inequality, and the
export-led growth strategy and the expansion of employment based on
the primary sector do not improve income distribution. As for the set of
policies underlined in the second stage of the economic liberalization
process, domestic efficiency can help to reduce the adverse effect of FDI
on income distribution, while it can also help to obtain some benefits
from trade liberalization in income distribution. Moreover, a stronger
state is important to decrease inequality, and the set of socio-political
norms enclosed in the post-Washington Consensus approach improves
income distribution. The study also suggests that further supranational
mechanisms, beyond the scope of the state, are required to socialize the
flow of trade and investment.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
6 Introduction

From a Classical point of view, Sergio Cesaratto explores the


economic impact of ageing on the labour market and on the future of
social security organized along pay-as-you-go (PAYG) lines. Assuming as
a working hypothesis the persistence of employment at the present
levels (with output growth resting on productivity growth), it is argued

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
that the falling supply of domestic labour population is not necessarily
a constraint on output growth, insofar as it is compensated by higher
participation rates (including later retirement), technical progress,
immigration and, perhaps, some recovery in fertility. But in any
case, there will be a rise in the PAYG burden on social output which,
however, can be covered if wages and/or profits share part of the pro-
ductivity gains with the retirees. The crucial question the chapter asks is,
whether the additional sources of labour supply are such as to preserve
an industrial reserve army and if capitalism can live without a signifi-
cant labour reserve army that keeps the whip of competition on the
workers. It is in this sense that a shrinking labour supply may affect
employment and growth by inducing deflationary policy choices in
order to ensure the existence of an industrial reserve army.
The relation between distribution, investment and employment is
empirically investigated by Ozlem Onaran and Engelbert Stockhammer.
They use a Kaleckian-Post-Keynesian macroeconomic model building
on the approach by Bhaduri and Marglin (1990). This model is able to
generate profit- and wage-led regimes within an aggregate demand frame-
work. Applying a structural vector autoregression (SVAR) approach to three
developed and two developing countries the authors confirm the
Keynesian/Kaleckian hypotheses about the labour market: accumulation
and capacity utilization/growth have a strong impact on employment.
Goods-market variables have a strong effect on unemployment and the
economy is driven by investment expenditures. The neoclassical hypothe-
ses of the labour market are not validated. There is little evidence of
employment reacting to wages, and no evidence for factor substitution.
The findings also suggest that productivity growth does play an important
role. It is not distributionally neutral and causes unemployment. However,
no statistically significant effect of the 'profit share' was found on invest-
ment and growth in developed countries, as well as in one of the develop-
ing country cases. There is hence no general result in terms of the
distinction between wage-led and profit-led growth regimes for the inves-
tigated set of countries.
In a theoretical paper, Thomas I. Palley expands the Cambridge-
Kaleckian model of income distribution and growth to include a labour-
market class-conflict channel that is distinct from the product-market

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Eckhard Hein, Arne Heise and Achim Truger 7

competition channel. The labour-market channel works through


conflict over distribution of the wage-bill, whereas product market com-
petition impacts the mark-up and the profit share. The distinction
between wage share and wage-bill distribution has important theoretical
and policy implications. At the theoretical level, it explains why

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
economies can simultaneously exhibit both 'wage-led' and 'profit-led'
characteristics. Redistributing the wage-bill to workers always raises
aggregate demand and economic activity by raising consumption.
However, lowering the profit share can retard activity by lowering
investment spending in a 'profit-led' regime. At the policy level, it sug-
gests that progressive policy should focus on altering the distribution of
the wage-bill, rather than the profit share as has been the traditional
focus. Lastly, the dual wage-led-profit-led characteristic helps make
sense of developments in the US economy over the last three decades.
In the final chapter Amit Bhaduri develops a synthesis of the
Kaleckian model of distribution and growth, as presented in Bhaduri
and Marglin (1990), which relies on constant distribution and variable
capacity utilization in order to allow for an adjustment of savings to
investment, and the Kaldorian model assuming full utilisation of capac-
ity and a variable distribution of income. If neither the profit share nor
the rate of capacity utilisation is treated as exogenously fixed, the behav-
iour and the stability of the system will depend on relative speeds of
adjustment. Bhaduri shows that neither the profit-led nor the wage-led
course of expansion is unambiguously stable.
The papers in this volume were presented at the 8th Workshop of the
Research Network 'Alternative Conceptions of Macroeconomic Policies
under the Conditions of Unemployment, Globalization and High Public
Debt' on 'Wages, Distribution and Growth' which took place in Berlin,
29-30 October 2004. Further papers have been published in a book by
Metropolis Verlag (Marburg). We would like to thank the contributors
to this volume for their cooperation, and the participants in the confer-
ence for the stimulating discussions. Special thanks go to Barbara
Schnieders for assistance in the editing process and to the Hans Boeckler
Foundation for organizational and financial support for the workshop
and the publications.

Notes
1 On New Classical and New Keynesian models see Snowdon, Vane and
Wynarczyk (1994). On the 'new consensus' models see Arestis and Sawyer
(2003), Clarida, Gali and Gertler (1999) and Meyer (2001).
2 See Aghion, Caroli and Garcia-Penalosa (1999) or Alesina and Perotti (1996).

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
8 Introduction

References
Aghion, P., Caroli, E. and Garcia-Penalosa, C. (1999) 'Inequality and Economic
Growth: The Perspective of the New Growth Theories', Journal of Economic
Literature, 37: 1615-60.
Alesina, A. and Perotti, R. (1996) 'Income Distribution, Political Instability, and
Investment', European Economic Review, 40: 1203-28.

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Arestis, P. and Sawyer, M. (2003) ' "New consensus", New Keynesianism and
the Economics of the "Third Way" ', in E. Hein, A. Heise and A. Truger (eds),
Neu-Keynesianismus - der neue wirtschaftspolitische Mainstream? (Marburg:
Metropolis).
Atkinson, A.B. (1997) 'Bringing Income Distribution In from the Cold', The
Economic Journal, 107: 297-321.
Atkinson, A.B. (2000) 'The Changing Distribution of Income: Evidence and
Explanations', German Economic Review, 1: 3-18.
Bhaduri, A. and Marglin, S. (1990) 'Unemployment and the Real Wage: The
Economic Basis for Contesting Political Ideologies', Cambridge Journal of
Economics, 14: 375-93.
Clarida, R.; Gali, J. and Gertler, M. (1999) 'The Science of Monetary Policy: A New
Keynesian Perspective', Journal of Economic Literature, 37: 1661-707.
Meyer, L.H. (2001) 'Does Money Matter?', Federal Reserve Bank of St. Louis Review,
83(5): 1-15.
Snowdon, B., Vane, H. and Wynarczyk, P. (1994) A Modern Guide to
Macroeconomics. An Introduction to Competing Schools of Thought (Cheltenham:
Edward Elgar).

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
1
Labour-Market Flexibility and
Economic Expansion

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Amit Bhaduri

Introduction

The major economies of Western Europe have been facing almost


chronic unemployment, often hovering around double digits, for nearly
two decades now. This is especially true of Germany, which acted as the
economic powerhouse in the consolidation of the European Union. It is
all the more paradoxical that advanced market democracies can live
with high unemployment placing relatively little emphasis on this prob-
lem, except perhaps at times of elections. A deep change in the climate
of opinion has occurred in striking contrast to the post-Second World
War years, when the pursuit of full employment was the agreed policy
objective of almost all shades of political opinion. History is seldom
mono-causal; several causal factors and processes usually coincide at a
particular juncture of time to bring about such a dramatic change in the
climate of opinion.
To a significant extent, this change of opinion began as a reaction
against the Keynesian style of demand management in pursuit of high
employment, which gradually lost ground for several interrelated rea-
sons. First, the Keynesian argument was almost self-consciously set in
the context of an economy closed to international trade and capital
flows. The intention might have been to emphasize the importance of
domestically oriented economic policies for fighting unemployment.
The disastrous consequences of 'beggar- my- neighbour' policies of try-
ing to export unemployment through competitive devaluation of the
national currencies during the interwar period (until the stand-still
agreement of 1936) was still fresh in memory. At the same time, the
prestige of the City as the financial centre of the world was in jeopardy,
and its views propounding the virtues of 'sound finance' was in ruins.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
10 Labour Market Flexibility and Economic Expansion

It was rather natural in that context to look inward for a solution to the
problem of unemployment which relied more on domestic industries,
and less on 'high finance' (Bhaduri and Steindl 1985).
Secondly, although the contest between the competing systems of
capitalism and socialism was most visible in the arms race during the

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
cold war, its ideological dimension was essentially economic. The socialist
system appeared capable of providing full employment through deliber-
ate state policies, although much of it was neither satisfactory to the
employees nor socially useful. (A common joke of the time in these
countries was, 'they pretend to pay, we pretend to work'.) The capitalist
market economies, where the level of employment depended largely on
the decision of private business, had the opposite problem. The employ-
ment it provided had to be necessarily gainful, that is profitable for the
private employer, even if private and social gain differed. But the level of
economic activity was prone to cyclical fluctuations, at times resulting
in severe unemployment. Given this visible difference between the two
systems, even initiatives like the Marshall Plan tended to be influenced,
at least partly, by the economic competition between the two systems
(Hobsbawm 1994). It was around this time that the welfare state also
found wider political acceptance. Its theoretical rationale derived from
Keynesian demand-management policies, whereas the rising real wage
with near-full employment, leading to rapidly improving standards of
living for the working population under this new style of economic
governance in most Western democracies, posed a counter-challenge to
the socialist ideology.
Thirdly, the very success of demand management, high employment
and rising mass consumption, which had ushered in a 'golden age' of
welfare capitalism through rapidly expanding domestic markets for
nearly a quarter century, came to be troubled by its own contradictions
(Marglin and Schor 1990). Years of high employment had reduced the
fear of job-loss for the workers leading to higher wage claims. It was in
this context that the experiences of the two major oil price shocks of the
1970s served as almost the watershed years. They made it clear that the
burden of such shocks could no longer be passed on easily to the work-
ers. The model of cooperative capitalism of the welfare state was giving
away to the model of conflictive capitalism, in which conflict over the
distribution of income tended to manifest itself through inflationary or
stagflationary price rise (Rowthorn 1977). Even more problematically,
the fiscal policy of the state itself got entangled in this distributive con-
flict, as both workers and their employers tended to pass on the addi-
tional tax burden to one another (Bhaduri 1986: ch. 6). Understandably,

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Amit Bhaduri 11

targeting inflation rather than unemployment became the new focus


of policies under these changed circumstances, and new economic
doctrines, at time reviving old ideas that had been pushed aside by
the success of Keynesian economic policies, returned in academic cir-
cles and policy discussions under the broad heading of Monetarism.

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Ironically, Monetarism reinvented the Marxian idea that a 'reserve
army' of labour is needed to keep a check on the real wage. Kalecki
(1943, reprinted in 1971) had already foreseen towards the end of the
Second World War that 'political trade cycles' would be imposed delib-
erately, particularly in the name of sound finance (read, no deficit
financing) to inflict unemployment from time to time on the workers,
in order to keep control over the workers. These ideas returned in the
orthodox monetarist framework of economic theory as the 'natural rate'
of unemployment (Friedman 1968), or as the non-accelerating inflation
rate of unemployment, NAIRU (Layard, Nickell and Jackman 1991).
They all had a common theme insofar as they propounded the view
that, keeping inflation and inflationary expectations under control
requires accepting a certain, it may even be a fairly high, rate of unem-
ployment. In particular, it requires giving up demand management poli-
cies intended at keeping the rate of unemployment lower than that
'natural' or NAIRU rate. Since deficit financing by the government had
been the most potent instrument used for demand management, unsur-
prisingly it came under special attack. The doctrine of the virtues of a
balanced budget, and the evils of a fiscal deficit by a self-seeking gov-
ernment in the name of 'public choice theory' were propounded as
general truths, applicable to almost all countries under all circumstances.
The wider process of globalization that gathered almost irresistible
momentum with the deregulation of national capital markets since
the mid-1970s in OECD countries also contributed to this change in
the climate of opinion to a significant extent. The greater economic
opening-up in goods and services trade has meant an increase in the
relative importance of the foreign or external market compared to the
domestic or internal market. It encourages countries to stimulate
demand through export surplus, rather than through demand manage-
ment by the fiscal policies of the government. As a result, each country
tries to be more price competitive compared to its rivals, by cutting unit
cost through wage 'flexibility' on the one hand, and raising labour pro-
ductivity on the other. Yet there is an obvious 'fallacy of composition' in
this strategy; because all countries cannot achieve export surplus at the
same time as the export surplus of some must be matched by the import
surplus of others. But even for any particular country, which does

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
12 Labour Market Flexibility and Economic Expansion

manage to achieve an export surplus in this zero-sum game, such


policies may turn out to be counter-productive if the contraction in the
size of its internal market more than outweighs the expansion of its
external market.
The danger of over-contraction of domestic demand from such

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
policies is serious in many countries. Wage restraint depresses consump-
tion by the working people, while labour productivity growth brought
about by the corporations through the downsizing of the domestic
labour force reinforces this depressive effect. Analytically, this may be
identified as the wage- or consumption-led regime in which the depres-
sive effect on consumption of wage restraint and downsizing outweigh
its possible stimulating effect on investment and export (Bhaduri and
Marglin 1990). And yet, this danger of a sharp decline in domestic
demand tends to be overlooked in policy discussions, not due to igno-
rance, but for reasons that have become intrinsic to the current phase of
globalization.
Successive waves of deregulation and liberalization of capital markets
since the mid-1970s has unleashed private trade in foreign exchange on
an historically unprecedented level. In comparison to its daily volume
of some 1.2 trillion dollars (BIS 2001, 2002), the total foreign exchange
reserve of all the central banks is insignificantly small, while foreign
trade and investment together do not account for even 4 per cent of it.
The economic policies of national governments seem overwhelmed by
the power of the financial markets. In particular, this has generally
meant that expansionary fiscal policies for fighting serious unem-
ployment through budget deficits are generally not favoured options
by financial markets; and therefore by national governments. Similarly,
easy money policies with lower interest rates are not favoured either,
insofar as they stimulate consumption and investment expenditure (for
example hire-purchase, inventory accumulation, housing and so on),
and tend to strain the current account of the balance of payments
through lower borrowing cost. Moreover, financial markets often read
the lowering of the interest rate as signalling the government's intention
of embarking on expansionary fiscal policies. The fear that private capi-
tal would vote with its feet in the foreign exchange market by crossing
borders on a massive scale has crippled fiscal and monetary policies
required to fight growing unemployment.
However, there has been a significant difference in this respect
between the United States and Europe. Due to the reserve currency
status of the dollar, the United States manages to run growing current-
account deficits and a relatively free national demand-management

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Amit Bhaduri 13

policy, while the rest of the world, especially Japan and Germany as
important export surplus countries, continues to finance the US. import
surplus in the form of capital inflow into the USA. This is no longer
because of the international 'store of value' property of the US dollar;
these export surplus countries are almost locked into this arrangement,

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
because economically they have come to depend too heavily for their
export surplus on the US market for maintaining demand, while politi-
cally they perceive their national defence capability as being dependent
on the USA.
The consequence has been a peculiar paradox about national employ-
ment policies which is seldom commented upon. The current strength
of the euro has been bought at the cost of restrictive fiscal policies
imposed by the Maastricht 'stability pact', and even more restrictive
monetary policies pursued by an independent European Central Bank.
But if the euro becomes an international 'store of value' through its
strength replacing largely the dollar, capital inflows to the USA would be
sharply reduced, reducing in turn the export market, particularly for the
large export surplus nations like Germany or Japan. As a result, they
would face an even more serious unemployment problem. Thus, with
the present international financial arrangements and policies, the para-
dox is that the euro and the yen seem doomed to remain 'strong' cur-
rencies without becoming alternative international stores of wealth.
Consequently, the privilege to run large trade deficits continuously by
issuing paper liabilities accepted by the rest of the world - the privilege
based on the international store of value status of a currency - which
England had before the First World War, and the USA has enjoyed since
the Second World War, needs to be denied to these countries. The wis-
dom of following restrictive monetary and fiscal policies that inflict
high unemployment also needs to be examined in the light of this
paradoxical international financial arrangement.

Theories and counter-theories for economic policies

The macroeconomic perspective justifying restrictive monetary and


fiscal policies is neoliberal in its essence. Since its general philosophy is
to roll back the economic role of the state, it is natural that it would view
adequate employment creation as the job of the private sector, and
not of the state. At best, the role of the state is seen as creating the 'right'
climate for the private sector by increasing its profitability, so that it
offers sufficient employment. In this respect it has three distinguishing
characteristics contrary to the Keynesian theory.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
14 Labour Market Flexibility and Economic Expansion

First, unlike Keynes and Kalecki, who identified the lack of effective
demand in the product market as the central cause of unemployment,
these theories locate the central cause of unemployment in the labour
market. In terms of policy, this means demand management in the
product market becomes less important than correcting the malfunc-

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
tioning of the labour market. This deemphasizes the role of demand
management on the one hand, and brings to the forefront the impor-
tance of labour-market 'flexibility' on the other.
Secondly, even the branch of theory relatively sympathetic to the
Keynesian view (for example the neo-Keynesians), tends to make a
distinction between the 'short run' and the 'long run'. It is claimed that
there may be insufficiency of demand in the product market in the short
run, but somehow the demand problem is resolved in the long run by
the market. This is starkly visible in almost all versions of neo-classical
long-run growth theory which assumes away the demand problem as a
short-run phenomenon.
Thirdly, and perhaps most basic to this neo-liberal way of thinking, is
its foundation of 'methodological individualism'. It uses some proce-
dure of optimization by the individual agent as the central organizing
principle of macroeconomic theory. Approaching macroeconomic prob-
lems exclusively in this way has many serious ramifications, and insofar
as the unemployment problem is concerned perhaps its most serious
consequence is to blur the distinction between 'voluntary' and 'invol-
untary' unemployment. Thus, all unemployment, even mass unem-
ployment, begins to look voluntary in this framework, because it has
to be explained through some optimizing decision by the individual
worker, but not by a failure of the system which requires intervention.
This then becomes attributable to the imperfect functioning of the price
mechanism for giving wrong signals to the worker. In some cases gov-
ernment interventions, mostly related to the welfare state, are supposed
to distort the signals carried by wages. An important case in point is the
criticism levelled against social safety nets like unemployment benefits
provided by the government. A widely prevalent view, propagated by
the media as well as some academic economists, is that it raises the
'reservation wage' so much that the worker chooses voluntarily to
be unemployed. Another version with a similar thrust is to claim that
the worker refuses to take up employment at the prevailing wage, either
because (s)he is misinformed and spends all his or her time in search of
a better job, or waits until the wage rate becomes higher. For a non-
economist with some common sense these theories lie only a little dis-
tance away from the view that some are unemployed because they are

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Amit Bhaduri 15

lazy or unreasonably ambitious who do not know their 'market worth'.


In this framework either the jobless worker becomes responsible for his
or her own situation; or the blame is put on government intervention.
Neo-liberalism relying on the market rather than the government to
deal with problems like unemployment has a different kind of advan-

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
tage which is seldom recognized. Nothing in economic theory specifies
how long it might take to reach 'equilibrium', through the market
mechanism, even under the most idealized circumstances of perfect
competition, when the competitive equilibrium has all the desirable
properties of so-called Pareto-optimality by the 'fundamental theorem'
of welfare economics. In democracies, the government and the politi-
cians in power remain accountable to their people at regular time inter-
vals through the elections, but the market mechanism has no such
definite time horizon for showing results. And, this ambiguity helps in
sustaining the market ideology insofar as it can always be claimed
that, given 'sufficient' time and 'sufficiently' wide-ranging pro-market
reforms, the desired results would materialize without having to specify
how long is sufficiently long. The market mechanism, like a dictator,
can always promise without actually delivering!
The virtues of sticking adamantly to pro-market policies as the solu-
tion to the unemployment problem get further strengthened by the role
of the media. The enormous power wielded by electronic media and tel-
evision images in shaping public opinion was noticed by perceptive
commentators even in the very early stage of current globalization (see
for example McLuhan 1960). It helps in spreading a sort of a popular
culture in economic policy, policies that can be easily comprehended by
'practical' men and women. In this respect, most appealing is the anal-
ogy with the individual. Since spending beyond means is considered
bad for the individual, a budget deficit of the government is also con-
sidered bad; because the demand for apples can be raised by lowering its
price, the demand for labour can also be raised by lowering the wage
rate; hard work helps, so a corporate manager helps the economy
by downsizing its labour force. Underlying these pronouncements,
popularized by the media, is the foundation of 'methodological individ-
ualism'. Unfortunately, politicians become victims of it, even if some
of them understand better, because political expediency often demands
that they do not swim against the current of popular 'understanding'.
Nevertheless, this popular culture of viewing macroeconomic policies
exclusively through the neo-liberal glass of methodological individual-
ism goes against the very logic of having macroeconomics as a distinct
branch of enquiry. The latter is justified, precisely in those situations

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
16 Labour Market Flexibility and Economic Expansion

when the analogy with the individual is misleading. And, serious


unemployment is perhaps the most important example of such a situa-
tion, as the Keynesian theory demonstrated. The common mistake in all
arguments based on methodological individualism is to fall into the
fallacy of composition. Thus, cutting wages may improve the profit mar-

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
gin for one firm, but the total volume of sales in the economy is likely
to suffer. Downsizing the labour force may reduce unit cost for one firm,
but when followed by many firms it may reduce total employment and
the volume of sales in the economy. Even export surplus for one coun-
try must be balanced by the import surplus of another; so the strategy of
export-led expansion cannot work for all. These familiar examples show
why the argument that holds for the individual part may fail to hold for
the whole system. Keynesian macroeconomic theory is based precisely
on this premise, and the macroeconomic policies to fight unemploy-
ment follows from the fundamental insight that, unlike in the case of
the individual, the macroeconomy is characterized by a circular flow
from expenditure to income. Expenditure by creating demand may
determine output and employment in situations of serious unemploy-
ment and excess capacity.

Policies for high employment

It is against this intellectual background that we need to reconsider


policies for attaining and sustaining higher employment. However, at
the very outset of this policy discussion it is worth pointing out that
many different types of projects may be devised for generating employ-
ment. By and large these projects would be country-specific, dependent
on geography, stage of development, degree of openness, and so on. The
purpose of economic theory in this context is not to list a set of such
projects irrespective of the country-specificity, but to indicate the direc-
tion in which an employment-generating and employment-sustaining
programme can become feasible.
The problem of servicing debt exists, but it is far from clear why a
government cannot take recourse to further borrowing to service its
debt, so long as its higher spending is effective in raising the employ-
ment level and the rate of growth. The well-known result that so long as
the growth rate exceeds the interest rate in real terms, the debt to
income ratio tends to stabilize, might be invoked in this context to serve
as a guide. On the whole it seems misplaced to argue that a growing
debt-servicing burden necessarily strains the credibility of a government
under all circumstances. A government's economic credibility might

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Amit Bhaduri 17

indeed suffer more due to persisting high unemployment and low


growth. Therefore, the real issue to be faced is whether government
spending can be effective in raising productive employment opportuni-
ties and growth prospects.
This is indeed the crux of the problem, and needs to be decomposed

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
in two parts. First, to the extent the economy faces insufficiency of
demand with excess capacity, the traditional Keynesian remedy should
be reasonably effective provided the composition of demand, governed
mostly by income elasticities, is broadly in line with the pattern of
excess capacities in various industries. In that case, until it really begins
to overheat the labour market, even the financial markets might not
react unfavourably. Anyway, with considerable excess capacity, inaction
is no solution in the face of growing unemployment.
The second part of the problem is to devise suitable programmes for
government spending. Going against received wisdom, it needs to be
pointed out that extensive labour training schemes of various sort can
not be of much help. Because, although it might improve labour pro-
ductivity, without strong demand-side policies such training would only
raise the bar for entering the labour market. Thus, while the better
trained persons would probably have a higher chance of getting a job,
macroeconomically it would not be a very effective policy for the entire
labour force. It would be rather like changing the positions of the can-
didates in a long queue for a fixed number of jobs, without being able to
shorten the length of the queue.
A better policy with respect to training might be to have a govern-
ment and private sector forum to agree on the types of training needed
in view of both the composition of demand and cost-effectiveness. The
training programmes would be partly financed by the government on
the understanding that the private firms would simultaneously provide
on-the-job training as the rest of the financing.
Second, instead of large centralized projects, it might be better to
decentralize them. This would at least have the advantage of identifying
the regions in which government spending has been relatively more
effective in fighting unemployment, and would thus introduce an
element of competition into the allocation of public funds.
Finally, in the post-industrial societies with aging populations, there
might be considerable scope for creating innovative services for the
aged. Note that better care of the aged is a demand that cannot be easily
saturated, and therefore provides a longer-term perspective on the direc-
tions of training and public spending. For financially sustaining these
programmes, the government and beneficiaries might share the costs

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
18 Labour Market Flexibility and Economic Expansion

through voluntary schemes of using part of pension and insurance


funds. It is the increased quality and productivity of labour in these so-
called non-tradable areas of services, directed at the domestic rather
than the external market, in which the future for a better quality of life
with higher employment in post-industrial societies may well lie.

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Beyond this suggested long-term path of development of the service
industry, the problem of labour-market flexibility also has another long-
run aspect which is overlooked in current policy discussions. Higher
labour productivity tends to reduce prices by reducing unit costs in the
more competitive industries. For given money wages, this raises the
product wage rates in these industries, normally implying an increase in
the overall real wage rate. Therefore, unless labour productivity rises
more or less in line with increasing real wage, the share of wages in
income would tend to increase over time. However, the observed rela-
tive stability of the wage share over the longer run suggests that produc-
tivity growth itself adjusts to rising real wages, just as the wage claims of
workers have to adjust to the growth in labour productivity. In the long
run, therefore, a two-way relationship seems exist between the growth
rate in wage and in productivity which might be a central driving force
of 'endogenous' growth, making technology appear as 'neutral' to dis-
tribution in successful market economies (Bhaduri 2006). In this virtu-
ous circle of positive feedbacks or cumulative causation, growth in
productivity tends to raise the real wage rate by lowering unit costs and
prices due to competition among firms, while rising real wages induces
firms to increase productivity for maintaining their profit margins.
Viewed from this longer-run perspective, artificially imposed wage
restraints might well weaken this complex positive feedback mechanism
between productivity and real-wage, generating economywide increas-
ing returns. This process has been recognized as a potent dynamic
force in the successful development of capitalism, at least since Adam
Smith identified division of labour as a main source of the wealth of
nations. To ignore this dynamic force propelling long-term growth in
the name of labour-market flexibility might turn out to be counter-
productive, and not a sign of wisdom for the successful management of
capitalism.

References
Bank for International Settlements (2001) Central Bank Survey of Foreign Exchange
and Derivatives Market Activity in April 2001, preliminary global data, press
release (9 October).
Bank for International Settlements (2002) 'Quarterly Review of International
Banking and Financial Markets Development', Statistical Annex (March 2002).

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Amit Bhaduri 19

Bhaduri, A. (1986) Macroeconomics: The Dynamics of Commodity Production


(London: Palgrave Macmillan).
Bhaduri, A. (2006) 'Endogenous Economic Growth: A New Approach', Cambridge
Journal of Economics (forthcoming).
Bhaduri, A. and Marglin, S. (1990) 'Unemployment and the Real Wage. The
Economic Basis of Contesting Political Ideologies', Cambridge Journal of

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Economics, 14: 375-95.
Bhaduri, A. and Steindl, J. (1985) 'The Rise of Monetarism as a Social Doctrine', in
P. Arestis and T. Skouras (eds) Post-Keynesian Economic Theory (Sussex:
Weatsheaf).
Friedman, M. (1968) 'The Role of Monetary Policy', American Economic Review,
58: 1-17.
Hobsbawm, E. (1994) The Age of Extremes. A History of the World, 1914-1991 (New
York: Pantheon Books).
Kalecki, M. (1971) 'Political Aspects of Full Employment' [1943], in M. Kalecki,
Selected Essays in the Dynamics of the Capitalist Economy (Cambridge: Cambridge
University Press).
Layard, R., Nickell, S. and Jackman, R. (1991) Unemployment, Macroeconomic
Performance and the Labour Market (Oxford: Oxford University Press).
Marglin, S. and Schor, J. (eds) (1990) The Golden Age of Capitalism (Oxford: Oxford
University Press).
McLuhan, M. (1960) Explorations in Communications (Ontario: University of
Toronto Press).
Rowthorn, R. (1977) 'Conflict, Inflation and Money', Cambridge Journal of
Economics, 1: 215-39.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
2
The Causes of High
Unemployment: Labour-Market

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Sclerosis v. Macroeconomic Policy*
Thomas L Palley

The debate over the causes of high u n e m p l o y m e n t

The economies of Western Europe remain afflicted by high and


intractable rates of unemployment, with the European Union unem-
ployment rate standing at 8.1 per cent as of August 2004, while the
unemployment rate in the 12-country euro-zone area was 9.0 per cent.
Moreover, European unemployment rates have been stuck at these
levels for several years. In stark contrast, the US unemployment
rate touched a 30-year low of 3.9 per cent in September 2000, rose to
6.3 per cent in the most recent recession, and was back down to 5.4 per cent
in September 2004. This divergence in performance has opened a great
debate that is being especially vigorously pursued in Germany.
One side claims that Europe's unemployment is the result of rigid scle-
rotic labour markets that have rendered it incapable of adjusting to
technological advance and change in the international economy.
Unemployment benefits are too generous and their duration too long,
unions are too strong, and employee protections are such that firms are
discouraged from hiring workers. This contrasts with the US economy,
which is marked by flexible dynamic labour markets that have adjusted
to these developments and used them to create new jobs.

* This chapter was originally published in Challenging the Market: The Struggle to
Regulate Work and Income, Jim Stanford and Leah E Vosko (eds), McGill-Queen's
University Press, 2004. My thanks go to McGill-Queen's University Press for
permission to reprint this material.

20

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Thomas I. Palley 21

The other side claims that Europe's unemployment problem is


significantly attributable to bad macroeconomic policy (Baker and Schmitt
1998; Palley 1998, 1999; Solow 1994), which has resulted from mistaken
adherence to the theory of the natural rate of unemployment. This has
prompted policy-makers to adopt austere macroeconomic policies aimed at

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
reducing inflation regardless of the unemployment cost or the underlying
cause of inflation. Currency market concerns have also played an adverse
role. In the 1980s and 1990s the persistent threat of currency speculation
induced European governments to raise rates so as to defend their curren-
cies and guard against imported inflation. Subsequently, arrangements
leading up to the introduction of the euro aggravated the problem as coun-
tries were forced to satisfy strict fiscal convergence criteria that called for
policies of austerity irrespective of economic conditions. The net result has
been a persistent contractionary bias to policy, and policy has also exhib-
ited insensitivity to the state of the business cycle. Contrastingly, US macro-
economic policy has been relatively flexible and counter-cyclical (Palley
1999). Both the US budget deficit and Federal Reserve monetary policy have
exhibited clear counter-cyclical fluctuation, and in the recession of 1990-91
the Fed lowered short-term nominal rates such that the real rate was zero.
Moreover, this sharp difference in macroeconomic policy persists
through to the present. Thus, when the last recession began in 2001 the
US Federal Reserve slashed its interest rate in the first six months of the
year by over 40 per cent, lowering rates from 6.5 per cent in January to
3.75 per cent in June. Side-by-side, fiscal policy shifted into expansion-
ary mode with a significant tax cut, albeit one tilted toward the affluent.
And these policy shifts were undertaken despite the fact that the unem-
ployment rate was still below 4.5 per cent and the inflation rate had
actually increased above 3 per cent. Moreover, when robust recovery
failed to take hold the Fed lowered rates further to 1.25 per cent in
November 2002 and has only recently begun a marginal reversal of
these rate reductions. In stark contrast, the European Central Bank was
slow to lower interest rates as global recession set in, and the ECB has
kept its interest rate above that of the Fed despite the fact that Europe's
unemployment is significantly higher than that of the United States.
The bottom line is that the monetary and fiscal policy have displayed
greater ease and counter-cyclicality in the USA than in Europe.
These two accounts of unemployment have enormously different
policy implications. If the labour-market flexibility hypothesis is correct,
Europe needs to adopt the US model and introduce policies of labour-
market flexibility that render wages downwardly flexible, reduce
employee protections, and reduce unemployment benefits and other

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
22 The Causes of High Unemployment

social protections. If the macroeconomic policy hypothesis is correct,


Europe should adopt expansionary macroeconomic policies predicated
on lower real interest rates. It also needs to adopt policy rules that ensure
monetary and fiscal policy move in counter-cyclical fashion.
The outcome of this controversy is not only germane to the countries

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
of the OECD, it is also relevant for the developing economies which
are marked by a parallel debate. Thus, the Washington Consensus -
which represents the developing-country analogue of the Euro-sclerosis
hypothesis - maintains that employment and output growth in the
developing world depends upon the adoption of policies of labour-
market flexibility. Supporters of this consensus therefore counsel devel-
oping countries to resist calls for international labour standards since
such standards would promote worker rights of freedom of association
and collective bargaining.
These observations spotlight the critical nature of the debate over the
causes of unemployment. How it is resolved promises to have deep last-
ing impacts on policy in both developed and developing countries. This
chapter provides some new evidence on the relative contributions of
macroeconomic factors and labour market institutions to unemploy-
ment in the OECD. The principal empirical innovation here is that we
combine macroeconomic time series variables that capture the stance of
macroeconomic policy with microeconomic labour-market institution
variables. This means that the effects of both labour-market institutions
and macroeconomic policy are taken into account in statistical exami-
nations of the causes of higher unemployment. The principal findings
are that macroeconomic policy variables consistently and robustly mat-
ter for the evolution of country unemployment rates, and that macro-
economic policy affects unemployment rates in the manner expected.
High real interest rates and slow growth raise unemployment, as does a
slowdown in export growth. With regard to the microeconomic labour
market variables the evidence is more problematic. Unemployment-
benefit duration and union density are both consistently insignificant.
The level of wage bargaining coordination and the extent of union cov-
erage matter consistently, but they need not raise unemployment if they
are appropriately paired with other policies. Finally, the significance of
other microeconomic variables (employment protection, unemployment-
insurance wage replacement rate, tax burden) is unstable and not
robust to changes in specification.
These findings lead to the conclusion that high unemployment in
Western Europe is principally the result of self-inflicted dysfunctional
macroeconomic policy. European policy-makers adopted a course of
disinflation, high real interest rates and slower growth that raised

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Thomas I. Palley 23

unemployment. Moreover, they all adopted this course at the same


time, thereby generating a wave of trade-based cross-country spillovers
that generated a continent-wide macroeconomic funk and further raised
unemployment.
A last important finding is that real interest rates have tended to be

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
systematically higher in countries with high union density despite the
lack of any evidence that high union density raises inflation. This sug-
gests that central banks have systematically raised interest rates in
countries with high union density.

Evidence o n the causes of OECD u n e m p l o y m e n t

As noted above, the principal contribution of the current study is to fully


incorporate both microeconomic labour-market institution variables and
macroeconomic variables, thereby allowing for a proper assessment of
the relative contributions of labour market institutions and macroeco-
nomic policy to higher unemployment. This section describes the data,
the empirical model and the empirical findings.

Data
Data for the labour market institutional variables were supplied by
Nickell, and are described in his widely cited study on the impact of
labour market rigidities on unemployment (Nickell 1997). Data for the
macroeconomic variables were drawn from the annex tables in the 1999
OECD Economic Outlook, the World Bank CD-ROM, and IMF CD-ROM.1
Further details regarding the data are provided in the Appendix.
The macroeconomic variables are annual time series data so that
there is one observation per year for each variable for each country.
Contrastingly, the labour-market institution variables correspond to
fixed effects. For each type of labour-market institution a six-year average
measure was constructed for each country covering the periods 1983-88
and 1989-94. Thus, for each institution in each country there are two
observations - one for the period 1983-88, and the other for the period
1989-94. Lastly, data for the following countries was used in the
regressions: Austria, Belgium, Denmark, Finland, France, Germany,
Holland, Ireland, Italy, Norway, Portugal, Spain, Sweden, Switzerland, the
UK, Australia, New Zealand, Japan, the USA and Canada (see Table 2.1).
Table 2.1 shows average macro data and labour-market institution
data for these 20 countries for the periods 1983-88 and 1989-94. The
macroeconomic data are average standardized unemployment rate (%),
average real GDP growth (%), average inflation rate (%), average short-
term nominal interest rate (%), and average short-term real interest

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
2014-03-15
Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect -
T-H oo CM L o r o c M r o ^ r o ^ L O T f H H N ON N
O ^ 1-1 0 \ H H O O N N 0 0 r 0 O H 0 0 X ^ O N
V O ^ I ^ ^ N N V O ^ I O V O V O N ^ N ^ V O I O N ^ H
CM ON oo i^- LO c \ CM N N N N H i ^ N r f i vo " ^ ro
10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim Truger

L O C M r o ^ L o r ^ ^ r o ^ L O ^ O r o ^ ^ o LO ^ H IT) ^
r^ oo r^ 00 fO N C O V O C O C O ^ N ' t f O
o o \ o v O H v o o \ r o q N i o o N ^ a v q
00 N C) H 00' N K o \ H ^ ' o \ d r o N H ^ d o \ i o
M >0 H M N \ D N r^ CM CM ^f T--I CM LO r^ oo CM oo
vo ro CM ro Tf r^ io T t ^ i n H f O N N N O \ ^ r O O
CM to r>! C* O CM* ON i n H ^ i n N r o ' ^ o o ' d r o d c o
00 CO N rO N l O N N r O i ^ C O N N K N O O N r O
r^. no oq H L O oq r^ uo t o o \ ^ # p p p vq t^. p ^ T-H oq
ro'ro*cM"ro'cM*ro*cM*no*cM'cM'Lo*cMro*ro'a*LO*voro'Lo'ro
N r o uo N N i o r o r^ r^ CM CM r^ CM oo uo
v o o o o o ^ o ^ c s i i i o ^ v o o o c s i r ^ i - i T f uo r f N VO ^
N N r O ^ i O U O U O H H U O O O H O N N O O ^ N ' t P O
LO ^O ^ O \ ^ O > N ^ ' t a H 0 0 H 0 0 0 0 0 0 t^.vo
vq r->. f 0 ^ f S j v O O \ X , t N ( O O N i J ; i o q r O N q q
( S J N H H H O H N ' M ' U O CM H CM csi CM o T-H H CM
COON ^ H H O \ i f ) K N r O 0 0 H i O N N \ O M O 0 0
"* p CM r^ ^ H no ^ # r^# r^ ON oq ro T-H CM uo © o\ vq
TJ" CM CM ^ CM ro CM CM CM CM CM ro H ro ro ro CM CM* ro ro
LO r o r o H T—i LO vo VOLOLOXVOrHVOrH VO
O ^ O O O O u O O N f O O r O
( ^ r o N Q\ CO 0 \ O
"^ ^ °°. ^ °. ^ P ^ °) ^
L o v d ^ O N C M o o L O L o o ^ L o c M o o v d
^O X rH 00 T-H T-iOOOOOOOOOOrorov£>r-HLOvO
i * CO r o 0> ' t ^O N u 0 ^ 0 0 , O , t N \ O H K C 0 N H
00 CM O ON vo LO Os VO X LO 00 CM T f CM l \ O CM o
T—1 T-H CM
as

New Zeal
.2

Portugal
1-1

Norway

Sweden
as T3 as £ T3

Spain
C/5
a ^ as
N
•M
3 3 D fl <u
g
*H
a3 CD
TS
O """"3 OH
< < C Q U Q
a in
as
|—i OO
2014-03-15
Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect -
25
<v
TJH CM CM 00 ^ 00 ON >/) I N . oq ^ oq oq LO NO H NO
©* NO T-H LO* T-H CM ON CM* L O ON 00 L O "3* NO T-H T-H CM' NO ON L O
p
^ ^ L o r o r ^ i ^ f O N ^ r o M ^ i O f O H C O N r O H .5
O
U
10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim Truger

r>. CM vq ON IN. ON oq ro r# r# H ro ^ LO en T-H vq oq


^rHoouo'ro©ro'^'dro*rf*X*©'vdvdxT-Hod^'oN
H H i O i O f O N N H f O f O l O ^ N l O i O ^ H O O N ^ r H
0s ^
ON
CM roi/}©TfLOC^CM©OOCMT-HOOO\rovoiN.r-H
T-H
<* T-HT-HT-H T-HCM T-HT-HT-HT-H
oo
ON
X ' T-H
so
^3
•S
+-1
s
^
00
°p 00
ro«r)©TfLOONCM©XCMrHOOONro^OtN,T-H

s ON
T-HT-HT-H T-HCM T-HT-HT-IT-H
"-N ^
ON
V5 1 LO Lq LO LO 00 LO CM LO
3 ON
O
£> 00
ON
C N i f N l P O ^ t M ^ d d ^ t H d p O H H T f d
*43
Q
?*.
3
^+j 00
°p
^0 | , ! l , !
LO LO
O O ^ H
LO LO LO CM
O f O H H
«
^>
cq
oo
ON
T t - ^ ^ L o c M ^ r O ' ^ t ^
^ 0 0 0 > O o o N r o O N O O O i O u o o O O c o O
r o L O M D L o o > ^ L o v o r N r o c M ' O r o N O N O i N . o o i N s r o L O
O vN O O v O O ^ > N f O O O O O C, O i O vO O O O v O O
ro OMD OONi NLOVorNLOCM OroNO OOOXtNsroLO
T3
as
2 -S
*ZJ (1) ±2
<<
2014-03-15
Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect -
r O N O r f C M ^ L O ^ L o ^ c M ^ ^ r O N O ^ r o ^ ^ C M C M
10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim Truger

r O N O ^ c M N O N O ^ L O T f c M r o ^ r O N O ^ t r o ^ O ^ C M c M
r o r o r o c M r o r o r o r o r o r o r o c M C M r o r o r o r o c M C M T - H
r o r o r o c M r o r o r o r o r o r o r o c M c M r o r o r o r o c M r O T - H
r o o o T t ' L o O N d o o L o v d o ^ o ^ N O ^ o o ^ O N O O N d
T-H r-1 T-H CM T-H r H r H L O
H is en NO ^ CM ON C M r H T ^ T t t L O O N C M L O oq ON
^t* 00 © MD © 00 IN." rO
I N . i s oq I N . r o LO oq L O r o ON r o oq NO NO CM i s NO oq oq
o 6 r o o ^ c M ^ O N r o r o N d ^ c M N O ^ o 6 l s ^ © o d © r o
c M L O ^ ^ ^ v O N O L O L o r o N O r o r o ^ r O L O i s r o ^ T f
oq LO so oq oq NO oq NO r o ^ CM T-H r o ON LO T-H ON v q NO
© T^' ts' ^ 06 ON CM CM ON r o I S r o LO ON r o © 00 © rjn CM'
r O L O ^ r o ^ L O N O L O L o r O L o r o r o ^ r O L O N O ^ ^ " ^
a
X3 as bO
H H M-H OS G rrt
if c o
aS >^ as
< <« u s £ a i ^ s ^ ^ PH 00 V3 UO
Thomas I. Palley 27

rate (%) defined as the difference between the short-term nominal


interest rate and inflation rate. The labour-market institution data are
the wage replacement rate (%), unemployment benefit duration (years),
an index of employment protections (1-20), union density (%), the
overall tax rate (%), index of spending on active labour market pro-

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
grammes, index of union wage coverage (1-3), and an index of coordi-
nation in wage bargaining (2-6).
There are a number of features worth remarking. First, with regard to
unemployment rates the USA is in the bottom half of the distribution,
but many countries also had lower unemployment rates over the period
1983-94. Second, inflation rates were much higher in Europe in the first
half of the sample, but they fell significantly in the second half. Third,
average short-term real interest rates have been very much lower in the
USA than in the rest of the world. These two features, disinflation and
higher real interest rates in Europe, are indicative of the more difficult
macroeconomic conditions that have confronted European economies.
With regard to the labour-market institution data, the USA clearly has
the most laissez-faire markets as indicated by its low wage replacement
rate, low benefit duration, low level of employment protections, low
union density, low tax rate, low spending on active labour market pro-
grammes, low union wage coverage, and low level of coordination of
wage bargaining. Many of these features carry over to the Anglo-Saxon
economies of the UK, Canada and New Zealand - particularly regarding
employment protection, tax rates, labour market spending, union wage
coverage, and coordination of wage bargaining. However, despite hav-
ing deregulated labour markets, Australia, Canada, New Zealand and the
UK all tended to have unemployment rates that clustered at the top of
the distribution.

An empirical model
The empirical model used to estimate the causes of unemployment is
given by:

UNEMP;-t = a0 + ajUNEMP^! + a2UNEMP;-f_2 + a3EMPROTM


+ a4REPRATE/t + a5BENDUR;-1 + a6UNIONDEN;-1
+ a7UNIONCOV;-1 + a8COORD;-1 + a9TAXRATE;-1
+ a10ALMPROG/t + anDINFLATEM + a^REALINT,^
+ a13GDPGROW/#t + a14GDPGROW; ^ + a15EUROPEN/t
+ a16CANUS;-t + a17IREDUM;-f + a18SPADUMM + uj>t
(2.1)

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
28 The Causes of High Unemployment

The definition of variables is as follows:

UNEMP, t standardized unemployment rate in country / in


year t
EMPPROT; t index of employment protection (1-20) in country /.

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
REPRATE; t unemployment insurance wage replacement rate (%)
in country /.
BENDURy1 benefit duration (years) in country /.
UNIONDEN;-1 union density (%) in country /.
UNIONCOV, t extent that union wage coverage extends to
non-union workers (1 = less than 25%, 2 = 25-70%,
3 = greater than 70%) in country /.
COORD; t extent of coordination (index = 2-6) of wage bar-
gaining amongst unions and employers in country /.
TAXRATE; t total tax rate (sum of average payroll, income and
consumption tax rates) in country /.
ALMPROG;1 measure of active labour market policy (spending per
unemployed worker as a per cent of the potential
output per worker) in country /.
DINFLATE; t change in the CPI inflation rate (%) in country / in
year t.
REALINT; t real interest rate (%) in country / in year t
GDPGROW,- t rate of real GDP growth (%) in country / in year t.
EUROPEN, t measure of exposure of individual European countries
to intra-European trade in year t (0 for non-European
countries).
CANUS7/1 measure of exposure of the Canadian economy to
trade with the USA in year r (0 for all countries except
Canada).
IREDUM dummy variable capturing effects specific to unem-
ployment in Ireland.
SPADUM dummy variable capturing effects specific to unem-
ployment in Spain.
u7/1 residual.

The variables can be broken down into three sets. The microeconomic
labour market variables consist of EMPROT, REPRATE, BENDUR,
UNIONDEN, UNIONCOV, TAXRATE, COORD and ALMPROG. The
macroeconomic variables consist of DINFLATE, REALINT, GDPGROW,
EUROPEN and CANUS. The significance of the EUROPEN and CANUS
variables is discussed below, and the construction of these variables is
described in the Appendix. Lastly, the IREDUM and SPADUM capture

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Thomas I. Palley 29

fixed effects that are specific to Ireland and Spain. Both of these
economies had much higher unemployment rates over the sample
period, reflecting their position as quasi-developing economies on the
periphery of the European Union.2
With regard to the specification of the empirical model, the inclusion

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
of two lags of the unemployment rate as explanatory variables reflects
the fact that adjustment in labour markets tends to be gradual as it takes
time for workers to reallocate and for firms to create new jobs. As a result,
all economies exhibit considerable persistence to unemployment shocks.
With regard to the macroeconomic variables, the effects of macro-
economic policy and conditions is captured by the change in the infla-
tion rate (reduced inflation corresponds to tight policy), the level of real
interest rates (high real rates corresponds to tight policy), and the rate of
real GDP growth. The inclusion of the economic openness variables,
EUROPEN and CANUS, is especially important. These variables capture
the cross-country Keynesian multiplier effects that operate through
international trade. Within the European economy it is critical to con-
trol for cross-country growth spillover effects owing to the high degree
of economic integration among countries. Just as an explanation of
unemployment in Texas would need to control for developments in the
US economy, so too a similar logic holds in Europe where countries are
very integrated economically with each other. This same logic also holds
for Canada which is highly integrated into the US economy. Such effects
are noticeably absent from other studies examining the causes of higher
European unemployment (Nickell 1997; Blanchard and Wolfers 2000).
The Appendix describes the construction of the EUROPEN and CANUS
variables.

Empirical findings
Table 2.2 reports several regression estimates of equation (2.1) based on
a two-stage least-squares analysis for the sample period 1983-94.3
Column (a) reports the benchmark regression equation which contains
just two lags of country unemployment rates. In this model there are
assumed to be absolutely no differences between countries, and both
micro institutions and macro policy and performance factors are absent.
Despite this, the model has considerable explanatory power as measured
by the adjusted R2 which indicates the goodness of fit of the model with
the data. This highlights the fact that persistence in unemployment
rates is a feature common to all economies, and it should therefore be
incorporated in all models of unemployment.
Column (b) expands the benchmark equation to include labour-
market institution variables. The coefficients of the replacement rate

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
2014-03-15
Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect -
OS ^ 00 CM ON IN.
is 00 CM IS CM ON
© LO T-H CM © ^
© ^ Tf ^ H ^
10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim Truger

OS is © CM is 00
H T-H CM 00 CM 00
T-H 00 T-H H © Tf
w
<S I i
^ ^ CM ^ vo ^ IS ^ IS ^ IS ^-N IN. ^ © ^ VO ^ CM ^
ro © ^ NO ON FN © OS o © o rN © is T^H Tf 00 00 T-H ro
^ en T-H is CM CM © ^ © CM © TJH LO NO CM CM © ON
© H T-H" d d LO* © ©^ © d © TfH © rH d ro © ro © rH
i^
ro^vis^^ro^-vOs^-^ro^^^O^-^ro ^•v LO ^ X ^ ro ^
OOrorOT-HrH©cMMDT-HONT-HCM© vo T-H os Os ^h © ©
en T-H CM © r o r o p O s p N O © ^ ©
LO ^ ^q CM CM © ^#
©T-HrHCM'©LO©T-H©CM'©d© © d CM* © ro © d
w w w w
cs I i w w
I I I I
* * * * * _ ! ^ CM ^ ^ T-H ^ r o
L O ^ r o ^ O O ^ r o ^ i s ^ T-H © T-H 00 NO ^ NO O r o
O S T - H I S V 0 ^ V O C M L O © X * © © uo r o T f CM i s © x
^OCMCMCMrqOS©LO©lsNO © d d © CM © C M © ©
©CMT-Hro'©LO©T-H©r-H^
w
£i i i ^ ^ i
^ M ^ ^ ^ LO LO
is" o
NO vo NO o LO
LO T-H CM LO ro CM
r o OS T-H TjH os © CM LO
T-H r q T-H NO rq © CM © OS O LO © 00 LO 00 LO 00 ©
© T-H is © CM* © © © © © o © CM o £S©1 1
O ^
o o ^ i o ^ t s ^ ^ ^ r o ^ ON , - v 00 ^ LO ^ P 0 ^ ©
© C M O O O I S O O N D V O T - H C M
ISLOISONT-HlSrO^fT—I ^ © LO © rq rq 00 ^ T-H p
en os ^ ON vq r-j p oq o
CM d d © T-H © T-H © ^' ©
© d T - H l S © T - H © T - H ©
w w w w , !
OS „ CM ^ ^ ^
CM OS NO OS
ro Os LO Tf LO CM
d CM T-H oo © d
- si T
CM >
Q PJ
O O Q
s fee Q O u O
P-)
PJ O X
Z PH o
u
Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect -
ALMPROG -0.014 -0.029*** -0.002 -0.006 -0.019***
(-1.56) (-3.23) (-0.230) (-0.81) (-2.73)
DINFLATE -0.084*** -0.077*** -0.064*** -0.080*** -0.0861
(-3.54) (-3.27) (-2.86) (-3.51) (-3.58)
REALINT(-l) 0.070*** 0.061*** 0.046*** 0.046*** 0.0401
(3.85) (3.39) (2.70) (2.97) (2.41)
GDPGROW -0.236*** -0.245*** -0.225*** -0.257*** -0.274'
(-10.23) (-9.30) (-9.01) (-10.65) (-10.78)
GDPGROW ( - 1 ) -0.055* -0.67*** -0.103*** -0.079** -0.040
(-1.68) (-2.08) (-3.31) (-2.48) (-1.21)
EUROPEN -0.227*** -0.269*** -0.167** -0.135'
(-2.62) (-2.97) (-2.51) (-2.00)
CANUS -0.318 -0.031 -0.057 -0.288
(-1.49) (-0.15) (-0.29) (-1.44)

2014-03-15
IREDUM 1.028 1.332*** 1.196***
(3.07) (4.84) (4.87)
SPADUM 2.440 1.536*** 1.229***
(5.74) (4.49) (4.56)
Adj.i? 2 0.956 0.959 0.964 0.977 0.978 0.981 0.979 0.976
S.E. 0.930 0.896 0.840 0.664 0.655 0.615 0.641 0.682
N= 240 240 239 239 239 239 239 239

Notes: T-statistics are in parenthesis.


*** significant at the 1% level; ** significant at the 5% level; * significant at the 10% level.
Source: See Appendix.

OO

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim Truger
32 The Causes of High Unemployment

(REPRATE) and the overall tax rate (TAXRATE) are both statistically
significant at the 5 per cent level, and both raise unemployment. The
extent of wage bargaining coordination (COORD) is significant at the
1 per cent level and lowers unemployment. Employment protections
(EMPPROT) and union coverage (UNIONCOV) are both significant at

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
the 10 per cent level, and both raise unemployment. Lastly, benefit
duration (BENDUR), union density (UDEN) and active labour market
programmes (ALMPROG) are all insignificant at the 10 per cent level.
Regression (c) further expands the model to include country-specific
effects for Ireland (IREDUM) and Spain (SPADUM). Both of these country-
specific effects are statistically significant and positive at the 1 per cent
level, and their inclusion dramatically changes the significance of other
explanatory variables. Now, both the employment protection index and
replacement rate become statistically insignificant at the 10 per cent
level, but union density and spending on active labour market pro-
grammes now both become statistically significant at the 1 per cent
level. This is indicative of coefficient instability among the microeco-
nomic labour-market institution variables.
Regression (d) begins the task of incorporating macroeconomic vari-
ables by including the change in inflation (DINFLATE), the lagged real
interest rate (REALINT(-l)), and the current and lagged real output
growth (GDPGROW and GDPGROW(-l)). Inclusion of these variables
dramatically improves the quality of the regression estimate as indicated
by the jump in the adjusted R2 statistic and the fall in the standard error
of the regression equation. The variables DINFLATE, REALINT(-l) and
GDPGROW are all statistically significant at the 1 per cent level, while
GDPGROW(-l) is statistically significant at the 10 per cent level. All are
signed in a manner consistent with conventional understandings of the
impact of macroeconomic policy on unemployment. Disinflation raises
unemployment, as do higher real interest rates.4 Faster growth lowers
unemployment.
As regards the labour-market institution variables, inclusion of the
macro variables causes major changes. First, the union density coeffi-
cient becomes insignificant - a feature which is examined in greater
detail below. Second, the statistical significance and magnitude of the
tax coefficient falls considerably. Third, the variables EMPROT and
REPRATE now become significant at the 1 per cent level, which is indica-
tive of coefficient instability surrounding these variables. This too is fur-
ther discussed below.
Column (e) further augments the model by including the interna-
tional trade exposure variables EUROPEN and CANUS. The former is

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Thomas I. Palley 33

significant at the 1 per cent level, while the latter is only significant at
the 14 per cent level. Both are negatively signed. The large magnitude
and clear statistical significance of the coefficient of EUROPEN indicates
the importance of interdependence amongst European economies.5 The
signs of the other macro variables (DINFLATE, REALINT(-l), GDPGROW,

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
GDPGROW(-l)) remain unchanged, and all coefficients are statistically
significant at the 1 per cent level. The coefficients of these macroeconomic
variables are robust and stable with regard to changed model specification,
lending confidence to their importance for explaining unemployment.
With regard to the micro variables, BENDUR, UNIONDEN, TAXRATE and
ALMPROG are all statistically insignificant at the 10 per cent level.
REPRATE, UNIONCOV and COORD are statistically significant at the 1 per
cent level, while EMPROT is significant at the 6 per cent level.
Column (f) reports the findings for the full model that includes all
labour-market institution variables, all macroeconomic variables, and
the Ireland and Spain country fixed effect variables. The coefficients of
all the macroeconomic variables remain same-signed, and all except the
CANUS variable are statistically significant at the 1 per cent level. The
Ireland and Spain country fixed effects are also both positive and statis-
tically significant at the 1 per cent level. However, most of the labour-
market institution variables are statistically insignificant. This holds for
the employment protection index (EMPROT), the wage replacement
rate (REPRATE), benefit duration (BENDUR) and union density (UDEN).
The full model therefore suggests that none of these variables matter for
explaining unemployment. Spending on active labour market pro-
grammes (ALMPROG) is statistically significant at the 1 per cent level,
and it contributes to lower unemployment. The overall tax rate
(TAXRATE) is also significant at the 10 per cent level, and higher taxes
rates raise unemployment.
This fully specified model helps understand a number of features.
First, both union wage coverage (UNIONCOV) and the extent of coordi-
nation in wage bargaining (COORD) are significant at the 1 per cent
level, and both variables are statistically significant in most of the other
regressions. These variables have opposite signs with the former being
positive, while the latter is negative. The UNIONCOV variable takes
values of 1-3, while the COORD variable takes values of 2-6. These two
variables are strongly positively correlated, having a correlation coeffi-
cient of 0.49, and a regression of COORD on UNIONCOV yields:

UNIONCOV = 1.897 + 0.197 COORD, Adj. R2 = 0.235 (2.2)


(25.53) (11.11)

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
34 The Causes of High Unemployment

Figures in parentheses are t-statistics. Thus, if COORD = 2, the predicted


value of UNIONCOV = 2.3: if COORD = 6, the predicted value of
UNIONCOV = 3.1. The two variables therefore co-move strongly and
systematically, and should best be thought of as a 'system of industrial
relations'. Coordination in wage bargaining lowers unemployment,

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
while union wage coverage raises it. As long as these two features are
appropriately paired, there need be no negative impact on unemploy-
ment. 6 Problems only emerge when there is extensive union wage
coverage unaccompanied by wage bargaining coordination. This find-
ing is consistent with the work of Calmfors and Drifill (1988).7
Second, the inclusion of the country dummy variables for Ireland and
Spain causes the EMPROT and REPRATE variables to become statistically
insignificant. Inspection of Table 2.1 shows that Spain had extremely
high unemployment rates, and it also had an extremely high level of
employment protection and a very high replacement rate. The statistical
significance of these two institutional variables therefore appears to be
entirely related to Spain - that is, it is an outlier phenomenon. When a
Spain dummy is included, they become insignificant. This phenome-
non holds for both the full model (compare regressions (e) and (f)) and
for the restricted model with just labour-market institution variables
(compare regressions (b) and (c)). The policy implication is that existing
employment protections and wage replacement rates have not been
a contributory factor to European unemployment, except perhaps in
Spain.
Finally, regressions (g) and (h) provide estimates of the restricted
model with just macroeconomic variables. These regressions are pre-
sented to give additional evidence of the significance of macroeconomic
factors for explaining unemployment. The coefficients of the macro
variables continue to be highly statistically significant, they remain
same-signed and their magnitude is little changed. At the same time, the
restricted regressions with just macro variables perform very well in
terms of adjusted R2 and standard error of the regression, being only
marginally worse than the full model which includes the labour-market
institutional variables.

Further interpreting the results


In sum, the regressions reported in Table 2.2 provide clear evidence of
the importance of macroeconomic forces for unemployment. This con-
clusion is robust to empirical specification. Based on regression (f), per-
manently lowering the inflation rate by 1 percentage point increases
unemployment by 0.4 percentage points. An increase in real interest

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Thomas I. Palley 35

rates of 1 percentage point increases unemployment by 0.3 percentage


points. Lowering the rate of real output growth by 1 percentage point
increases unemployment by 2.1 percentage points. This latter finding
implies an Okun coefficient of one-half. This is fully in accordance with
existing estimates of the Okun coefficient (Palley 1993), which lends

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
additional support to the results presented. For a European country that
exports 20 per cent of its GDP, a 1 percentage point increase in the
growth rate of other European economies results in a 0.35 percentage
point decrease in that country's unemployment rate.
Regression (f) indicates that a one hundred basis point increase in the
real interest rate increases the unemployment rate by 0.4 percentage
points. During the period 1989-94, the US real interest rate averaged
1.80 per cent. In Canada the real interest rate averaged 4.7 per cent,
raising the Canadian unemployment rate relative to the USA by 1.2 per-
centage points. In Germany the real interest rate averaged 4.03 per cent,
raising the German unemployment rate relative to the USA by 0.9 per-
centage points. In France it averaged 6.12 per cent, raising the French
unemployment rate relative to the USA by 1.7 percentage points.
Finally, in the Scandinavian countries (Denmark, Finland, Norway and
Sweden) the real interest rate averaged 5.87 per cent, raising the
Scandinavian unemployment rate relative to the USA by 1.6 percentage
points.
With regard to the labour-market institution variables, the regressions
provide no evidence that lowering employment protections, replace-
ment rates or benefit durations will reduce unemployment. Nor will
lowering union density. However cutting taxes can. A 10 percentage
point reduction in tax burdens (which in most countries means reduc-
ing taxes by about one-fifth) lowers the unemployment rate by only
0.8 percentage points. Increasing spending on active labour-market poli-
cies has a much bigger bang for buck. Increasing active labour market
spending per unemployed worker by an amount equal to 10 per cent of
potential output per worker lowers the unemployment rate by 1.2 per-
centage points. Spending on job training and placement programmes
for the unemployed is therefore a more cost-effective fiscal approach to
the problem of unemployment.
Finally, if properly paired, coordination of wage bargaining and union
wage coverage can actually lower unemployment. If both of these insti-
tutions were maximally implemented (UNIONCOV = 3, COORD = 6),
then the unemployment rate would be reduced by 0.6 percentage
points. Of course if there is widespread union wage coverage and no
coordinated wage bargaining, then unemployment rates will rise.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
36 The Causes of High Unemployment

Quantifying the causes of changed u n e m p l o y m e n t rates

The previous section reported several estimates of structural equations


determining the causes of unemployment. This section changes the
focus of analysis and uses these estimates to identify what caused coun-

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
try unemployment rates to change between 1983 and 1994. For this pur-
pose, the preferred equation is that reported in column (f) of Table 2.2.
According to this equation, the contribution of microeconomic institu-
tional factors to unemployment is given by:

0.007 EMPPROT,-1 + 0.007 REPRATE,-1


+ 0.007 BENDUR; t + 0.007 UNIONDEN,t
MICRO,-1 = /0.154 (2.3)
+ 0.541 UNIONCOV; t - 0.286 COORD, \
+ 0.012 TAXRATE, t - 0.019 ALMPROG,'t

The change in unemployment rates attributable to changes in labour


market institutional factors is then computed as:

DMICRO = MICROy - M I C R O y (2.4)

Table 2.3 reports an analysis that decomposes the actual change in


country unemployment rates between 1983 and 1994 into those parts
attributable to micro and macro factors. Columns (1) and (2) detail the
country unemployment rates ruling in 1983 and 1994 respectively, while
column (3) reports the change in country unemployment rates between
1983 and 1994. Column (4) then details that part of the change attribut-
able to changed microeconomic institutional settings. Finally, column
(5) details the change in unemployment rates attributable to macro-
economic factors. This macroeconomic component is computed as:

DMACRO = DUNEMP - DMICRO (2.5)

The table has a number of interesting and important findings. First,


DMICRO is negative in 13 out of 20 countries, indicating that most
countries have pursued policies designed to flexibilize labour markets.
Second, DMACRO is positive in 15 out of 20 countries, indicating that
over the period 1983-94 most countries experienced negative macro-
economic outcomes that raised unemployment rates. Third, in Europe's
three biggest economies (France, Germany, Italy) these negative macro
shocks were quantitatively large. In all three economies the direction of
microeconomic change was such that unemployment should have

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Thomas I. Palley 37

Table 2.3 Decomposition of the causes of changing unemployment rates into


factors due to changing labour-market institution (DMICRO) and macro-
economic slowdown (DMACRO)

d) (2) (3) (4) (5)


UNEMP UNEMP DUNEMP DMICRO DMACRO

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Country 1983 (%) 1994 (%) (2) minus (1) 1983-94 1983-94

Austria 3.8 3.8 0.0 -0.79 0.79


Belgium 11.1 10.0 -1.1 -0.51 -0.59
Denmark 10.3 8.2 -2.1 -0.26 -1.84
Finland 6.1 16.8 10.7 2.29 13.17
France 8.1 12.3 4.2 -0.34 4.54
Germany 6.9 8.4 1.5 -1.61 3.11
Holland 9.7 7.1 -2.6 -0.89 -1.71
Ireland 14.0 14.3 0.3 -0.69 0.99
Italy 7.7 11.4 3.7 -1.68 5.38
Norway 3.5 5.5 2.0 -0.77 2.77
Portugal 7.8 7.0 -0.8 -1.69 0.89
Spain 17.5 24.1 6.6 -0.64 7.24
Sweden 3.7 9.4 5.7 0.23 5.47
Switzerland 0.9 3.8 2.9 1.63 1.27
UK 11.1 9.6 -1.5 3.80 2.3
Australia 10.0 9.7 -0.3 -0.38 0.80
New Zealand 5.8 8.1 2.3 0.40 1.90
Canada 11.9 10.4 -1.5 0.20 -1.70
USA 9.6 6.1 -3.5 0.05 -3.45
Japan 2.7 2.9 0.2 0.25 0.05

Source: See Appendix.

fallen, but instead unemployment rose owing to the large scale of


macroeconomic shocks. Fourth, US unemployment fell by 3.5 percent-
age points, and this decline was almost entirely due to favourable
macroeconomic conditions. Fifth, Finland, Sweden and Spain all suf-
fered large increases in unemployment rates, and in all three instances
the increase was almost entirely due to unfavourable macroeconomic
forces. Sixth, Belgium, Denmark and Holland experienced reductions in
unemployment rates, and favourable macroeconomic developments
explain more than 50 per cent of the decline in these three cases.
In sum, almost all of the decline in the USA is attributable to positive
macro forces, while almost all of the increase in Europe is attributable to
negative macro forces. And in those few instances in Europe where
unemployment rates fell, macro forces were again primarily responsible.
The policy implication is clear. Rather than engaging in a wholesale re-
making of labour-market institutions and arrangements, European

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
38 The Causes of High Unemployment

governments should correct the dysfunctions that have driven macro


economic policy over the last two decades. That these dysfunctions
remain in place is clearly evident in the different policy responses of
the Federal Reserve and the European Central Bank to the economic
slowdown of 2001.

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
The political e c o n o m y of monetary policy: have
central bankers waged war o n unions?

Both Nickell (1997) and Scarpetta (1995) report that union density has
a statistically significant positive impact on unemployment rates. This
contrasts sharply with the findings reported in the current study, and it
is worth enquiring as to the source of this difference.
One clue to this difference comes from a comparison of regressions
(2.c) and (2.d) in Table 2.2 in which the inclusion of macroeconomic
variables appears to undo the unemployment impact of union density.
In the regressions reported by Nickell (1997) the only macro variable is
the change in inflation rates. This suggests that the effect may be related
to the inclusion of real interest rates.
To test this hypothesis, union density was regressed against the aver-
age measure of country real interest rates shown in Table 2.1. The result-
ing pooled least-squares regression, with and without a time dummy to
capture changes in financial market conditions across the periods 1983-88
and 1989-94, is given by:

REALINT, = 3.505 + 0.032 UNIONDEN,


(5.33) (2.27)
2
Adj.R = 0.096
N = 40 (2.6a)

REALINT. = 2.943 + 0.035 UNIONDEN, + 0.923 TIMEDUMMY


(4.12) (2.49) (1.77)

2
Adj.R = 0.145
N = 40 (2.6b)

Figures in parentheses are t-statistics. In both regressions, with and


without a dummy for the period 1989-94, the coefficient of UNIONDEN
is positive and statistically significant at the 5 per cent level. According
to these regressions, a 10 per cent absolute increase in the union density
rate therefore results in a 0.3 percentage point increase in the real
interest rate.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Thomas I. Palley 39

To test for robustness, this union interest rate hypothesis was also
tested in a simple time-series model with one lag of the real interest rate
and with a union density fixed effect for the period 1983-94. The result-
ing regression was:

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
REALINT, t = 1.822 + 0.483 REALINT.-1_! + 0.018 UNIONDEN,
(4.82) (10.11) ' (2.36)
2
Adj.JR = 0.333
N = 238 (2.7)

The union density coefficient is positive and statistically significant at


the 5 per cent level, and according to equation (2.7) the net effect of a
10 per cent increase in union density is to raise real interest rates by 0.35
per cent. This almost exactly matches the results from regressions (2.6a)
and (2.6b).
Prima facie, regressions (2.6a), (2.6b) and (2.7) suggest that central
bankers may have raised rates in economies where union density is
high. However, it is possible that union density causes inflation and cen-
tral banks were really aiming to lower inflation. To test this hypothesis
union density was regressed on average consumer inflation (as reported
in Table 2.1) yielding:

INFLATION, = 3.854 + 0.023 UNIONDEN,


(3.25) (0.89)
2
Ad).R =- 0.005
N = 40 (2.8a)
INFLATION, = 4.839 + 0.019 UNIONDEN, - 1.633 TIMEDUMMY
(3.76) (0.74) (-1.74)
2
Ad].R = 0.045
N = 40 (2.8b)

Both regressions (2.8a) and (2.8b), with and without a dummy for the
period 1989-94, show that there is no statistical relation between inflation
and union density. This conclusion was further tested by a simple auto-
regressive pooled time-series model of country inflation rates given by:

INFLATION,, = 0.514 + 0.776 INFLATION,t_t + 0.001 UNIONDEN,


(1.59) (26.69) ' (0.14)

2
Ad].R = - 0 . 0 0 5
N= 240 (2.9)

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
40 The Causes of High Unemployment

Again, figures in parentheses are t-statistics, and again union density has
no explanatory power regarding inflation. In sum, regressions (2.8a),
(2.8b) and (2.9) all show no statistical relation between inflation and
union density. This challenges the defence that real interest rates were
higher in countries with higher union density because unions cause

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
inflation. Instead, it looks as if central banks systematically raised inter-
est rates in countries with high union density. This is fully consistent
with the idea that monetary policy is a site of class conflict and policy
has largely been captured by interests antagonistic to unions (Palley
1997).

Labour-market institutions and income inequality

Thus far, attention has focused on the impact of labour-market institu-


tions on unemployment. This section shifts the focus to income
inequality, and Table 2.4 presents some data on patterns of cross-country
income distribution.
Simple pooled regressions using the data in Tables 2.1 and 2.4, appro-
priately matched by year, yield the following relations between income
distribution, union density and employment protection:

= 2
(In) 06.04 - 0.581 UDEN,
\50/,- (27.78) (-3.60)
2
Adj.i? = 0.40
N = 19 (2.10a)

f £ ) = 188.03 - 0.612 EMPPROT,


50// (22.43) (-3.60)
Ad].R2 = - 0.02
N = 19 (2.10b)

§£ J = 212.35 - 0.582 UDEN, - 0.621 EMPPROT,


50// (22.95) (-3.64) (-1.12)
Ad).R2 = 0.41
N=19 (2.10c)

(IH) = 4-539 " °- 0 2 3 UDEN,


\10// (12.56) (-2.91)
Adj.£2 = 0.29
N=19 (2.10d)

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Thomas I. Palley 41

Table 2.4 Measures of country income distributions

Ratio

Country Year 10th/50th 90th/50th 90th/10th Gini

Australia 1989 .45 1.93 4.3 0.308

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Austria 1987 .56 1.87 3.34
Belgium 1992 .58 1.63 2.79 0.23
Canada 1991 .47 1.83 3.9 0.285
Denmark 1992 .54 1.55 2.86 0.239
Finland 1991 .57 1.58 2.75 0.223
France 1984 .55 1.93 3.48 0.294
Germany 1989 .54 1.72 3.21 0.263
Holland 1991 .57 1.73 3.05 0.249
Ireland 1987 .50 2.09 4.18 0.328
Italy 1991 .56 1.76 3.14 0.255
Japan 1992 .46 1.92 4.17 0.315
N.Zealand 1987-88 .54 1.87 3.46
Norway 1991 .56 1.58 2.8 0.233
Portugal
Spain 1990 .49 1.98 4.04 0.306
Sweden 1992 .57 1.59 2.78 0.229
Switzerland 1982 .54 1.85 3.43 0.311
UK 1991 .44 2.06 4.67 0.335
USA 1991 .36 2.08 5.78 0.343

Source: Mishel, Bernstein and Schmitt (2000).

| ^ l = 4.196 - 0.060 EMPPROT,


/ (12.16) (- 2.03)
2
Adj.R1 = 0.15
N= 19 (2.10e)

§-"« j (13.11)
0.023 UDEN, - 0.061 EMPPROT,
(-3.38) (-2.59)
2
Adj.ff 0.47
N = 19 (2.10f)

10
45.987 + 0.141 UDEN,
50/,- (i3.ii) (-3.38)

Adj.R2 = 0.18
N= 19 (2.10g)

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
42 The Causes of High Unemployment

s o ) = 46.265 + 0.552 EMPPROT,


W U
// (19.29) (2.66)
2
Adj.jR = 0.25
N= 19 (2.10h)

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
0.141 UDEN, + 0.554 EMPPROT,
50// (13.74) (2.78) (3.16)
2
Adj.£ = 0.47
N= 19 (2.10i)

GINI, = 0.332 - 0.001 UDEN,


(19.07) (-3.38)

Adj.£ 2 = 0.39
N= 17 (2.10J)

GINI, = 0.307 - 0.003 EMPPROT,


(15.06) (-1.56)

Adj. JR2 = 0.08


(2.10k)
N= 17

GINI, = 0.360 0.001 UDEN, - 0.003 EMPPROT,


(17.59) -3.74) (-2.12)

Adj.R2 = 0.51
N=17 (2.101)

where

(90/50), = ratio of income of 90th percentile household to median


household in country /.
(90/10),- = ratio of income of 90th percentile household to 10th
percentile in country /.
(10/50), = ratio of income of 10th percentile household to median
household in country /.
GINI = Gini coefficient in country /'.
UDEN,- = u n i o n density (%) in country /.
EMPPROT, = index of employment protections (1-20) in c o u n t r y ; .

Figures in parentheses are r-statistics. The regressions show that b o t h


union density and employment protections work unambiguously to
reduce income inequality. Union density reduces income inequality
between the top and the middle (90/50), and the top and the b o t t o m

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Thomas I. Palley 43

(90/10).8 Employment protections reduce inequality between the top


and the bottom (90/10), but have no impact on the top and the middle
(90/50). Both union density and employment protections reduce
income inequality between the bottom and the middle (10/50). In sum,
unions appear to equalize income across both left and right tails of the

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
income distribution, while employment protections seem to operate on
just the left tail.

Beyond the wasteland: towards fair and full


employment for all

The conventional wisdom is that the cause of high European unem-


ployment lies in a job market that is rigid and inflexible. These rigidities
include excessive employment protection, too generous replacement
rates, too long benefit durations, and high rates of unionization. The
empirical results reported in this chapter challenge this received
wisdom.
These results are based on an empirical model of unemployment that
includes both microeconomic labour-market institution variables and
macroeconomic variables. The evidence clearly shows that macroeco-
nomic factors matter for unemployment, and these factors are robust to
changes in the empirical specification of the model. However, when it
comes to microeconomic factors the evidence is much more problem-
atic. The level of wage-bargaining coordination and the extent of union
coverage matter consistently, but they need not raise unemployment if
they are appropriately paired. The level of benefit duration and the level
of union density are both consistently insignificant. The significance of
other microeconomic variables (employment protection, replacement
rate, tax burden) is unstable and not robust to changes in specification.
Moreover, none of these variables is significant in a fully specified model
that takes account of country-specific fixed effects related to Ireland and
Spain.
This leads to the conclusion that high unemployment in Western
Europe is the result of a self-inflicted macroeconomic policy. European
policy-makers adopted a course of disinflation, high real interest rates
and slower growth that raised unemployment. Moreover, since all
adopted this course at the same time, they generated a wave of trade-
based cross-country multipliers that further raised unemployment and
contributed to a continent-wide macroeconomic funk. The policy impli-
cations are clear. Lowering European unemployment will require a
period of sustained expansionary macroeconomic policy, and this

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
44 The Causes of High Unemployment

Labour Market Policy


Regulate Flexibilize

(0 A B
c

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
o
"55
c progressive consensus USA
(0

,0 [jj

aoo
O .
(0 >
2 15 C D

Europe Laissez-faire consensus

Figure 2.1 The economic policy menu

policy needs to be pursued by all countries. Flexibilizing labour-market


institutions will not lower unemployment as these institutions are not
the cause of unemployment. Indeed, if it involves just reducing the
extent of wage bargaining coordination, it could raise unemployment.
The above analysis of the causes of unemployment and factors influ-
encing income distribution is consistent with the two dimensional
macroeconomic-microeconomic policy framework presented by Palley
(1998) which is shown in Figure 2.1. 9
In this framework unemployment is caused by macroeconomic
factors. Microeconomic labour-market institutions protect workers by
giving them voice and bargaining power which impact distributional
outcomes. Weakening these institutions therefore worsens income
distribution but has little impact on unemployment. In the USA macro-
economic policy has been expansionary but labour-market institutions
protecting workers have eroded: the result has been low unemployment
and increased income inequality. In Europe macroeconomic policy
has been contractionary but labour-market institutions protecting
workers remain intact: the result has been high unemployment but
relatively unchanged income inequality. Restoring the golden age
economic prosperity of the post-Second World War era will require
expansionary macroeconomic policy combined with labour-market
institutions that protect workers' voice and bargaining power.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Thomas I. Palley 45

Unfortunately, the laissez-faire 'Washington' consensus that dominates


policy-making recommends the exact opposite combination.

Appendix

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
This Appendix details the data that was used in the regression reported in
Tables 2.1-2.2.
All data for the labour-market institution variables (EMPROT, REPRATE,
BENDUR, UNIONDEN, UNIONCOV, COORD, TAXRATE, ALMPROG) were pro-
vided by Nickell and are as described in Nickell (1997). The macroeconomic data
were taken from OECD Economic Outlook (1999), the World Bank CD-ROM, and
the IMF CD-ROM. The series on real GDP growth was taken from the World Bank
series of that name on the CD-ROM. Updates for 1998 were taken from the World
Bank's homepage. These series match the real GDP growth figures reported in the
June 1999 OECD Economic Outlook, Annex table 1.
Short-term interest rates are from the IMF CD, series 60B, money market rates.
For Ireland, series 60C, Treasury Bills, was used due to the unavailability of the
money market series. Missing values for New Zealand 1978-82 and Australia
1996-98 were filled in using 60C values.
The measures of inflation are the per cent change in consumer prices drawn
from the OECD database's purchasing power parity figures for private consump-
tion, updated to match the OECD's published 1999 figures. DINFLATE is com-
puted as the first difference of the annual inflation rates. The real short-term
interest rate was computed as the difference between the short-term nominal
interest rate and the CPI inflation rate.
Standardized unemployment rates were drawn from the Statwise database
where available, and completed manually from the OECD Economic Outlook
(1999) Annex table 22, with which these figures are in accordance.
To extend the series to include values back to 1977, the June 1999 OECD
Economic Outlook numbers were supplemented by values from the June 1994
OECD Economic Outlook. However, these two series are not always identical owing
to adjustments made by the OECD. To achieve compatibility, the 1994 figures
were adjusted hard copy from the OECD. The series were adjusted for compati-
bility according to the following:

1979 Adj. std. unemp = 1979 std. unemp OECD June 1994
^ 1980 std. unemp OECD June 1999
1980 std. unemp OECD June 1994

Thus, earlier measures of the standardized unemployment rate were converted to


the new basis by multiplying the old series by an adjustment factor. This adjust-
ment factor was computed as the ratio of the first year of the new series to the old
measure of standard unemployment in that year. The first year of the series in
Annex table 22 is 1980.
A similar scaling method was used to create standard unemployment rate
values for countries for which they were unavailable. In these instances, values
for the commonly used definition of unemployment rates (Annex table 21) were

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
46 The Causes of High Unemployment

adjusted according to:

std.unemp t+1
Adj. std.unemp t = common unemp. * —
Common unemp f+1

where the adjustment factor was calculated for the earliest year for which the

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
standard unemployment series was available. The countries to which this was
applied are: Denmark, Austria, Portugal, Ireland; New Zealand had a scalar of 1.
The cross-country Keynesian multiplier openness variable is designed to cap-
ture the impact of growth in the rest of the European economy on each European
country. Canada is especially exposed to growth in the USA, and a similar vari-
able was therefore also constructed for the Canadian economy. The European
country openness variable is defined as:

n( EMP/f \
EUROPEN,,, = s x ; , t | ( T 5 ^ G Y , (

where sx; = export share of GDP for country /; EMPZ = employment in country
i(i = /); TOTEMP; = total employment in all European countries excluding coun-
try/; and GY, = growth of real output in country i(i = j).
The logic of this openness variable is as follows. The sx;- component measures
the export openness of a country, while the rest of the term measures real growth
outside the country. This real growth component is the employment-weighted
average of country growth rates. For all non-European countries EUROPEN takes
on a value of zero. The Canadian openness variable is defined as:

CANUS, = sxCAN/tGYuv

where sxCAN t - Canadian export share of GDP; and GYUS t = US real GDP growth
rate. For all countries other than Canada it is zero.

Notes
1 The OECD continually changes its reported measure of standardized
unemployment, and as a result the measures used here do not match earlier
measures used by Nickell (1997). The current measures are drawn from the
OECD's Economic Outlook, December 1999.
2 Over the period 1983-94 Spain had average standardized unemployment of
19.15 per cent, while Ireland had average standardized unemployment
of 15.32 per cent. The next country after these two was Belgium with an aver-
age standardized unemployment rate of 11.33 per cent.
3 Two-stage least-squares was needed because the ALMPROG variable is defined
as the percentage of GDP spent on labour market policies normalized on the
unemployment rate. The instrument for this variable was spending as a
percentage of GDP normalized on the average unemployment rate in 1977-79
(see Nickell 1997: 64).
4 The statistical significance of REALINT is at odds with results reported by
Scarpetta (1995) which have informed much OECD policy analysis. This

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Thomas I. Palley 47

difference likely stems from differences in the measure of real interest rates.
Scarpetta used a measure of world real interest rates based on a GDP-weighted
average of domestic long term rates. The current estimate uses the short-run
country interest rate which is the appropriate rate for purposes of assessing the
impact of country macroeconomic policies on country unemployment rates.
5 Though negatively signed, the Canadian openness variable (CANUS) is only

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
significant at the 14 per cent level. This may be because the impact of the US
economy on the Canadian economy is fully incorporated in the domestic
GDP growth variable.
6 Indeed, given the coefficients in column (c) of Table 2.2, a properly
constructed system of coordinated wage bargaining and extensive union cov-
erage can lower unemployment. The coefficient of COORD is -0.298, while
that of UNIONCOV is 0.415. However, the value of COORD is twice that of
UNIONCOV.
7 Ireland suffers especially from having high coverage and low coordination
(UNIONCOV= 3, COORD- 2). The UK, Canada and New Zealand also suffer,
albeit less so (UNIONCOV- 2, COORD- 2).
8 These cross-country findings are consistent with the time-series data for just
the USA, which show that union density dramatically lowers US income
inequality (Palley 1999).
9 Stanford (2000) uses a similar framework to compare Canadian economic
policy with that of other countries.

References
Baker, D. and Schmitt, J. (1998) The Macroeconomic Roots of High European
Unemployment: The Impact of Foreign Growth (Washington, DC: Economic Policy
Institute).
Blanchard, O. and Wolfers, J (2000) 'The Role of Shocks and Institutions in the
Rise of European Unemployment: The Aggregate Evidence', The Economic
Journal, 110: C1-C33.
Calmfors, L. and Drifill, J. (1988) 'Bargaining Structure, Corporatism, and
Macroeconomic Performance', Economic Policy, 6: 13-61.
Mishel, L., Bernstein, J. and Schmitt, J. (2000) The State of Working America,
2000/2001 (Ithaca, NY: Cornell University Press).
Nickell, S. (1997) 'Unemployment and Labour Market Rigidities: Europe versus
North America,' Journal of Economic Perspectives, 11: 55-74.
Organisation for Economic Co-operation and Development (1998) OECD
Economic Outlook (Paris: OECD).
Palley, T.I. (1993) 'Okun's Law and the Asymmetric and Changing Nature of the
U.S. Business Cycle', International Review of Applied Economics, 7: 144-62.
Palley, T.I. (1997) 'The Institutionalization of Deflationary Policy Bias', in
H. Hagemann and A. Cohen (eds), Advances in Monetary Theory (Kluwer
Academic Publishers).
Palley, T.I. (1998) 'Restoring Prosperity: Why the US Model is Not the Answer for
the US or Europe', Journal of Post Keynesian Economics, 20 (Spring): 337-54.
Palley, T.I. (1999) 'Arbeitslosigkeit und makrookonomische Weichenstellung', in
S. Lang, M. Meyer and C. Scherrer (eds) Jobwunder USA - Modell fiir Deutschland
(Mtinster: Westfalisches Dampfboot).

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
48 The Causes of High Unemployment

Scarpetta, S. (1996) 'Addressing the Role of Labour Market Policies and


Institutional Settings on Unemployment: A Cross-Country Study', OECD
Economic Studies, 26(1): 44-98.
Solow, R. (1994) 'Europe's Unnecessary Unemployment', International Economic
Insights (March/April).
Stanford, J. (2000) 'Canadian Labour Market Developments in International

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Context: Flexibility, Regulation, and Demand', Canadian Public Policy 26: 27-58.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
3
Is Capital Stock a Determinant
of Unemployment?

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Philip Arestis, Michelle Baddeley and Malcolm Sawyer

Introduction
This chapter seeks further to examine the proposition that the size of the
capital stock is an important variable in the determination of the rate of
unemployment. The argument is along the lines of recent contributions
that are able to confirm the hypothesis that capital stock affects
the non-accelerating inflation rate of unemployment (NAIRU) (see, for
example, Arestis and Biefang-Frisancho Mariscal 1997, 1998, 2000), but
here we focus on EMU countries in our empirical estimations.
An important implication of the hypothesis that unemployment and
capital stock are closely related, concerns current mainstream macro-
economic thinking. The proposition is made whereby monetary policy
is concerned with the nominal side of the economy, and specifically
with inflation, and supply-side policies are designed to address the real
side of the economy. The supply side of the economy is often repre-
sented in terms of an unchanging supply-side equilibrium, with the
NAIRU assumed to summarize it. The estimates provided for the NAIRU
are often presented as a single (and hence implicitly unchanging) num-
ber. A less extreme view would be that the supply-side equilibrium may
change over time but not in response to the demand side of the econ-
omy. Changes in labour-market institutions and laws, for example,
would be predicted to lead to changes in the supply-side equilibrium,
and these changes are assumed to take place independently from any
variations in the level of economic activity. But when the capital stock is
a major influence on the level of the NAIRU, then there is a factor con-
tinually changing the NAIRU (since net investment is always causing
the capital stock to vary) and the pace and structure of investment will
be influenced by the level of economic activity (and by variables such as

49

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
50 Is Capital Stock a Determinant of Unemployment?

profitability which are correlated with economic activity). Hence the


variability of the NAIRU will be influenced continually by the path of
aggregate demand.
The focus of this chapter is to examine the proposition that invest-
ment and the capital stock can contribute substantially to the determi-

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
nation of the rate of unemployment and real activity, and thereby to the
determination of the NAIRU in a range of EMU countries. We deal first
with the existing theoretical and empirical work and attempt to sum-
marize existing literature on the matter. This is followed by estimation
and testing of the theoretical model as postulated in Arestis and Biefang-
Frisancho Mariscal (1997, 1998, 2000). The following four EMU coun-
tries were selected for the empirical investigation, solely on the basis of
data availability and consistency: Austria, Belgium, Germany and Spain.
It is argued here that negative demand shocks affect employment and
investment adversely. When shocks reverse, unemployment may not
fall to previous levels due to insufficient capital equipment. Our empir-
ical results show that the general rise in the unemployment rate in the
countries over the past 30 years was to a large and significant extent due
to insufficient investment, leading to a lower (than otherwise) capital
stock. A final section summarizes the argument and concludes.

Review of the literature

Labour market flexibility


Rigidities in labour markets (for example employment protection laws,
union density and so on), coupled with an inadequately skilled work-
force, are widely held to play a key role in the explanation of high
unemployment rates, especially in Europe in the 1980s and 1990s (see
for example OECD 1994; Nickell 1997; Siebert 1997; Elmeskov, Martin
and Scarpetta 1998; Layard et al. 1991; Nickell et al. 2002; IMF 2003).
Another branch of literature has focused on the particular labour-market
institutions and rigidities that are responsible for high levels of unem-
ployment (representative examples include, Nickell 1997; Layard and
Nickell 1999). Related contributions attempt to examine the evolution
of unemployment across countries using as their guide the hypothesis
that 'flexible' institutions are more adaptable to adverse shocks (for
example, Blanchard and Wolfers 2000). The unemployment policy
implications that follow from the 'labour-market flexibility' thesis
include 'weakening the power of trade unions, cutting taxes on labour,
deregulating labour markets, spending on active labour-market policies,
cutting the value and duration of unemployment benefits and reducing

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Philip Arestis, Michelle Baddeley and Malcolm Sawyer 51

the relative value of minimum wages' (Kapadia 2004: 4). Britain and the
Netherlands are usually thought of as two examples where labour-
market reforms have worked according to the postulates of the theory.
There is, however, a great deal of controversy over the thesis that
labour-market reforms lower unemployment, and the OECD (1999)

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
study is actually damning to the thesis of 'labour-market flexibility'. The
study covers the period late 1980s to late 1990s and includes new and
improved data on employment legislation in 27 OECD countries. It uti-
lizes multiple regression analysis and techniques, in a way that it is able
'to control for other factors that can influence unemployment' (OECD
1999: 88). The study demonstrates that employment protection legisla-
tion (a measure of labour-market flexibility) has small or no impact at all
on total unemployment. The employment protection legislation is
defined broadly and covers all types of employment-protection meas-
ures resulting from legislation, court rulings, collective bargaining or
customary practices. The OECD study considered a set of 22 indicators,
summarized in an overall indicator on the basis of a four-step procedure
(OECD 1999: 115-18). This confirms that dismantling employment pro-
tection would not solve the current unemployment malaise in the
27 countries considered in the study.
There are other studies that are equally, if not more, critical. Ball
(1999) concludes that the British and Dutch experience with labour-
market reforms does not constitute a favourable case for the labour-market
flexibility thesis. Those reforms were relatively minor compared to the
reduction in unemployment these two countries experienced. There are
other countries, though, such as Belgium, Canada and Spain, where
labour-market reforms were undertaken on a larger scale than the British
and Dutch equivalent, but failed to reduce unemployment in the 1990s
(although it is true to say that Spain did experience a fall in unemploy-
ment, from 18.1 per cent in 1994 to 11.3 per cent in the middle of
the 1990s, well before any labour-market reforms were introduced, but
unemployment in Spain has stuck at over 10 per cent ever since; the
Netherlands experienced a fall in unemployment from 7.6 per cent in
1994 to 2.0 per cent in 2001 but then rose back to 5.0 per cent in 2004
took place). It is also argued in the same study that countries like
Portugal and Ireland that did not undertake major changes in their
labour-market institutions and laws, were still successful in reducing
unemployment (less so it must be said in the case of Portugal, from a
peak of 7.2 per cent in 1995 to 4.0 per cent in 2000, followed by a rise to
6.6 per cent to 2004 only because it sought to adhere to the Stability
and Growth Pact). The comparison between Portugal and Spain is

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
52 7s Capital Stock a Determinant of Unemployment?

particularly apt in view of the argument (see Blanchard and Jimeno


1995) that labour-market institutions are very similar in the two countries
(see, also, Kapadia 2004). Glyn (2002) compares the 1990s experience of
Ireland, where unemployment rates fell in the absence of labour-market
reforms, and New Zealand, where despite major reforms only small

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
changes in unemployment took place, and concludes that 'excessive
labour-market deregulation is neither a necessary nor a sufficient condi-
tion for a radical improvement in employment' (Glyn 2002: 16).
The study by Baker et al. (2004) provides empirical evidence that
suggests the results of the well-known studies that support the 'labour-
market flexibility' thesis are not robust, especially with respect to the
variables employed and the time period utilized. The Baker et al. (2004)
study covers 20 OECD countries spanning 40 years, 1960-99. l Different
time periods are utilized and different combinations of variables. The
most comprehensive measure of labour-market institutions and policies
utilized can only account for a minor part of the differences in the evo-
lution of unemployment. An index of the extent of labour-market
deregulation in the 1990s is constructed, but this variable too, showed
no meaningful relationship between labour-market deregulation and
shifts in the NAIRU. The same study poses the question of 'reverse
causality' to conclude that 'While clearly not universal, this evidence of
reverse causation provides serious grounds for viewing test results show-
ing a correlation between high unemployment and long benefit dura-
tion' (Baker et al. 2004: 28) with caution. Consequently, the evidence in
Baker et al. (2004) provides little or no support for the thesis that con-
centrates on labour-market rigidity, and in more general terms on
labour-market institutions. Palley (2001) by accounting for micro- and
macro-economic factors, and also for cross-country economic spillovers,
concludes that unemployment in Europe emanates from 'self-inflicted
dysfunctional macroeconomic policy' (Palley 2001: 3), rather than from
labour-market 'rigidities'.2
Furthermore, there is the argument (Kapadia 2004) that in many of
the more 'successful' countries a more promising relationship might be
between high levels of investment and unemployment reduction. The
cases of Portugal (following accession to the European Economic
Community in 1986), and Ireland (following the large increase in for-
eign direct investment in the mid-1990s), are very telling on this aspect.
The Netherlands, too, experienced increased investment in the early
1990s, from -3.2 per cent in 1990, -4.7 per cent 1991, to +6.9 per cent
in 1992 and 1.2 per cent in 1993. Indeed, the British economy experi-
enced strong investment performance in the early to mid-1990s. In all

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Philip Arestis, Michelle Baddeley and Malcolm Sawyer 53

these cases, increased investment activity was followed by substantial


decreases in unemployment. The British example may be particularly
relevant in this respect in that increased investment activity in the early
parts of the 1990s was associated with falling unemployment in the sec-
ond half of the 1990s. Similarly, earlier periods in the UK (early 1960s to

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
early 1990s) are characterized by high unemployment rates being
accompanied by lower levels of investment (Kitson and Michie 1996).
These experiences confirm Hahn's (1995) puzzlement that 'one can only
be amazed at the neglect of investment and of the capital stock in theo-
ries of the natural rate' (Hahn 1995: 52). They also point to another
important consideration, namely that reduction in unemployment in
all these cases is sustained even after investment rates have fallen.
Capital stock may have some impact in NAIRU. We turn to this aspect
next.

Degree of substitutability between capital and labour


Part of the debate on the relationship between capital shortage and
unemployment concerns the degree of substitutability between labour
and capital (Bean 1989; Jerger 1991; Rowthorn 1999). A larger capital
stock will permit a higher level of aggregate demand, a higher level of
employment and a lower rate of unemployment, without inflation
tending to rise. But this is not always accepted in the literature. When
capital investment takes the form of increasing the average capital stock
per enterprise, and where the elasticity of substitution between capital
and labour is unity and the wage equation shifts up in line with the rise
in output (and hence the labour share in national income remains a
constant), then the NAIRU can involve a substantial level of unemploy-
ment, and it cannot be shifted through the expansion of the capital
stock (see, for example, Layard, Nickell and Jackman 1991: chapter 2).
However, when the elasticity of substitution is below unity, or when the
wage equation does not shift up in line with the rise in output, or when
capital investment takes the form of more enterprises, then the NAIRU
could be guided into compatibility with full employment through
capital investment.
If a Cobb-Douglas production function is assumed, which implies an
elasticity of substitution of unity between labour and capital, then a
1 per cent change in relative factor costs is predicted to lead to a 1 per cent
change in relative factor use. Furthermore, there is empirical evidence
that real wages rise in line with rises in labour productivity, so that the
increase in labour productivity due to higher capital intensity tends to
lead to higher real wages. Consequently, as argued by the critics of the

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
54 Is Capital Stock a Determinant of Unemployment?

capital shortage hypothesis, it is the downward inflexibility of real


wages that prevents a return to previous employment levels (Lindbeck
1993; Blanchard and Summers 1986). No policy measures discriminat-
ing in favour of investment are necessary, for there is no obvious market
failure involved (Bean 1989). It follows that unemployment is best

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
viewed as part of labour-market reform given an existing capital stock.
However, Rowthorn (1999) shows that the bulk of the empirical
evidence suggests that the elasticity of substitution between capital and
labour is considerably below unity, and hence a rising capital-labour
ratio reduces the equilibrium level of unemployment. Rowthorn
(op. cit.) cites 33 studies which show that the elasticity of substitution
between labour and capital is well-below one. In these circumstances,
capital accumulation can affect the equilibrium level of employment,
even when real wages rise following an increase in the capital stock. The
rise in capital stock will reduce a firm's ability to increase their profit
share and the wage share will rise. Due to the limited substitutability
between capital and labour, employment rises with the increase in capi-
tal accumulation. Wage models show that a fall in unemployment is
accompanied by upward pressure on (real) wages (Layard, Nickell and
Jackman 1991). However, since the wage share has increased, trade
unions will be less inclined to demand higher (real) wages in response to
lower unemployment. Consequently, the economy can operate at a lower
level of unemployment. Furthermore, as argued above, slow capital accu-
mulation in the recovery phase prevents unemployment from falling.
It also reduces the effective labour supply through hysteresis effects,
thereby reinforcing the problem of high unemployment. Therefore, poli-
cies that address the speed with which new capital is installed become
necessary in order to prevent insider effects in the labour market.
Rowthorn (1999) assumes a Constant Elasticity of Substitution (CES)
technology. If a Cobb-Douglas production technology is assumed and
capacity constraints are introduced, the same result is obtained. Kapadia
(2004) actually shows that within a Cobb-Douglas production technol-
ogy with imperfectly competitive arrangements in the labour market, of
the type assumed in, for example, Layard et al. (1991), capital stock
would still affect NAIRU. Introducing capacity into the standard Cobb-
Douglas production technology, so that when the capital-labour ratio
drops below a certain threshold, the returns on capital increase while
those of labour fall by the same amount shows that, over a certain range,
NAIRU depends on capital stock. This assumes that capital stock is
exogenous. When capital stock is endogenous, the NAIRU depends on
the real user cost of capital over a certain range.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Philip Arestis, Michelle Baddeley and Malcolm Sawyer 55

Capital stock and unemployment


Gordon (1997) finds that countries that experienced the largest slow-
down in capital accumulation per labour hour faced the highest unem-
ployment rates in the 1990s. He concluded that the European countries
did not have sufficient capital in the 1990s to equip all employees who

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
would have had a job had unemployment rates remained the same as in
the 1970s. Bean (1989) predicted a similar result. The work of Dow (1998)
can be used to illustrate the effects of recession on the future capital stock
and thereby on future employment prospects. He argued that

in a major recession underemployment results in the deterioration


and premature scrapping of physical equipment, and that
disbandment or underemployment of a firm's workforce similarly
results in the partial destruction of working practices and working
relations. The latter constitute the intangible capital of a firm, the
value of which is an important fraction of its market value as a going
concern. The capital stock, physical and intangible, takes time to
build up, and it cannot be destroyed rapidly; in effect, therefore, the
destruction is quasi-permanent. In this way demand shocks impact
on supply. A major recession causes a downward displacement of the
growth path of productivity (or potential or capacity output); after
the recession, the 'stable growth' mechanism described by the first
mechanism will in the absence of further shocks start to operate
again, i.e. normal growth will be resumed from the low point of the
recession. (Dow 1998: 369)

Dow produces estimates of the impact of five major recessions in the


UK, implying that substantial (negative) changes in capacity can occur
as the result of recession, with long lasting effects. These findings are
consistent with the views we have outlined above.
The increase in the level of unemployment experienced in the 1980s
and in the 1990s as compared with say the 1960s (especially in Europe)
was substantial. The oil price shock of the mid-1970s could be seen as
effectively reducing capacity as heavy energy-using plant and equip-
ment became unprofitable. High levels of unemployment and excess
capacity can be expected to have led to a substantial reduction in the
capital stock below its full employment level. In some respects, this says
little more than enterprises will adjust the capital stock to the prevailing
demand for output (and level of employment). But it, nonetheless, sug-
gests a clear mechanism through which the level of (un)employment
experienced will be reflected in the estimated NAIRU.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
56 Is Capital Stock a Determinant of Unemployment?

Estimates of a NAIRU (or similar) are generally derived from the


estimation of wage and price equations from which an 'equilibrium' rate
of unemployment is solved (based on expectations being fulfilled and
real wages growing in line with productivity). This equilibrium is equiv-
alent to income distribution between wages and profits remaining

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
unchanged, and hence in the absence of pronounced trend in the dis-
tribution of income any successful estimation of the NAIRU will fall
in the range of observed unemployment. It is, though, often observed
that the NAIRU is a 'weak attractor'. For example, Blanchard argues that
'the natural rate is at best a weak attractor' and that 'the natural rate is
often as much an attractee as it is an attractor' (Blanchard 1995: xiii),
where the term 'natural rate' is used in the way we have used the term
NAIRU and that the estimates of the NAIRU tend to follow the actual expe-
rience of unemployment (see, for example, Worswick 1985). The variation
in the estimated value of the NAIRU, the tendency of estimates of the
NAIRU to follow actual unemployment, and the NAIRU being a 'weak
attractor' are consistent with the view put forward in the current paper.
Sawyer (2004) considers the experience of the 1990s with declining
unemployment without inflation in Canada, the UK and the USA, con-
trasting it with the high levels of unemployment in countries such as
France and Germany. He argues that while unemployment of labour was
falling in the first group of countries, capacity utilization did not rise
substantially as there was something of an investment boom, which led
to a rising capital stock. There is evidence to support this view, though
the empirical work to which we refer has been undertaken within frame-
works that have some similarities to the one outlined here, albeit with
some differences as well. Rowthorn finds that 'when manufacturing
and services are combined, capital stock has a large, statistically signifi-
cant impact on employment' (Rowthorn 1995: 33), in equations esti-
mated for a cross-section of OECD countries. So that, Rowthorn
concludes, 'The problem of unemployment is ultimately one of invest-
ment' (Rowthorn 1995: 38). In an introduction to a recent symposium
on unemployment, Dixon concludes that 'although the papers [in the
Controversy section] reflect the wide disagreement which exists
amongst economists, they suggest that the increase in the real interest
rate and the decline in the investment ratio are partly to blame for the
high unemployment rate in the OECD countries' (Dixon 1998: 781).
Arestis and Biefang-Frisancho Mariscal (1997) conclude from their
empirical work for the UK over the period 1966 to 1994 that 'unem-
ployment is significantly determined by capital shortages. Capacity is
not fixed and investment depends on expected profitability and the

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Philip Arestis, Michelle Baddeley and Malcolm Sawyer 5 7

expected long-term rate of interest' (Arestis and Biefang-Frisancho


Mariscal 1997: 191). In a subsequent paper, they find that 'the NAIRU is
determined by long-term unemployment, worker militancy and the
capital stock' (Arestis and Biefang-Frisancho Mariscal 1998: 202). In
work on Germany and the UK those authors conclude that 'adverse

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
demand shocks affect employment and investment. When shocks
reverse, unemployment may not fall to previous levels due to insuffi-
cient capital' (Arestis and Biefang-Frisancho Mariscal 2000: 487).
Miaouli utilizes annual data from the manufacturing sector of five
European countries and concludes that 'the empirical analysis verifies
that in all countries, accumulated investment in the private sector influ-
ences employment in a positive way' (Miaouli 2001: 23). Rowthorn
finds that 'when manufacturing and services are combined, capital stock
has a large, statistically significant impact on employment' (Rowthorn
1995: 33) in equations estimated for a cross-section of OECD countries.
Stockhammer aims 'to contrast and test the NAIRU hypothesis and a
Keynesian explanation of unemployment in a time-series context. For
the NAIRU explanation, wage push variables are key to explaining the
rise of unemployment in Europe, for Keynesians the slowdown in capi-
tal accumulation is' (Stockhammer 2004: 21) the key. This proposition is
tested using data from the mid-1960s to the mid-1990s for Germany,
France, Italy, the UK and the USA. He concludes that 'the NAIRU speci-
fication performed poorly, with only the tax wedge having a positive
effect on unemployment as predicted. As to the Keynesian approach,
the role of capital accumulation was confirmed. Whereas capital
accumulation is robust to the specification and can be pooled across
countries, the tax wedge is not. In the Keynesian specification the tax
wedge has the incorrect sign, however replacement ratios are significant
with the predicted sign' (Stockhammer 2004: 21). Further support
comes from Alexiou and Pitelis (2003), where a panel-based study is
undertaken for the period 1961 to 1998 for a number of European coun-
tries. They conclude that their 'analysis and empirical findings suggest
that one of the potential factors behind the high and persistent
European unemployment is insufficient growth of capital stock and
inadequate aggregate demand' (Alexiou and Pitelis 2003: 628).
In this chapter we utilize a model that aims to explain the determina-
tion of wages and unemployment. Specific to this model is the effect of
the capital stock on wages and unemployment, and consequently on
the NAIRU. We follow the theoretical framework put forward in Arestis
and Biefang-Frisancho Mariscal (1997, 1998, 2000) for the purposes of
empirical investigation.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
58 is Capital Stock a Determinant of Unemployment?

Estimable equations

Our empirical work concerns the estimation of two relationships, one


for real wages and one for unemployment, involving the capital stock,
and the basic equations derived in Arestis and Biefang-Frisancho

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Mariscal (1997, 1998, 2000) will be used throughout. We begin with
equation (3.1), where lower-case variables are in logarithms:

w = ps + j32s - p3u + p4ltu + psk (3.1)

where the dependent variable w, defined as w = (wn - p - Ip) with wn


being nominal wage, p the GDP price deflator and Ip labour productiv-
ity, is the real wage per unit of output, which implies that productivity
neutrality with respect to unemployment is assumed. Although the the-
oretical model (Arestis and Biefang-Frisancho Mariscal 1997, 1998, 2000)
itself does not impose this restriction, this assumption is backed by
empirical results (Manning 1992; Elmskov 1993). Furthermore, r=(wu- p)
is real level of unemployment benefits,3 where wu is unemployment
benefits, 5 stands for strike activity, u for unemployment, Itu for long-
term unemployment, and k for capital stock. This equation indicates
how the rise in capital stock can reduce inflationary conflict over
income distribution, and thus affect the real wage per unit of output.
This can be explained in the following two ways (Rowthorn 1995). First,
a higher capital stock would mean that capacity utilization will be lower
for any given level of aggregate demand, and this lower capacity utiliza-
tion restrains firms' ability to raise prices. Second, the rise in capital
stock leads to a fall in conflict over income distribution, which produces
lower domestic prices, which may lead to the economy operating with a
higher real exchange rate; this increases the amount of resources avail-
able for domestic use. Each of these effects allows the real wage to rise
and helps to reduce inflationary conflict.
The second equation derived is:

u =
Jir + Jis ~ 73^ + yJAu (3.2)

It can be seen that the equilibrium rate of unemployment is a declining


function of capital stock. The effect of changes in capital stock on unem-
ployment can be explained as follows. If capital stock is reduced, or
taxes or import cost rise, conflict over income distribution rises. Taking
as an example the rise in oil prices in the mid-1970s, the resulting con-
flict over who should bear the cost led to accelerating inflation. In order

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Philip Arestis, Michelle Baddeley and Malcolm Sawyer 59

to bring inflation down, governments introduced policies that curbed


demand. As a result, unemployment rose. This can be explained if a
fall in demand is accompanied by a fall in capacity utilization. Firms
respond by reducing investment, leading to a lower (than otherwise)
capital stock, which increases the NAIRU even further. When oil prices

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
return to their previous level, inflationary pressure and unemployment
fall. However, due to a lower capital stock, the previous level of employ-
ment cannot be attained.

Empirical investigation and empirical evidence

The first part of the empirical section discusses the data utilized for
the four countries concerned. This is followed by the estimation of the
cointegrating relationships corresponding to equations (3.1) and (3.2).

Variable definition and data


The selected countries Austria, Belgium, Germany and Spain exhibit
consistency in the data. The model is estimated using quarterly data for
different periods in the four countries as indicated in Table 3.1. The
definition of variables for all countries are as follows: real hourly average
earnings per employee is denoted by (w), deflated by the GDP price
deflator (p) and labour productivity (pf); (r) is real benefits, that is nom-
inal benefits (wu) deflated by (p); (5) is the number of strikes; (u) is the
number of unemployed; the long-term unemployed (Itu) are those unem-
ployed for more than 52 weeks in relation to the labour force; and capi-
tal stock (k) is the business-sector capital stock. As suggested above, all
variables are measured in logarithm form.

Empirical results
We apply the standard Johansen (1988, 1995) maximum likelihood esti-
mation procedure to estimate the number of linearly independent coin-
tegrating vectors (see, also, Johansen and Juselius 1990). Identifying
restrictions on the cointegrating vectors are then imposed and tested for
lying in the cointegrating space spanned by the cointegrating vectors.
The results along with relevant explanatory notes are given in
Tables 3.1-3.3.
All variables were tested for the level of integration, applying the stan-
dard Augmented Dickey-Fuller (ADF) unit-root tests with the result that
all variables have unit roots.4 The Akaike (AIC) and Schwarz-Bayesian
(SBC) information criteria were applied in order to enable us in the
model selection procedure. These criteria are designed to select the

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect -
ON

Table 3.1 Cointegrating vectors

Austria Belgium Germany Spain


1994Q4 to 2002Q4 198SQ4 to 2000Q4 1983Q4 to 2002Q4 1979Q4 to 2002Q4
Normalized on Normalized on Normalized on Normalized on

Wages Unemployment Wages Unemployment Wages Unemployment Wages Unemployment

-0.03 n/a -0.08 n/a -0.02 nidi -0.38 nidi


[-2.62] n/a [-0.41] n/a [-5.25] nidi [-1.85] n/a.

2014-03-15
0.03 -1.04 -0.41 6.40 0.25 3.73 0.29 0.50
[2.34] [-2.46] [-0.50] [1.00] [3.51] [ 1.41] [ 1.88] [4.30]
0.00 0.00 0.12 0.50 n/a n/a 0.22 -0.87
[2.31] [0.31] [5.06] [3.14] n/a n/a [1.39] [-18.07]
0.04 -0.16 0.18 -2.22 0.001 -0.001 0.85 -1.99
[5.53] [-1.51] [0.67] [-1.20] [5.20] [-1.94] [2.31] [-17.69]
0.05 0.55 0.31 3.57 0.25 -13.46 -0.63 -0.14
[4.38] [2.26] [1.45] [2.11] [4.37] [-6.90] [-5.94] [-1.75]

Notes: w=number of unemployed; r = real benefits; s = number of strikes; k = business sector capital stock; Itu = long-term unemployed; all variables
are measured in logarithm form, r-ratios are in square brackets.
Sources: OECD Quarterly Labour Force Statistics, OECD Business Sector Database, ILO.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim Truger
Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect -
Table 3.2 Model specification statistics

Austria Belgium Germany Spain

Wages Unemployment Wages Unemployment Wages Unemployment Wages Unemployment

Adj.R2 0.92

2014-03-15
0.45 0.10 0.38 0.20 0.33 0.08 0.63
F-statistic F(14,33) = 3.19 F(12,33) = 39.50 F(14,61) = 1.40 F(12,61) = 3.47 F(14,71) = 2.32 F(12,71) = 3.87 F(14,87) = 1.58 F(12,87) = 14.21
AIC -7.70 -2.37 -6.42 -2.54 -6.29 1.95 -4.99 -4.62
SBC -7.11 -1.87 -5.85 -2.04 -5.84 2.38 -4.58 -4.27
x2 430.64(14) 258.71(12) 852.80(14) 638.20(12) -33.84(14) -310.79(12) 1444.84(14) 1179.41(12)

Notes: AIC = Akaike information criterion; SBC = Schwarz-Bayesian information criterion; all variables are in logarithms; numbers in brackets denote the
numbers of degrees of freedom.
Sources: OECD Quarterly Labour Force Statistics, OECD Business Sector Database, ILO.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim Truger
62 is Capital Stock a Determinant of Unemployment?

Table 3.3 Key to variables and statistics/diagnostics as in Tables 3.1 and 3.2

w \n(w-p-pr), with p being the GDP price deflator and pr labour


productivity
u ln(no. of unemployed), for Germany is the rate of unemployment
k ln(X) 1995 prices

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
s ln(strikes)
r ln(real unemployment benefits)
Itu ln(long-term unemployment)
n/a not applicable
F-statistic F-test of explanatory power, with its critical value at 95%
confidence interval being approximately equal to 2.0
AIC Akaike Information Criterion is a model selection criterion based
around the residual variance which is equal to Tln(RSS)+2iC,
where K = number of regressors, T = number of observations,
and RSS = residual sum of squares
SBC Schwarz-Bayesian Information Criterion; another model selection
criterion, which selects the most parsimonious model = Tln(MSE) +
iCm(T), where MSE is the mean-square error
^(ri) log-likelihood function, which captures the probability that the
model is data consistent. The critical values for a 95% confidence
interval are: *2(14) = 23.68 and *2(12) = 21.03

model with the optimal empirical lag length in the specification of the
vector error-correction model (VECM). They also require that the result-
ing model has residuals that correspond to the absence of serial correla-
tion. It is well-known that the SBC is d o m i n a n t w h e n selecting lag
length and cointegrating rank. The F(nlf n2) statistic suggests that the
explanatory power of the equation is satisfactory. The results of all these
tests, along with those of the relevant R2 and ^iri) tests, are reported in
Table 3.2. 5 The R2s are reasonable, and the x2(n) is significant in all cases,
which suggests that the model conforms with the theoretical priors.
Most importantly, the AIC and SBC statistics are remarkably similar and
in most cases above their critical values. In fact, all the results of the tests
conducted for the purposes of Table 3.2 are satisfactory from the point
of view of the model selection aspect of the exercise.
The two cointegrating vectors, as discussed in the last sub-section,
may not be meaningful in an economic sense. They span a space in
which any linear combination is another cointegrating relationship.
Since the Johansen's reduced rank regression procedure only determines
how m a n y unique cointegration vectors span the cointegration space,

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Philip Arestis, Michelle Baddeley and Malcolm Sawyer 63

and since any linear combination of the stationary vectors is also a


stationary vector, the estimates produced are not necessarily unique.
Therefore, it will be necessary to impose restrictions to obtain unique
vectors lying within that space (Harris 1995). Consequently, general
joint restrictions on the cointegration vectors and the speed-of-adjustment

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
parameters, according to prior theoretical premises, were imposed and
tested for significance (see, for example, Harris 1995; Hunter 1992; and
Mosconi and Giannini 1992). The resulting equations thereby derived
for the four countries are reported in Table 3.1, which should be read in
conjunction with Table 3.2.6 In Table 3.1 the values in square brackets
are the t-values.
The first equation for each country describes the average hourly real
wage per unit of output (wn—p-lp), determined by the logarithm of
unemployment (u), the logarithm of real benefits (r), strike activity rep-
resented by the logarithm of the number of strikes (s), capital stock (k)
and the logarithm of long-term unemployment (Itu). The second equa-
tion explains unemployment by real benefits, number of strikes, capital
stock and by the number of long-term unemployed. Most variables are
correctly signed and statistically significant at the conventional levels.
In particular, it may be noted that for all the countries in this investiga-
tion the real wage per unit of output is negatively related to unemploy-
ment, and hence positively related to employment. Furthermore, and
for the purposes of this paper, the capital stock variable is correctly
signed and significant with varying degrees in the eight relationships
reported in Table 3.1. The empirical performance of the rest of the vari-
ables support the theoretical contention postulated in this chapter.

Summary and conclusions

We have employed an aggregate wage model which incorporates a num-


ber of ideas on wage determination, and the model is concerned with
explaining the (long-run) level of unemployment. In this model, the
effect of demand on unemployment and wages is emphasized. The
mechanism works through changes in capacity utilization, which them-
selves depend on economic activity and the stock of capital. Wage and
unemployment relationships are estimated and tested with comparable
data sets for the four EMU countries included in our sample. The data-
sets provided two cointegrating vectors, which could be identified as
unemployment and wage relationships. The empirical results, based on
quarterly time-series data for the four countries, confirm Stiglitz's state-
ment that 'it takes capital and entrepreneurship to create new firms and

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
64 Is Capital Stock a Determinant of Unemployment?

jobs' (Stiglitz 2002: 59). Consequently, policies to promote investment


also help to tackle unemployment. The current focus on labour-market
reforms and flexibility is overstated. The policy implications of our find-
ings are that in view of the strong effect of long-term u n e m p l o y m e n t o n
the NAIRU, programmes that enhance the skills of displaced workers

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
may allow t h e m to regain access to the labour market. Furthermore, the
results suggest that capital stock significantly affects unemployment.
This is important since investment-enhancing policies can lead the way
in reducing unemployment.

Notes
1 It is important to note that this data series covers the late 1990s when
unemployment rates fell sharply in the 20 countries included in the study's
sample.
2 It is worth commenting on Germany's current labour-market reforms, which
do not appear to be delivering a reduction in unemployment, although it
should be readily noted that they have not been in place for all that long.
Financial Times (20 October 2004) reports that six leading German economic
institutes have criticized labour-market reforms. Their critique rests on the
failure of the reforms to reduce unemployment, which if anything has been
increasing over the period. More importantly, though, the six institutions
warn that Germany's labour reforms might fail to curb high unemployment.
However, this has not prevented the majority of them from demanding fur-
ther reforms.
3 We note that the variable real benefits (wu-p) differs from the variable replace-
ment ratio, with the latter being (wu-w).
4 The results of these tests can be obtained from the authors upon request.
5 The results of the vector error-correction model are not reported in the chap-
ter, but can be obtained from the authors upon request.
6 The results reported in Tables 3.1 and 3.2 were estimated using EViews 4.0 ©
1994-2000 Quantitative Micro Software, LLS, California, USA.

References
Alexiou, C. and Pitelis, C. (2003) 'On Capital Shortages and European
Unemployment: A Panel Data Investigation', Journal of Post Keynesian
Economics, 25(4): 613-31.
Arestis, P. and Biefang-Frisancho Mariscal, I. (1997) 'Conflict, Effort and Capital
Stock in UK Wage Determination', Empirica, 24(3): 179-93.
Arestis, P. and Biefang-Frisancho Mariscal, I. (1998) 'Capital Shortages and
Asymmetries in UK Unemployment', Structural Change and Economic Dynamics,
9(2): 189-204.
Arestis, P. and Biefang-Frisancho Mariscal, I. (2000) 'Capital Stock, Unemploy-
ment and Wages in the UK and Germany', Scottish Journal of Political Economy,
47(5): 487-503.
Baker, D., Glyn, A., Howell, D. and Schmitt, D. (2004) 'Labour Market Institutions
and Unemployment: A Critical Assessment of the Cross-Country Evidence', in

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Philip Arestis, Michelle Baddeley and Malcolm Sawyer 65

D. Howell (ed.), Fighting Unemployment: The Limits of Free Market Orthodoxy


(Oxford: Oxford University Press).
Ball, L. (1999) 'Aggregate Demand and Long-run Unemployment', Brookings
Papers on Economic Activity, 2: 89-251.
Bean, C.R. (1989) 'Capital Shortage and Persistent Unemployment', Economic
Policy, 4(1): 12-53.

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Bean, C.R., Layard, R.G. and Nickell, SJ. (1986) 'The Rise in Unemployment: A
Multi-Country Study', Economica, Supplement, 53(210): Sl-22.
Blanchard, D.G. (1995) 'Preface', in R. Cross (ed.), The Natural Rate of Unemployment:
Reflections on 25 Years of the Hypothesis (Cambridge: Cambridge University
Press).
Blanchard, OJ. and Jimeno, J.F. (1995) 'Structural Unemployment: Spain versus
Portugal', American Economic Review, 85(2): 212-18.
Blanchard, OJ. and Summers, L.H. (1986) 'Hysteresis and the European
Unemployment Problem', in S. Fischer (ed.), NBER Macroeconomics Annual 1986
(Cambridge, Mass.: MIT Press).
Blanchard, O.J. and Wolfers, J. (2000) 'The Role of Shocks and Institutions in the
Rise of European Unemployment: The Aggregate Evidence', Economic Journal,
110(462): C1-C33.
Dixon, H. (1998) 'Controversy: The Macroeconomics of Unemployment in the
OECD', Economic Journal, 108(448): 779-81.
Dow, J.C.R. (1998) Major Recessions: Britain and the World, 1920-1995 (London
and New York: Oxford University Press).
Elmskov, J. (1993) 'High and Persistent Unemployment: Assessment of the
Problem and its Causes', Economics Department Working Paper, 132 (Paris: OECD).
Elmeskov, J., Martin, J. and Scarpetta, S. (1998) 'Key Lessons for Labour Market
Reforms: Evidence from OECD Countries Experience', Swedish Economic Policy
Review, 5(2): 205-52.
Glyn, A. (2002), 'Labour Market Success and Labour Market Reform: Lessons from
Ireland and New Zealand', mimeo, Oxford University, http:/www.economics.ox.
ac.uk/Members/andrew.glyn/IrelandandNZ4.pdf.
Gordon, RJ. (1997) 'Is there a Trade-off Between Unemployment and
Productivity Growth?', in D.J. Snower and G. De la Dehesa (eds), Unemployment
Policy: Government Options for the Labour Market (Cambridge: Cambridge
University Press).
Hahn, F. (1995) 'Theoretical Reflections on the "Natural Rate of Unemployment" ',
in R. Cross (ed.), The Natural Rate of Unemployment: Reflections on 25 Years of the
Hypothesis (Cambridge: Cambridge University Press).
Harris, R.I.D. (1995) Using Cointegration Analysis in Econometric Modelling (London:
Prentice Hall).
Hunter, J. (1992) 'Cointegrating Exogeneity', Economics Letters, 34: 33-5.
IMF (2003) 'Unemployment and Labour Market Institutions: Why Reforms Pay
Off, World Economic Outlook (April, chapter IV): 129-50 (Washington, DC:
International Monetary Fund).
Jerger, J. (1991) 'Wage Gaps, Capital Formation and Unemployment', Jahrbuecher
fuer Nationaloekonomie und Statisik, 208(3): 264-71.
Johansen, S. (1988) 'Statistical Analysis of Cointegrating Vectors', Journal of
Economic Dynamics and Control, 12: 231-54.
Johansen, S. (1995) Likelihood-based Inference in Cointegrated Vector Autoregressive
Models (Oxford: Oxford University Press).

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
66 Is Capital Stock a Determinant of Unemployment?

Johansen, S. and Juselius, K. (1990) 'Maximum Likelihood Estimation and Inference


on Cointegration with Applications to the Demand for Money', Oxford Bulletin
of Economics and Statistics, 52: 169-210.
Kapadia, S. (2004) 'The Capital Stock and Equilibrium Unemployment: A New
Theoretical Perspective', mimeo (Balliol College, Oxford University).
Kitson, M. and Michie, J. (1996) 'Britain's Industrial Performance Since 1960:

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Underinvestment and Relative Decline', Economic Journal, 106: 196-212.
Layard, R., Nickell, S. and Jackman, R. (1991) Unemployment (Oxford: Oxford
University Press).
Layard, R. and Nickell, S. (1999) 'Labour Market Institutions and Economic
Performance', in O. Ashenfelter and D. Card (eds), Handbook of Labour
Economics, 3C: 3029-86.
Lindbeck, A. (1993) Unemployment and Macroeconomics (Cambridge, Mass.: MIT).
Manning, A. (1992) 'Productivity Growth, Wage Setting and the Equilibrium Rate
of Unemployment', CEP Discussion Paper no. 63 (London: London School of
Economics).
Miaouli, N. (2001) 'Employment and Capital Accumulation in Unionised Labour
Markets', International Review of Applied Economics, 15(1): 5-30.
Mosconi, R. and Giannini, C. (1992) 'Non-causality in Cointegrated System:
Representation, Estimation and Testing', Oxford Bulletin of Economics and
Statistics, 48:300-417.
Nickell, S. (1987) 'Why is Wage Inflation in Britain so High?', Oxford Bulletin of
Economics and Statistics, 49: 103-29.
Nickell, S. (1997) 'Unemployment and Labour Market Rigidities: Europe Versus
North America', lournal of Economic Perspectives, 11(3): 55-74.
Nickell, S., Nunziata, L. and Ochel, W. (2002) Unemployment in the OECD since the
1960s. What Do We Know (London: Bank of England). http:/www.nuff.ox.
ac.uk/users/nunziata/publications_files/nicknun02.pdf.
OECD (1994) OECD Jobs Study, Evidence and Explanations, Part I: Labour Market
Trends and Underlying Forces of Change (Paris: OECD).
OECD (1999) OECD Employment Outlook, June (Paris: OECD).
Palley, T.I. (2001) 'The Role of Institutions and Policies in Creating High
European Unemployment: The Evidence', Levy Economics Institute of Bard
College Working Paper Series, 336 (August).
Rowthorn, R.E. (1995) 'Capital Formation and Unemployment', Oxford Review of
Economic Policy, 11(1): 26-39.
Rowthorn, R.E. (1999) 'Unemployment, Wage Bargaining and Capital-labour
Substitution', Cambridge Journal of Economics, 23(4): 413-26.
Sawyer, M. (2004) 'The NAIRU, Labour Market "Flexibility" and Full Employment',
in J. Stanford and L. Vosko (eds), Challenging the Market: The Struggle to Regulate
Work and Income (McGill: Queen's University Press), 33-50.
Siebert, H. (1997) 'Labour Market Rigidities: At the Root of Unemployment
Problem', Journal of Economic Perspectives, 11(3): 37-54.
Stiglitz, J.E. (2002) Globalisation and its Discontents (London: Allen Lane).
Stockhammer, E. (2004) 'Explaining European Unemployment: Testing the
NAIRU Hypothesis and a Keynesian Approach', International Review of Applied
Economics, 18(1): 3-24.
Worswick, D. (1985) 'Jobs for All?', Economic Journal, 95(1): 1-15.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
4
Deflation Risks in Germany and
the EMU: The Role of Wages and

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Wage Bargaining*
Eckhard Hein, Thorsten Schulten and Achim Truger

Introduction

In late 2003 the deflationary dangers in Germany could hardly be


denied. The GDP deflator for Germany rose by 1.0 per cent in 2003 and
the forecast rise for 2004 was 0.8 per cent. Meanwhile, the consumer
price index rose by 0.9 per cent in 2003 and was forecast to rise by
1.2 per cent in 2004 (Institute 2003). These figures mean that inflation
in Germany has in principle already reached the level considered by the
European Central Bank (ECB) to be the minimum safety margin against
deflation in its reformulated monetary policy strategy for the whole of
the European Monetary Union (EMU) (ECB 2003). A further fall in infla-
tion would therefore significantly increase the danger of deflation and a
cumulative deflationary spiral. In its April 2003 Task Force Report, the
IMF named Germany alongside Japan, Taiwan and Hong Kong as one of
the economies most at risk from deflation worldwide (IMF 2003).l
Not least because of Japan's experiences in the 1990s, there is cur-
rently a broad consensus among economists that once a deflationary
spiral is underway it has very negative consequences for growth and
employment and is extremely difficult to stop. The causes of deflation-
ary processes can be found both on the supply side and the demand side

* An earlier version of this chapter was presented at the 8th International Post-
Keynesian Conference, 26-29 June, 2004, Center for Full Employment and Price
Stability, University of Missouri, Kansas City. For helpful comments we would
like to thank the participants in this conference and in the 8th Workshop of the
Research Network Alternative Macroeconomic Policies on 'Wages, Distribution
and Growth', 29-30 October, 2004, Berlin.

67

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
68 Deflation Risks in Germany and the EMU

(IMF 2003). However, while the price falls resulting from positive supply
shocks (such as innovations that boost productivity) are usually associ-
ated with an increase in economic activity, deflationary processes
caused by negative demand shocks go hand in hand with an overall fall
in economic activity. The real problem posed by deflation thus results

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
from a combination of falling demand and output with falling prices.
This leads to an expectation of further price cuts, an increase in the
value of real debts, falling share prices and stricter lending policies on
the part of commercial banks and financial intermediaries, all of which
ultimately may cause debtors to become insolvent and go bankrupt.
Owing to the risk of cumulative effects, there is also a broad consensus
that economic policy should take timely and decisive steps to combat
deflation, ideally as soon as the first signs of it emerge. This is particu-
larly important because as the nominal interest rate heads towards zero,
less and less can be achieved by monetary policy.2
However, there is a clear lack of consensus among economists with
regard to which instruments should be used to tackle (incipient) defla-
tion. While (post-)Keynesian authors have always stressed the key role of
wages policy as the nominal anchor for combating both inflationary and
deflationary tendencies, wages policy as an instrument in mainstream
new-Keynesian thinking is either non-existent or at best allocated a
highly ambivalent and ultimately contradictory role.
For example, the IMF (2003) study ranks the key indicators of defla-
tion risks as follows: (1) consumer and producer prices, (2) overcapacity
and output gaps, (3) share prices and property prices, and (4) credit and
money aggregates. Wages or unit-labour-cost trends are not explicitly
mentioned at all. This is hardly surprising, since at times of sustained
demand-led deflation, rigid nominal wages are considered to be an addi-
tional destabilizing factor that can lead to an increase in real wages and
a fall in employment and should therefore be avoided, according to the
IMF. On the other hand, during temporary demand shocks, rigid nominal
wages are considered to be a tried and tested means of preventing price-
cut expectations and their associated deflationary consequences from
arising in the first place. However, the point at which a temporary shock
becomes a sustained shock requiring nominal wages to be lowered is by
no means clear, nor is it evident how lowering nominal wages can halt
and reverse a deflationary process as described above once it is already
underway.
Similar inconsistencies with regard to the relationship between nomi-
nal wage rigidity, low inflation or deflation and employment are to be
found in a number of standard works on the problems of low inflation

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Eckhard Hein, Thorsten Schulten and Achim Truger 69

and deflation. For example, Akerlof, Dickens and Perry (1996) show that
because of employees' perceptions of what is fair and morally right,
nominal wage rigidities are inevitable, irrespective of what the existing
labour market institutions are. In their view, when inflation is low and
demand is falling these rigidities act as an obstacle to the necessary

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
downward adjustment of real wages, thereby destabilizing the whole
economy.3 At the same time, however, they also view downward
nominal wage rigidities as a means of braking cumulative deflationary
processes and consequently as something that promotes overall eco-
nomic stability! Also Bernanke (1995) blames insufficient downward
nominal wage flexibility together with debt deflation for the severity of
the Great Depression starting in 1929 which was characterized by major
deflationary processes without even acknowledging that downward
nominal wage rigidity might have stopped deflation. Finally, in an
otherwise highly informative study for the Board of Governors of the US
Federal Reserve System on the ultimate failure of Japan's economic
policy measures to prevent deflation in the 1990s, Ahearne et al. (2002)
make no mention whatsoever of wage trends or wages policy.
In contrast to these predominantly new-Keynesian analyses, more
recent studies of deflation risks in Germany have pointed to the desta-
bilizing effects of German wages policy (Flassbeck and Maier-Rigaud
2003; Kromphardt 2003).4 According to this approach, a policy of exces-
sive wage restraint has led to low increases in unit labour costs and con-
sequently to low inflation. To place wage trends at the core of the
analysis of deflation risks is to follow the line of reasoning outlined by
Keynes (1936: 257-71) in his General Theory, and at present it is only
post-Keynesian authors who continue to take this approach to its ulti-
mate conclusion. In contrast to the predominantly new-Keynesian stud-
ies alluded to above, Keynes and post-Keynesian theory view rigid
nominal wages and stable unit labour costs as the indispensable basis
for price stability in a monetary production economy. Consequently,
rather than disturbing the market system and threatening to prevent it
from functioning optimally, rigid nominal wages resulting from trade
union wages policy or statutory minimum wages are in actual fact con-
sidered to be a requirement for the functioning of capitalist monetary
economies. This is because to remove the wage anchor is to remove the
last barrier against cumulative and disruptive deflationary processes.
Of course, in the post-Keynesian view the wage anchor also prevents
cumulative inflationary processes.
This chapter will follow Keynes's or the post-Keynesian line of reason-
ing in order to demonstrate that wages policy especially in Germany but

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
70 Deflation Risks in Germany and the EMU

also in the rest of the EMU is in danger of not fulfilling its stabilizing role
and that wage trends are causing deflationary risks. The second section
will summarize the key theoretical links between wages, prices and
employment from Keynes's and the post-Keynesian perspective. The
third section will present an empirical study of the relationship between

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
unit labour cost trends and inflation in Germany and the EMU, and this
will be followed in the fourth section by an analysis of the causes of the
observed unit labour cost trends. The final section will discuss the
macroeconomic risks arising from the current wages policy in the con-
text of the EMU's monetary and fiscal policies.

Wages, prices and employment

The post-Keynesian approach to analysing prices and employment that


underpins this chapter differs fundamentally from mainstream think-
ing. In the neoclassical, neoclassical synthesis, monetarist and new
classical models, Say's Law and the classical dichotomy between the real
and the monetary sphere apply in the long term (and also in the short
term in new classical models) (Snowdon, Vane and Wynarczyk 1994).
Nominal wage settlements in the labour market affect real wages and
hence determine employment and output levels. Price levels are deter-
mined by the money supply, which is controlled by the central bank,
and inflation and deflation are purely monetary phenomena attributa-
ble to the central bank's monetary policy.
The new-Keynesian models (Snowdon, Vane and Wynarcyk 1994),
and in particular the 'new consensus models' (Arestis and Sawyer
2003; Clarida, Gali and Gertler 1999; Meyer 2001), do abandon the
assumption that the central bank can control the money supply. More
realistically, it is assumed that for a credit money economy the action
parameter of monetary policy is the money interest rate.5 Monetary
policies may have a short-term real effect on output and employment.
In the long term, however, unemployment is determined by the NAIRU
(non-accelerating inflation rate of unemployment) which itself depends
on structural factors of the labour market, the wage bargaining and the
social security system. As such, the NAIRU describes the unemployment
rate at which, in imperfect labour and commodity markets, the distribu-
tion claims by employees and employers do not result in any increase or
decrease in the inflation rate. When unemployment falls below the
NAIRU, inflation always rises, and when unemployment climbs above
the NAIRU the result is disinflation and finally deflation. By resorting
to the interest-rate tool, monetary policy is always able to stop both

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Eckhard Hein, Thorsten Schulten and Achim Truger 71

cumulative inflationary and deflationary processes and to bring about a


stable inflation rate, according to this view.
The post-Keynesian approach presented in this chapter has for several
decades already been arguing the case for the endogeneity of money in a
modern credit money economy as recently 'discovered' by the new-

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Keynesian consensus models (Kaldor 1970, 1982, 1985; Lavoie 1984,
1992: 149-216, 1996; Moore 1988, 1989): the central bank's control
instrument in a credit money economy is the key interest rate, and the
money supply arises endogenously through commercial banks supply-
ing creditworthy credit demand at a given rate of interest and the cen-
tral bank accommodating the required cash. In such a model, the price
level and hence inflation or deflation cannot be determined by the
quantity of money supply. In an imperfect commodity-market scenario,
the price level is instead the result of mark-up pricing on unit costs.
Post-Keynesian research has put forward various theories with regard
to the underlying unit costs (full costs or variable costs) and the factors
determining the mark-up (competition, internal finance requirements,
interest rate).6 One simple version of this approach, which draws on the
work of Kalecki (1954: 11-27), suggests that businesses in the industrial
sector of a closed economy set their prices by charging a mark-up on
unit labour costs, which are taken to be constant until full capacity out-
put (Hein 2004: 178-87). The size of the mark-up is determined on the
one hand by the degree of price competition on the commodity markets
and on the other by the extent to which the trade unions are able to
achieve significant nominal wage increases when profit levels are high.
If the size of the mark-up is fixed, then it is unit labour costs that deter-
mine price levels.
Cumulative inflationary processes come about if the trade unions
attempt to increase employees' share of the national product by negoti-
ating nominal wage increases that exceed the neutral scope for distribu-
tion given by the sum of productivity growth and inflation, and when
businesses are able to pass these increased unit costs on to consumers by
raising prices. However, upward pressure on inflation also arises when
businesses attempt to increase their mark-ups7 or when the bargaining
parties fail to foresee a fall in productivity growth. Disinflation or defla-
tion arise if wages policy is either unwilling or unable to make full use of
the growth in productivity plus inflation.
Nominal wage growth will now be affected by the employment or
unemployment rate, in particular, and also by the degree of wage-
bargaining coordination with a higher degree of coordination being
conducive to stable unit labour costs in the face of upward but also

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
72 Deflation Risks in Germany and the EMU

downward movements in employment. 8 In post-Keynesian models, the


employment rate depends in both the short and long term on effective
demand for goods, which is governed mainly by private investment, the
level of which is in turn determined by the ratio of the expected profit
rate to the monetary interest rate. In contrast to the new-Keynesian

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
'new consensus models', the post-Keynesian approach sees no reason to
assume that the unemployment rate determined by the commodity
market will adjust to the NAIRU (Sawyer 2001, 2002). On the contrary,
the post-Keynesian model implies that at best the NAIRU constitutes a
short-term employment barrier enforced by monetary policy. But if
unemployment exceeds the NAIRU, there is no guarantee that expan-
sive monetary policies are sufficient to stimulate the economy when
profit expectations are low and firms are hit by debt-deflation.9
According to Keynes (1936: 262-71), price levels can be expected to
drop, albeit not necessarily to the same extent owing to specific price
rigidities in the commodity market, if sustained high unemployment
results in falling nominal wages or unit labour costs.10 If reductions in
unit labour costs are not fully passed on to consumers in the shape
of price cuts, the result is also a redistribution at the expense of wage
earners and a concomitant fall in this group's consumption demand.
However, if domestic prices fall in an open economy, the balance of
trade improves assuming the Marshall-Lerner condition to be fulfilled.
But this short-run improvement is likely to be counteracted by a nomi-
nal appreciation of the domestic currency or by nominal wage modera-
tion and hence real devaluation abroad, so that the overall effect of
falling unit labour costs on foreign demand is quite uncertain in the
medium run.
In order to assess the effects of disinflation or deflation on invest-
ment, the development of interest rates, debt and profit expectations
have to be taken into account. Here, the effect of falling wages and
prices on the interest rate postulated by Keynes (1936: 263) as a result
of falling transactions demand for money can only come about if the
money supply is largely exogenously determined and does not adjust
endogenously to the demand for money through credit creation or
destruction and therefore constitutes net wealth.11 However, this is not
the case in modern credit money economies where money is largely
created by creditor-debtor relationships and every asset therefore has a
corresponding liability. In this scenario, falling wages and prices can
only affect interest rates in the event of a discretionary key interest rate
cut by the central bank. Even in such cases, however, the potentially
expansive effect on investment (and consumption) is counteracted by

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Eckhard Hein, Thorsten Schulten and Achim Truger 73

the fact that in a credit money economy where prices are falling, there is
a redistribution of wealth from debtors to creditors with the associated
risk of over-indebtedness. This debt deflation effect that was accorded
central importance by Fisher (1933) and as well by Keynes (1936: 264)
serves to dampen investment (and consumption) if the realistic assump-

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
tion is made that creditors are less inclined to spend than are debtors.
Furthermore, it is more difficult to obtain credit to finance spending in
a debt deflation scenario, since banks' and financial intermediaries'
lending policy is determined by the creditworthiness of firms (and
households) applying for loans, and their indebtedness is an important
indicator of how creditworthy they are.12 Taking these considerations
into account, the new-Keynesian view of a symmetrically effective mon-
etary policy which is always capable of adjusting actual unemployment
to the NAIRU seems to be overly optimistic and the neglect of the stabi-
lizing role of nominal wage rigidities seems to be unwarranted.
If one realistically assumes the characteristics of a modern credit
money economy as described in the post-Keynesian approach to hold, it
can thus be said that in times of recession, rigid nominal wages are
the anchor to prevent deflationary processes, even if the monetary
policy response also favours growth and employment. Consequently,
any study of deflationary tendencies should pay particular attention to
wages policy and unit-labour-cost trends. This does not mean, however,
that monetary and fiscal policy are completely off the hook. On the one
hand, they should be used preventively to ensure that cumulative
downturns and the associated danger of the removal of the wage anchor
never come about in the first place. Moreover, decisive use of monetary
and fiscal policy should be made to combat downturns that are already
underway, thereby helping wages policy to fulfil its role as a nominal
stabilizer.

Inflation and unit-labour-cost growth

The previous section described how post-Keynesian models work on the


assumption that in imperfect commodity markets, prices come about
principally as a result of a mark-up being added to unit variable costs
with unit labour costs being a major part of these costs. If we accept this
as true, then the inflation rate ought to be mainly determined by
unit-labour-cost growth. Of course, starting from a theory of conflict
inflation, we do not expect that unit-labour-cost growth is the only
determinant of inflation, because in this view inflation in a real-world
economy will also be affected by distribution claims of the state through

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
74 Deflation Risks in Germany and the EMU

variations in the net tax rate, of foreign producers through changing


import prices or exchange rates, and of domestic firms through varia-
tions in the mark-up.13 However, in this section it will be briefly demon-
strated that a close - but incomplete - relation between unit-labour-costs
growth and inflation is in fact backed up by the empirical data between

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
1961 and 2003 for Germany and for the member countries of the EMU.
Following on from this, the consequences for functional income distri-
bution will be assessed as well and differences between Germany and
the EMU countries in the 1990s when the process towards EMU was
completed are highlighted.
Figures 4.1 and 4.2 show the percentage increase in unit labour costs
and the inflation rate in Germany and the EMU countries for the period
between 1961 and 2003. There is clearly a relatively close correlation
between the two values, in particular for the EMU countries but also for
Germany. Although the unit-labour-cost growth curve seems to show
more pronounced fluctuations than the fairly smooth inflation curve,
both curves nevertheless exhibit a rising trend until the mid-1970s and
a downward trend since then or since the early 1980s. The fluctuations
of unit-labour-cost growth are around the inflation rate. The deviations
of unit-labour-cost growth from inflation are more pronounced in the
upwards direction until the mid-1970s and more pronounced in the
downward direction since the early 1980s.
Sylos-Labini (1979) has presented a rationale for this partial adjust-
ment in an oligopolistic pricing framework for a specific industry char-
acterized by uniform output prices: Unit labour costs are not only
affected by variations in nominal wages which are uniform for all firms

14-i — o- • Unit-labour-cost growth


^ —*— Inflation rate (consumer prices)
12

10

8 - . \ P
o
6 \ I M
CD

/ /
^
V« \ /
i—i i i i i - - ' • I I I
r1 r1
CO
2 Jo>
CO
CD
CD
ID
CD
CD
CD
CD
CD
CO
CD CD
CO
CD
in
CD CD
o>
CD
00
CD
CO
CO
CD
m
00
CD
00
CD
CD
CO
CD
CD
CD
CO
CD
CD
in
CD
CD
tf
CD
CD
CD
CD
CD
CO

Figure 4.1 Unit-labour-cost growth and inflation rate (consumer prices) in


Germany, 1961-2003 (per cent)
Source: OECD (2003).

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Eckhard Hein, Thorsten Schulten and Achim Truger 75

o- - Unit-labour-cost growth
•*— Inflation rate (consumer prices)

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
i i i i i i i i i i i I'-'I i i i i i i i i i
- i - c o m i ^ c D T - c o i n i ^ - c D T - c o L n i ^ c D - i - c o m i ^ c D T - c o
C O C D C O C O C O I ^ I ^ I ^ I ^ I ^ O O O O O O O O O O C D C D C D C D C D O O
C D C D C D C D C D C D C D C D C D C D C D C D C D C D C D C D C D C D C D C D O O
• l - T - - l - T - l - T - l - T - - l - l - l - l - t - T - T - l - r - T - T - T - C V J C M

Figure 4.2 Unit-labour-cost growth and inflation rate (consumer prices) in EMU,
1961-2003 (per cent)
Source: OECD (2003).

within an industry, but also by productivity which is different between


firms, the price-setting firm being the one with the highest rate of pro-
ductivity growth. When nominal wage increases are completely shifted
to prices by the price-setting firm, the other firms and therefore the
industry as a whole cannot completely shift due to inferior productivity
growth. In a dynamic context of increasing unit-labour-cost growth, the
average growth rate of unit labour costs will therefore exceed inflation.
On the other hand, when nominal wages fall, the price-setting firm will
only have to decrease prices according to the lower productivity growth
of its competitors. In a dynamic context of falling unit-labour-cost
growth, this implies that average unit-labour-cost growth will now be
lower than inflation. We therefore get a partial adjustment of prices to
changes in unit labour costs for the industry as a whole based on an
asymmetric adjustment of the price-setting firm.14
In the case of Germany and the countries of the EMU, this partial
adjustment also seems to be valid for the economy as a whole and has
had an impact on functional income distribution which is shown in
Figure 4.3. In the period of high and accelerating unit-labour-cost
growth-under the conditions of high employment - from the early
1960s until the recession in the mid-1970s, partial adjustment of infla-
tion meant a rising tendency of the labour income share,15 in Germany
as well as in the other EMU countries. Profit margins declined and so did
the profit share. Since the early 1980s when under the conditions of
high and rising unemployment unit-labour-cost growth started to
decline considerably, partial adjustment of inflation has meant a

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
76 Deflation Risks in Germany and the EMU

78.0 Euro area - A - Germany


76.0
74.0
_ 72.0
§ 70.0 A A* ft

o 68.0

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
°" 66.0
64.0
62.0
60.0 I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I
O C M ^ t C D O O O C M ^ l - C D O O O C M ^ - C O O O O C M M - C O O O O C M
c D c o c o c D c i D r v - r ^ r ^ r ^ r ^ o o o o o D O D o o o j o j a j o j o j o o
0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 0
l - T - l - T - l - T - l - l - l - T - l - l - l - l - l - T - l - T - T - l - C M C M

Figure 4.3 Labour income shares in Germany and the EMU, 1960-2003 (per cent
of GDP at current factor costs)
Source: European Commission (2004).

tendency of the labour income share to fall, in Germany and also in the
EMU. Profit margins have increased and the profit share has recovered.
The close but imperfect correlation between the unit-labour-cost
growth rate and the inflation rate suggested by purely graphical analysis
can be confirmed statistically using regression analyses. If the inflation
rate is regressed on the unit-labour-cost growth rate it can be seen that
unit-labour-cost growth exerts a statistically significant positive influ-
ence on inflation both in Germany and the EMU. This influence is
greater in the case of the EMU than for Germany alone, since while in
Germany a 1 percentage point increase in unit-labour-cost growth leads
to a 0.39 percentage point rise in inflation, it leads to a 0.82 percentage
point rise in inflation in the EMU. The coefficient of determination (r2)
for Germany stands at 45 per cent, while it is as high as 85 per cent for
the EMU. If the previous year's unit-labour-cost growth rate is used as an
independent variable (this is a theoretically valid approach owing to
delays in the cost-based price adjustment by businesses), the impact of
unit-labour-cost growth on inflation is confirmed. For Germany a 1 per-
centage point increase in unit-labour-cost growth leads on average to a
0.44 percentage point rise in inflation in the following year. For the EMU
the figure becomes 0.79 percentage points. The extent to which increases
in unit labour costs explain increases in inflation is significantly higher
for Germany if the previous year's figure is taken into account, since this
gives an r2 coefficient of determination of 58 per cent. For the EMU the
figure becomes only slightly worse and amounts to 78 per cent.
The differences in adjustment of inflation to unit-labour-cost growth
between Germany and the EMU countries as a whole may be due to the

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Eckhard Hein, Thorsten Schulten and Achim Truger 77

different degrees of openness to foreign competition (Sylos-Labini


1979). Firstly, the more open an economy is, the higher will be the pro-
portion of imported raw materials and unfinished goods costs in unit
variable costs so that changes in unit labour costs have a smaller impact
on inflation than in closed economies with a smaller share of imported

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
raw materials and unfinished goods. Secondly, the more open an econ-
omy to foreign competition, the more difficult it is for domestic
producers to shift domestic cost increases to prices without loosing
international market shares. In the case of rising domestic unit labour
costs, foreign competition therefore acts as a brake upon prices. Since
the EMU economies as a whole are less open than the German economy
it comes with no surprise that we find a closer relation between unit-
labour-cost growth and inflation in the former than in the latter.
The statistically close correlation described between unit-labour-cost
growth and inflation clearly does not yet establish that it is increases in
unit labour costs that cause inflation to rise. It would in principle be
possible to imagine that the correlation could be the other way round,
that is that unit-labour-cost growth is driven by inflation. Indeed, the
regression analysis for Germany and the EMU does show a significant
degree of inverse correlation, although it is also true that the r2 value is
substantially lower. However, it is our belief that this inverse correlation
can be ruled out for two reasons. Firstly, it would cause theoretical prob-
lems in the framework of the post-Keynesian approach pursued in this
chapter, since if money is endogenous the model would no longer have
an explanation for price levels. Secondly, the results of a Granger causal-
ity test,16 even if we accept all the limitations of such tests, offer much
stronger support for the assertion that unit-labour-cost growth does in
fact influence inflation and not vice versa.17 It can thus be claimed both
theoretically and on the basis of empirical data that in both Germany
and Europe, an overall downward trend in unit-labour-cost growth since
the mid-1970s led to a similar downward trend in inflation.
If we take a closer look at the development since the early 1990s, the
final phase of European monetary integration, it can be noted that after
a brief period of more rapid growth at the beginning of this decade,
since 1995 unit labour costs in Germany have risen by consistently less
than in the EMU, with their annual growth remaining on average some
1.5 percentage points below the EMU average (see Figure 4.4). A simi-
larly clear difference is evident in Germany's inflation rate, which has
remained consistently below that of the EMU by an average of 0.6 per-
centage points per year (see Figure 4.5). In absolute terms, the average
inflation rate in Germany over the whole period in question was

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
78 Deflation Risks in Germany and the EMU

A-Germany -»-EMU

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Figure 4.4 Unit-labour-cost growth in Germany and the EMU, 1991-2003
(per cent)
Source: OECD (2003).

- A - Germany -•— EMU


/
•--..

^*—^A~T *""^»^^^

V
" A '

1 1 1 1 1 1 1 1 1 1 1
-

i- CM r» 00 O i- CM CO
O)
O)
O)
O)
O) O) o
o o
o
O
o oO
O) O)

Figure 4.5 Inflation rate (consumer prices) in Germany and the EMU, 1991-2003
(per cent)
Source: OECD (2003).

approximately 2 per cent, but this figure includes the unusually high
rates of 4 per cent experienced during the boom following reunification,
and the average figure since 1995 has been just 1.4 per cent. In contrast
to this, the average EMU inflation rate for the whole of the period being
examined was 2.8 per cent, reaching a high of almost 6 per cent at the
beginning of the 1990s, but falling to an average of just 2.1 per cent
since 1995. As will be argued below, the differences in unit-labour-cost
growth and inflation between Germany and the EMU as a whole cause
major macroeconomic problems and deflationary risks under the pres-
ent conditions of slow growth and low inflation in the currency union.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Eckhard Hein, Thorsten Schulten and Achim Truger 79

Wage trends and collective bargaining

The downward trend in the annual unit-labour-cost growth rate


observed in the EMU during the 1990s could theoretically be explained
either by particularly stringent wage restraint or by increasing produc-

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
tivity growth. However, since the average annual productivity growth
rate in the EMU during the 1990s stood at 1.5 per cent compared with
1.7 per cent during the 1980s (European Commission 2003: 52), it can
be seen that this figure has changed very little, and consequently it is to
wages policy that we should look for an explanation of this trend.
There has in fact been a clear downward trend in labour cost growth.
While during the 1980s the nominal remuneration per employee18 in
the EMU rose by an annual average of 6.9 per cent, the figure during
the 1990s was only 3.6 per cent (European Commission 2003: 88).
During the course of the 1990s, the employee remuneration growth rate
fell almost continuously and it was only at the end of the decade that a
slight upward trend emerged once more (see Figure 4.6).
This downward trend in the employee remuneration growth rate indi-
cates that in many European countries, against a background of sus-
tained mass unemployment, the collective bargaining power of trade
unions was substantially weakened. The most visible indicators of this
were falling trade union membership and a significantly lower number
of strikes and industrial disputes (Boeri et al. 2001; Schulten 2004). In
addition, the 1990s saw the emergence in many European countries of
new corporatist competitive structures which, as a result of national
social pacts and 'alliances for jobs', led to the trade unions becoming

-EMU --A--Germany
10
8H
c
CD 6

Figure 4.6 Remuneration per employee* in Germany and the EMU, 1991-2003
(annual increase in per cent)
* Wages and salaries plus employers' contribution to social security
Source: European Commission (2003).

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
80 Deflation Risks in Germany and the EMU

firmly tied into the political agenda and committed to a competitive


wages policy (Fajertag and Pochet 2000).
Since the mid-1990s, the annual nominal collectively agreed hourly
wage increases in the EMU have remained constantly below 3 per cent
(see Table 4.1). Interestingly, in most years actual hourly earnings rose

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
more rapidly than collectively agreed wages, resulting in a positive wage
drift for the EMU as a whole. In general, the scope for distribution
derived from the sum of productivity gains and inflation was clearly not
fully exploited by the growth of employee remuneration in the EMU
during the second half of the 1990s, and this resulted in disinflationary
tendencies. In contrast, labour cost trends in the EMU at the start of the
twenty-first century are typical for a cyclical downturn, since, owing to
the sharp fall in productivity growth, they are increasing at a rate that is
slightly higher than the scope for distribution.19
It should be pointed out, however, that wage trends in the individual
EMU countries were by no means uniform during the 1990s, and in fact
reflected the occasionally major differences in economic growth and
employment trends between countries. Wage increases were distinctly
higher than the EMU average principally in some of the smaller EMU
countries that achieved especially dynamic economic growth, such as
Ireland, the Netherlands and recently also Spain. This contributed to
higher than average inflation as a result of these countries exceeding the
national scopes for distribution, in some cases by a considerable margin
(Schulten 2002).
The situation was somewhat different in the larger EMU countries,
that is in France, Italy and Germany. While overall wage increases in
Italy were slightly higher than the EMU average and slightly lower
than the EMU average in France, in Germany they have remained
consistently below the EMU average since 1996 (see Figure 4.6).
Germany has thus been pursuing the most moderate wages policy in the
EMU for some eight years, and given that it is the largest economy in
the EMU, this has exerted a downward pressure on EMU average
wage increases.
The particularly low wage increases in Germany can firstly be attrib-
uted to a lessening of the trade unions' bargaining power. While at the
start of the 1990s the trade unions were still able to achieve exception-
ally high collectively agreed wage settlements on the back of the boom
following German reunification, since 1996 at the latest their collective
bargaining policy has been plunged into a major crisis and they have
been forced to accept collectively agreed wage increases of under 3 per
cent and on occasion even under 2 per cent (see Table 4.2).20

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect -
Table 4.1 Wage trends and extent to which the scope for distribution is exploited in the EMU

Scope for distribution^


Indicators of wage trends1 inflation +producitivity growth1 Extent to which
scope for
Collectively Actual Labour distribution is
agreed wages earnings per productivity exploited by
per employee employee Remuneration per Scope for employee
hour hour Wage drift2 per employee Prices3 employee distribution4 remuneration5

2014-03-15
1996 2.7 3.0 +0.3 2.3 2.2 1.1 3.3 -1.0
1997 2.3 2.6 +0.3 1.9 1.6 1.5 3.1 -1.2
1998 2.1 1.9 -0.2 1.2 1.1 1.1 2.2 -1.0
1999 2.3 2.5 +0.2 2.0 1.1 1.0 2.1 -0.1
2000 2.2 3.3 + 1.1 2.7 2.1 1.3 3.4 -0.7
2001 2.6 3.5 +0.9 2.8 2.3 0.2 2.5 +0.3
2002 2.7 3.3 +0.6 2.5 2.3 0.3 2.6 -0.1
2003 2.4 2.8 +0.4 2.4 2.1 0.2 2.3 +0.1

Notes: x increase to previous year in per cent; 2 difference between growth rate of actual earnings and growth rate of collectively agreed wages in
percentage points; 3 harmonized consumer price index (HCPI), 4 inflation rate + productivity growth rate; 5 difference between growth rate of
employee remuneration and growth rate of labour productivity plus inflation rate in percentage points.
Sources: ECB, authors' own calculations.

oo

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim Truger
Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect -
Table 4.2 Wage trends and extent to which the scope for distribution is exploited in Germany 00

Scope for distribution^inflation Extent to which scope for


Indicators ofwage trends1 +productivity growth1 distribution is exploited2

By
Collectively collectively
agreed Actual Wage Employee Labour Scope for agreed By actual By employee
wages3 earnings3 drift4 remuneration3 Prices5 productivity3 distribution6 wages earnings remuneration

1992 12.0 9.1 -2.9 9.2 5.1 2.7 7.8 +4.2 + 1.3 + 1.4
1993 7.5 6.1 -1.4 5.8 4.4 1.6 6.0 + 1.5 +0.1 -0.2
1994 3.4 2.1 -1.3 3.1 2.7 2.6 5.3 -1.9 -3.2 -2.2
1995 4.9 4.5 -0.4 4.9 1.7 2.5 4.2 +0.7 +0.3 +0.7

2014-03-15
1996 2.6 3.0 +0.4 2.8 1.5 2.3 3.8 -1.2 -0.8 -1.0
1997 1.5 1.0 -0.5 1.6 1.9 2.0 3.9 -2.4 -2.9 -2.3
1998 1.9 1.4 -0.5 1.5 0.9 1.3 2.2 -0.3 -0.8 -0.7
1999 2.9 2.3 -0.6 2.0 0.6 1.5 2.1 +0.8 +0.2 -0.1
2000 2.0 2.8 +0.8 3.3 1.4 2.2 3.6 -1.6 -0.8 -0.3
2001 2.0 2.7 +0.7 2.5 2.0 1.4 3.4 -1.4 -0.7 -0.9
2002 2.7 2.1 -0.6 2.2 1.4 1.3 2.7 0.0 -0.6 -0.5
2003 2.0 1.2 -0.8 1.5 1.1 0.8 1.9 +0.1 -0.1 -0.4

Notes:x increase to previous year in per cent, 2 percentage points, 3 per employee hour; 4 difference between growth rate of actual earnings and growth rate
of collectively agreed wages in percentage points; 5 Federal Statistical Office consumer price index, 6 inflation rate + productivity growth rate.
Sources: Bundesbank, Federal Statistical Office, authors' own calculations.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim Truger
Eckhard Hein, Thorsten Schulten and Achim Truger 83

The crisis of trade unions' collective bargaining policy is shown even


more clearly by actual earnings trends than it is by collectively agreed
wage trends. In contrast to most other EMU countries, wage trends in
Germany in the 1990s were mainly characterized by a negative wage
drift, with actual earnings growing even more slowly than collectively

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
agreed wages. This means that the trade unions were unable to ensure
that the wage increases they had negotiated were actually implemented
in all companies.
In addition to the trade unions' loss of political power, the negative wage
drift in Germany is also a consequence of fundamental changes in the
structure and operation of the German collective bargaining system.21 One
clear sign of this is the decline in the number of companies and employees
covered by collective agreements that has been observed since the mid-
1990s (Schnabel 2003). According to the IAB (Institut fur Arbeitsmarkt-
und Berufsforschung) figures for 2001, only 48 per cent of all companies in
western Germany and 71 per cent of all employees were bound by collec-
tive agreements, while in eastern Germany the figures were as low as 28 per
cent of companies and 56 per cent of employees (Bispinck 2003: 395). The
negative wage drift seems to suggest that wage increases in companies not
bound by collective agreements were significantly lower.
Furthermore, even within the German collective bargaining system
there are numerous signs to suggest that the binding nature of collective
agreements is being eroded, making negotiated collective wage increases
harder to implement in practice and consequently favouring a negative
wage drift. There is now a significant number of companies that are
formally bound by collective agreements but which in practice do not
comply with them. According to the results of the 2002 WSI Works
Council Survey, which probably only covers part of the problem, 10 per
cent of companies occasionally failed to comply with the terms of cur-
rent collective agreements, and a further 5 per cent did so frequently. In
the majority of these cases, the non-compliance involved failure to pay
the collectively agreed wages (Bispinck and Schulten 2003: 159).
In addition to the above, 'hardship' and 'opening-clauses' were intro-
duced into virtually all of the major sectoral collective agreements in the
1990s, allowing companies to deviate from the terms contained in col-
lective agreements under certain circumstances.22 Opening-clauses are
now used by more than a third of all companies, although it is true that
in the majority of cases these relate to the divergence of working time
organization from the collective agreement, and the use of opening-
clauses with regard to remuneration is for the time being still not very
widespread (Bispinck and Schulten 2003: 160).23

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
84 Deflation Risks in Germany and the EMU

One final significant cause of the negative wage drift is the reduction
of payments that are above the collectively agreed rate. A large number
of companies in Germany continue to pay wages that are higher than
those established in their collective agreement. The results of the IAB
company panel show that although the number of companies paying

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
more than the collectively agreed rate did decline in the 1990s, it still
stood at 48 per cent in 2000 (Schnabel 2003: 95). The wage spread, that
is the absolute difference between collectively agreed wages and actual
wages was on average found to be approximately 11 per cent (Schnabel
2003: 95). Nevertheless, during the course of the 1990s, several compa-
nies began to use 'company alliances for jobs' to 'compensate for' the
wage increases negotiated in collective agreements by cutting back on
payments above the collectively agreed rate. This led to the emergence
of a new form of concession bargaining in which employees agree to
give up established benefits in exchange for limited job security, thereby
contributing to a substantial reduction in labour costs.
Since the mid-1990s, the collectively agreed wage settlements achieved
in practice by Germany's trade unions have no longer been sufficient in
most years to fully exploit the scope for distribution (see Table 4.2). The
negative wage drift also indicates that the significance of trade union
collective bargaining policy has waned considerably, with the result that
actual wage increases have fallen still further behind the sum of infla-
tion and productivity increases. Even if overall employee remuneration
in the 1990s rose by slightly more than actual wages did, there can be lit-
tle question that on the whole wages policy developments in Germany
had clear disinflationary repercussions and must as such take a large part
of the responsibility for the low inflation rate in the largest economy in
the EMU.

Conclusion: the dilemma of wage policies in


Germany and Europe

Despite the sluggish growth currently being experienced in the EMU,


wages policy and trends have not yet caused acute deflationary risks.
However, there is no guarantee that this will continue to be the case if
the restrictive macroeconomic policy mix of the past continues to be
pursued.24 One of the main causes of the sluggish economic growth is
the 'anti-growth bias' in the ECB's monetary policy with its inflation tar-
get of 'below, but close to 2 per cent' (ECB 2003: 89) which is far too low
for a heterogeneous currency area with markedly different growth and
inflation rates - not to mention the fact that it is asymmetrical in nature

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Eckhard Hein, Thorsten Schulten and Achim Truger 85

and is exclusively geared towards ensuring that the inflation target is


not exceeded. The growth-unfriendly effect of this monetary policy is
magnified by the Stability and Growth Pact that forces the European fis-
cal policy to be pro-cyclical and to target budgetary consolidation via
spending cuts, something that is ultimately to the detriment of public

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
investment. If the economic stagnation resulting from these monetary
and fiscal policies persists, it is quite possible that the associated high
unemployment could increase the pressure on wages policy, leading in
turn to an increase in wages policy-driven deflationary risks.
The danger of deflation is already considerably higher in Germany
than in the EMU as a whole, since the stagnation caused by monetary
and fiscal policy is aggravated by Germany's excessive wage restraint.25
The unit-labour-cost growth rate has for some time now been signifi-
cantly lower than the EMU average, and this is to a large extent respon-
sible for an inflation rate that is also much lower than average.
Consequently, even a monetary policy that might be suitable for the
EMU as a whole is too restrictive for a country where growth is as low as
in Germany. Furthermore, the fact that nominal interest rates are the
same across the EMU and inflation in Germany is below average means
that German consumers and investors are faced with real interest rates
that are higher than the EMU average. On top of this, excessive wage
restraint has led to a falling labour income share, which has in turn
further weakened domestic demand.
The combination of a pronounced trend towards stagnation and
significant deflation risks - not yet actual deflation - in the largest EMU
country together with the ECB's overly ambitious inflation target for the
EMU as a whole represents a major challenge for wages policy in
Germany and in the rest of Europe.26 If Germany is to achieve an eco-
nomic recovery with the aid of wages policy, both the unit-labour-cost
growth rate and inflation will need to rise. However, if such a rise leads
to an EMU inflation rate that is higher than the ECB's inflation target
owing to the fact that other EMU countries have inflation rates that
exceed the ECB target by a considerable margin, then restrictive mone-
tary policy intervention is always going to be on the cards. What this
means is that if the ECB is not prepared to raise its inflation target sub-
stantially in order to allow the slowly growing larger economies more
room to achieve a recovery, then it will be necessary to reduce inflation
in the other EMU countries. It is therefore important for the bargaining
parties and in particular the trade unions to intensify their efforts
towards European-level effective coordination of wages policy. The aim
of this process should be for each country to increase wages on the basis

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
86 Deflation Risks in Germany and the EMU

of its long-term domestic productivity growth figures plus the ECB's


target inflation rate.27
If it proves impossible either to convince the ECB to raise its inflation
target or to coordinate wages policy in the EMU countries as described
above, then in the medium to long run Germany's stagnation and defla-

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
tion risks are likely to spread increasingly to the other EMU countries.
Excessive wage restraint in Germany will not only fuel national eco-
nomic stagnation, but will also put pressure on wages policy in the other
EMU countries in the medium term. The fact that inflation in Germany
is lower than the EMU average means that price competitiveness of
German producers in the European market is constantly increasing. It is
true that in recent years, growing export surpluses have prevented
Germany from sliding into a deep recession. However, it also means
rising import surpluses for the other EMU countries, something that
cannot be sustained for any length of time owing to the negative effects
on income and employment. Since the EMU countries can no longer
resort to a currency devaluation, it is inevitable that sooner or later
there will be a wages policy response, as witnessed in the Netherlands,
where the recent wage bargaining round ended with a zero wage
increase (Schulten and Muhlhaupt 2003). However, if wages policy starts
to be widely used to improve price competitiveness, then further redis-
tribution at the expense of labour, rising effective demand problems and
the threat of deflation will spread accordingly. If this happens, then
even a more growth-friendly monetary policy by the ECB might be inef-
fective and in the next downturn the deflationary risks may become
actual deflation.

Notes
1 The report defined deflation as a sustained fall in the consumer price index or
GDP deflator. The common technical definition of deflation as a fall over the
course of two consecutive quarters was not considered to be sufficient (IMF
2003: 6).
2 For a contrasting position on the powerlessness of monetary policy to combat
deflation, see for example Buiter (2003).
3 Consequently, monetary policy should aim for a positive inflation rate, albeit
a low one, rather than zero inflation.
4 While the Bundesbank (2003) does not believe that Germany is
experiencing deflation risks, it nevertheless includes unit labour cost trends in
its considerations.
5 At the simplest level, the 'new consensus models' are based on three equa-
tions: an aggregate demand function derived from optimization calculations
of consumers and businesses, which describes the output gap as an inverse
function of the real interest rate, a Phillips curve in which the inflation rate

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Eckhard Hein, Thorsten Schulten and Achim Truger 87

has a positive correlation with the output gap, and a central bank reaction
function, which relates the nominal interest rate set by the central bank to
the equilibrium real interest rate, the output gap and the deviation of inflation
from the inflation target (Taylor rule).
6 See, among others, Eichner (1976), Wood (1975), Harcourt and Kenyon
(1976), Sylos-Labini (1969, 1979). For surveys of post-Keynesian price theo-

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
ries see: Lavoie (1992: 129-48, 2001) and Lee (1998, 2003).
7 One cause of this can be a monetary policy decision to raise interest rates.
This leaves businesses facing higher interest costs, which they attempt to
pass on by increasing their mark-ups (Hein 2004a).
8 See Hein (2002, 2004a) for an attempt to integrate wage bargaining institu-
tions into a post-Keynesian model of wages, employment and inflation. This
attempt relies on the work of institutional political economists deriving the
beneficial effects of effective wage bargaining coordination on macro-
economic performance, in particular in interaction with independent central
banks (Soskice 1990; Hall and Franzese 1998; Franzese 2001, 2001a; Kittel
and Traxler 2001; Hein 2002a).
9 Post-Keynesians have also argued that in the long term the NAIRU adjusts
endogenously to the actual unemployment rate through different channels
(Hein 2002, 2004).
10 See also Kalecki (1969: 55-9).
11 The same is true of the positive effect of falling prices on real wealth and
consumer demand proposed by neoclassical theory: in order for the Pigou
effect to come about, it is necessary for the monetary wealth of the economy
as a whole to be exogenously determined net wealth.
12 This has already been made clear by Kalecki (1937, 1954: 91-5) in his
'principle of increasing risk' according to which the firm's access to external
capital on capital markets is largely determined by its entrepreneurial capi-
tal. Investment is therefore limited by finance which is in turn inversely
affected by the degree of indebtedness. A similar view was taken by
Robinson (1962: 86) and Steindl (1952: 107-38). Recent empirical work has
shown that business investment is strongly influenced by internal funds
which determine the access to external borrowing on imperfect capital mar-
kets (see Fazzari, Hubbard and Petersen 1988; Hubbard 1998; Schiantarelli
1996).
13 Changes in the mark-up may either be caused by changes in the intensity of
price competition in the goods market or changing power relations between
capital and labour in the labour market, on the one hand. Or changes in the
mark-up may be due to changing overhead costs, i.e. variations in salaries or
in capital costs, on the other hand.
14 On the relation between unemployment, wages and functional income
distribution in the European Union, see more explicitly Hein and Schulten
(2004).
15 The labour income share is calculated as compensation per employee as
percentage of GDP at factor costs per person employed.
16 See Gujarati (1995: 620-4).
17 In our Granger test, for a lag of 1, the growth rate of unit labour costs for
Germany had a significance level of 1 per cent and the significance level
for the EMU was still 25 per cent, making it Granger-causal for inflation.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
88 Deflation Risks in Germany and the EMU

In contrast, the inflation rate was not Granger-causal for the unit-labour-cost
growth rate in either Germany or the EMU.
18 The remuneration per employee figure includes gross earnings and salaries as
well as non-wage labour costs, i.e. employer social security contributions.
19 Since the ECB data for collectively agreed and actual wage increase is calcu-
lated on an hourly basis but the data for labour productivity is calculated on

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
the basis per employee, it is unfortunately not possible to discuss to what
extend the scope for distribution was exploited by collectively agreed wage
increases.
20 See also Flassbeck and Maier-Rigaud (2003).
21 When calculating the even higher overall figure for the negative wage drift
per employee, changes in actual working time (e.g. overtime or short-time
working) are of particular importance. However, these factors are not taken
into account in the figure for wage drift per employee hour used here, in
order to enable us to concentrate purely on the structural aspects of collec-
tive bargaining policy. For more on the current debate concerning develop-
ments in the German collective bargaining system, see for example Bispinck
(2003), Bispinck and Schulten (2003), Schnabel (2003) and the contributions
in Wagner and Schild (2003).
22 For a more detailed analysis and description of the main hardship and open-
ing-clauses, see Bispinck and WSI Tarifarchiv (2003).
23 For more on the debate surrounding 'Company Alliances for Jobs', see the
contributions in Seifert (2002).
24 See Hein (2003) and Hein and Truger (2005, 2005a) for a detailed analysis of
the EMU's restrictive policy mix.
25 For a more detailed analysis of the causes of stagnation in Germany see Hein
and Truger (2005b).
26 On the interaction of the ECB's monetary policy with wage bargaining in
Europe see Hein (2002).
27 For more detailed information on the current status and future prospects of
the various trade union coordination initiatives, see Schulten (2003, 2004)
and Traxler and Mehrmet (2003).

References
Ahearne, A. et al. (2002) Preventing Deflation: Lessons from Japan's Experience in the
1990s, Board of Governors of the Federal Reserve System, International Finance
Discussion Papers no. 729.
Akerlof, G.A., Dickens, W.T. and Perry, G.L. (1996) 'The Macroeconomics of Low
Inflation', Brooking Papers on Economic Activity, 1: 1-76.
Arestis, P. and Sawyer, M. (2003) 'New Consensus', New Keynesianism and the
Economics of the "Third Way" ', in E. Hein, A. Heise and A. Truger (eds),
Neu-Keynesianismus - der neue wirtschaftspolitische Mainstream? (Marburg: Metropolis).
Bernanke, B. (1995) 'The Macroeconomics of the Great Depression: A
Comparative Approach', Journal of Money, Credit and Banking, 27: 1-28.
Bispinck, R. (2003) 'Das deutsche Tarifsystem in Zeiten der Krise - Streit um
Flachentarif, Differenzierung und Mindeststandards', WSI-Mitteilungen, 56:
395-404.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Eckhard Hein, Thorsten Schulten and Achim Truger 89

Bispinck, R. and Schulten, T. (2003) 'Decentralization of German Collective


Bargaining - Current Trends and Assessments from a Works and Staff Council
Perspective', WSI-Mitteilungen, special English issue, 56: 24-33.
Bispinck, R. and WSI-Tarifarchiv (2003) Tarifliche Oeffnungsklauseln. Fine Analyse
von Regelungen aus 80 Tarifbereichen, Reihe Elemente qualitativer Tarifpolitik 52
(Diisseldorf: Haus Boeckler Foundation).

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Boeri, T., Brugiavini, A. and Calmfors, L. (eds) (2001) The Role of Unions in the
Twenty-First Century: A Report for the Fondazione Rodolfo Debenedetti (Oxford:
Oxford University Press).
Buiter, W. (2003) Deflation: Prevention and Cure, NBER Working Paper Series
no. 9,623 (Cambridge, Mass.: National Bureau of Economic Research).
Clarida, R., Gali, J. and Gertler, M. (1999) 'The Science of Monetary Policy: A New
Keynesian Perspective', Journal of Economic Literature, 37: 1661-707.
Deutsche Bundesbank (2003) 'Zur Diskussion liber Deflationsgefahren in
Deutschland', Monatsbericht, 55(6): 15-28.
ECB (2003) 'The Outcome of the ECB's Evaluation of its Monetary Policy
Strategy', Monthly Bulletin (June): 79-92.
Eichner, A. (1976) TheMegacorp and Oligopoly (Cambridge: Cambridge University
Press).
European Commission (2003) European Economy, Autumn 2003 (Brussels: DG
Economy and Finances).
European Commission (2004) AMECO Database, DG Economy and Finances,
http://europa.eu.int/comm/economy_finance/indicators/annual_macro_
economic_database/ameco_en.htm.
Fajertag, G. and Pochet, P. (eds) (2000) Social Pacts in Europe - New Dynamics
(Brussels: ETUI).
Fazzari, S.M., Hubbard, R.G. and Petersen, B.C. (1988) 'Financing Constraints and
Corporate Investment', Brooking Papers on Economic Activity, 1: 141-95.
Fisher, I. (1933) 'The Debt-Deflation Theory of Great Depressions', Econometrica,
1: 337-57.
Flassbeck, H. and Maier-Rigaud, R. (2003), 'Auf der schiefen Bahn - Die deutsche
Lohnpolitik verscharft die Krise', Wirtschaftsdienst, 83: 170-7.
Franzese, R.J. (2001) 'Institutions and Sectoral Interactions in Monetary Policy
and Wage/Price-Bargaining', in PA. Hall and D. Soskice (eds) Varieties of
Capitalism. Institutional Foundations of Comparative Advantage (Oxford: Oxford
University Press).
Franzese, R.J. (2001a) 'Strategic Interaction of Monetary Policymakers and
Wage/Price Bargainers: A Review with Implications for the European Common-
Currency Area', Empirica, 28: 457-86.
Gujarati, D.N. (1995) Basic Econometrics, 3rd edn (New York etal.: McGraw-Hill).
Hall, PA. and Franzese, R.J. (1998) 'Mixed Signals: Central Bank Independence,
Coordinated Wage Bargaining, and European Monetary Union', International
Organization, 52: 505-35.
Harcourt, G. and Kenyon, P. (1976) 'Prices and the Investment Decision', Kyklos,
29: 449-77.
Hein, E. (2002) 'Monetary Policy and Wage Bargaining in the EMU: Restrictive
ECB Policies, High Unemployment, Nominal Wage Restraint and Inflation
Above the Target', Banca Nazionale del Lavoro Quarterly Review, 55: 299-337.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
90 Deflation Risks in Germany and the EMU

Hein, E. (2002a) 'Central Bank Independence, Labour Market Institutions and


the Perspectives for Inflation and Employment in the European Monetary
Union', Political Economy. Review of Political Economy and Social Sciences, 10:
37-64.
Hein, E. (2003) Voraussetzungen und Notwendigkeiten einer europaischen
Makrokoordinierung, in: Angelo, S., Mesch, M. (eds) Wirtschaftspolitische

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Koordination in der Europaischen Wahrungsunion, Wirtschaftswissenschaftliche
Tagungen der Arbeiterkammer Wien, 7 (Vienna: LexixNexis).
Hein, E. (2004) 'Die NAIRU - eine post-keynesianische Interpretation',
Intervention. Journal of Economics, 1: 43-66.
Hein, E. (2004a) Verteilung und Wachstum. Eine paradigmenorientierte Einfuhrung
unter besonderer Berucksichtigung der post-keynesianischen Theorie (Marburg:
Metropolis).
Hein, E. and Schulten, T. (2004) 'Unemployment, Wages and Collective
Bargaining in the European Union', Transfer. European Review of Labour and
Research, 10: 532-51.
Hein, E. and Truger, A. (2005) 'Macroeconomic Co-ordination as an Economic
Policy Concept - Opportunities and Obstacles in the EMU', in E. Hein, T. Niechoj,
T. Schulten and A. Truger (eds) Macroeconomic Policy Coordination in Europe and
the Role of the Trade Unions (Brussels: ETUI).
Hein, E. and Truger, A. (2005a) 'European Monetary Union: Nominal
Convergence, Real Divergence and Slow Growth?', Structural Change and
Economic Dynamics, 16: 7-33.
Hein, E. and Truger, A. (2005b) 'What Ever Happened to Germany? Is the
Decline of the Former European Key Currency Country Caused by Structural
Sclerosis or by Macroeconomic Mismanagement?', International Review of
Applied Economics, 19: 3-28.
Hubbard, R.G. (1998) 'Capital-market Imperfections and Investment', Journal of
Economic Literature, 36: 193-225.
Institute (2003) Arbeitsgemeinschaft deutscher wirtschaftswissenschaftlicher
Forschungsinstitute, 'Die Lage der Weltwirtschaft und der deutschen
Wirtschaft im Herbst 2003', DIW Wochenbericht, 70: 643-86.
International Monetary Fund (2003) Deflation: Determinants, Risks and Policy
Options - Findings of an Interdepartmental Task Force (Washington, DC: IMF).
Kaldor, N. (1970) 'The New Monetarism', Lloyds Bank Review, 97 (July): 1-17.
Kaldor, N. (1982) The Scourge of Monetarism (Oxford: Oxford University Press).
Kaldor, N. (1985) 'How Monetarism Failed', Challenge (May/June): 4-13.
Kalecki, M. (1937) 'The Principle of Increasing Risk', Economica, 4: 440-7.
Kalecki, M. (1954) The Theory of Economic Dynamics (London: George Allen &
Unwin).
Kalecki, M. (1969) Studies in the Theory of Business Cycles, 1933-1939 (London:
Basil Blackwell).
Keynes, J.M. (1936) The General Theory of Employment, Interest, and Money, The
Collected Writings of J.M. Keynes, Vol. VII, 1973 (London, Basingstoke:
Macmillan - now Palgrave Macmillan).
Kittel, B. and Traxler, F. (2001) 'Lohnverhandlungssysteme und Geldpolitik',
Wirtschaft und Gesellschaft, 27: 11-40.
Kromphardt, J. (2003) 'Lohnpolitik bei moglicher Deflation', Wirtschaftsdienst,
83: 501-8.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Eckhard Hein, Thorsten Schulten and Achim Truger 91

Lavoie, M. (1984) 'The Endogenous Flow of Credit and the Post Keynesian Theory
of Money', Journal of Economic Issues, 18: 771-97.
Lavoie, M. (1992) Foundations of Post Keynesian Economic Analysis (Aldershot:
Edward Elgar).
Lavoie, M. (1996) 'Horizontalism, Structuralism, Liquidity Preference and
the Principle of Increasing Risk', Scottish Journal of Political Economy, 43:

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
275-300.
Lavoie, M. (2001) 'Pricing', in R.P.F. Holt and S. Pressman (eds), A New Guide to
Post Keynesian Economics (London, New York: Routledge).
Lee, F. (1998) Post Keynesian Price Theory (Cambridge: Cambridge University
Press).
Lee, F. (2003) 'Pricing and Prices', in J.E. King (ed.), The Elgar Companion to Post
Keynesian Economics (Cheltenham: Edward Elgar).
Meyer, L.H. (2001) 'Does Money Matter?', Federal Reserve Bank of St. Louis Review,
83(5): 1-15.
Moore, B.J. (1988) Horizontalists and Verticalists: The Macroeconomics of Credit
Money (Cambridge: Cambridge University Press).
Moore, B.J. (1989) 'The Endogeneity of Credit Money', Review of Political Economy,
1: 65-93.
OECD (2003) Economic Outlook, 73, Data from CD-ROM (Paris: OECD).
Robinson, J. (1962) Essays in the Theory of Economic Growth (London: Macmillan -
now Palgrave Macmillan).
Sawyer, M. (2001) 'The NAIRU: A Critical Appraisal', in P. Arestis and
M. Sawyer (eds), Money, Finance and Capitalist Development (Cheltenham:
Edward Elgar).
Sawyer, M. (2002) 'The NAIRU, Aggregate Demand and Investment',
Metroeconomica, 53: 66-94.
Schiantarelli, F. (1996) 'Financial Constraints and Investment: Methodological
Issues and International Evidence', Oxford Review of Economic Policy, 12: 70-89.
Schnabel, C. (2003) Tarifpolitik unter Reformdruck, Benchmarking Deutschland
Aktuell (Giitersloh: Bertelsmann Foundation).
Schulten, T. (2002) Tarifpolitik in Europa 2001/2002 - 2. Europdischer Tarifbericht
des WSI, Reihe WSI Informationen zur Tarifpolitik (Duesseldorf: Hans Boeckler
Foundation).
Schulten, T. (2003) 'Europeanization of Collective Bargaining: Trade Union
Initiatives for the Transnational Coordination of Collective Bargaining', in
H.-W. Platzer and B. Keller (eds), Industrial Relations and European Integration.
Trans- and Supranational Developments and Prospects (Aldershot, Hampshire:
Ashgate).
Schulten, T. (2004) Solidarische Lohnpolitik in Europa. Zur Politischen Oekonomie der
Gewerkschaften (Hamburg: VSA).
Schulten, T. and Muhlhaupt, B. (2003) 'Nullrunden in den Niederlanden', Die
Mitbestimmung, 49(12): 44-7.
Seifert, H. (ed.) (2002) Betriebliche Buendnisse fur Arbeit (Berlin: Edition Sigma).
Snowdon, B., Vane, H. and Wynarczyk, P. (1994) A Modern Guide to Macroeconomics.
An Introduction to Competing Schools of Thought (Cheltenham: Edward Elgar).
Soskice, D. (1990) 'Wage Determination: The Changing Role of Institutions in
Advanced Industrialized Countries', Oxford Review of Economic Policy, 4:
36-61.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
92 Deflation Risks in Germany and the EMU

Steindl, J. (1952) Maturity and Stagnation in American Capitalism, 2nd edn. 1976
(New York, London: Monthly Review Press).
Sylos-Labini, P. (1969) Oligopoly and Technical Progress, 2nd edn (Cambridge:
Cambridge University Press).
Sylos-Labini, P. (1979) 'Prices and Income Distribution in Manufacturing
Industry', Journal of Post Keynesian Economics, 11: 3-25.

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Traxler, F. and Mehrmet, E. (2003) 'Coordination of Collective Bargaining: The
Case of Europe', Transfer. European Review of Labour and Research, 9: 229-46.
Wagner, H. and Schild, A. (eds) (2003) Der Flaechentarif unter Druck. DieFolgen von
Verbetrieblichung und Vermarktlichung (Hamburg: VSA).
Wood, A. (1995) A Theory of Profits (Cambridge: Cambridge University Press).

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
5
The Influence of Unemployment,
Productivity and Institutions on

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Real Wage Trends: The Case of
Italy 1970-2000*
Enrico Sergio Levrero and Antonella Stirati

Introduction
The aim of this chapter is to describe and attempt to interpret the trends
of wages in Italy in the period 1970-2000. While it is an applied work, it
may be useful to provide at the beginning a brief clarification of the
broader framework implicit in the analysis. On consideration of the ana-
lytical faults found in neoclassical substitution mechanisms as a foun-
dation for decreasing factor demand schedules, we do not approach the
explanation of wages by assuming that the economic forces underlying
their determination can be described by the interaction between labour
demand and supply functions. More generally, we envisage no tendency
of the economy towards full employment or NAIRU (non-accelerating
inflation rate of unemployment) equilibria, such as is still assumed in
the models that replace the traditional labour demand and supply
curves with 'pseudo-curves' based respectively on firm pricing rules and
models of wage determination. 1 In line with the classical approach and
its modern revival, we instead expect wage trends to be affected by a set
of historical and current circumstances that can be broadly classified as
labour-market conditions, the degree of organization of the parties

*A longer and partially different version of this chapter was published in


Economia & Lavoro, 2004, 1. We wish to thank S. Cesaratto, G. Olini, M. Pivetti
and P. Vicard for their comments as well as the participants at the 2003 annual
International Working Party on Labour Market Segmentation (IWPLMS) confer-
ence, the conference 'Sraffa o un'altra economia' (Rome 2003), and the confer-
ence 'Wages, Distribution and Growth' (Berlin 2004). The usual disclaimer
applies.

93

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
94 Unemployment, Productivity and Institutions: Italy

involved, and broader economic and political/institutional factors. All


these circumstances affect wage determination through the same
channel, that is their influence on the ability of the parties involved to
establish favourable conditions for themselves in the distribution of
income (Stirati 1992; Levrero 2006). There is thus no a priori hierarchy

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
in their respective role, although empirical and historical analysis
may obviously show that in specific contexts certain factors have been
more important than others, and can also indicate their reciprocal
influences.2
An emphasis on the role of those circumstances can be found in vari-
ous streams of current economic literature, even though the analytical
standpoints adopted often differ greatly from the one just outlined.
This suggests that they are empirically relevant in affecting wages. For
example, the empirical literature on the 'wage curve' (Blanchflower and
Oswald 1994) shows the connection between labour market conditions,
the unemployment rate in particular, and real wages across regions and
industries within a country; changes in the 'bargaining strength' of the
parties involved (in the form of shifts in the pseudo demand and supply
curves) have been indicated as one possible explanation of changes in
income distribution in Europe (Blanchard 1997); and the political situa-
tion has been regarded as a central factor in determining the unemploy-
ment rate and income distribution across countries (Korpi 1991).
Adoption of the classical standpoint implies, however, that such factors
as institutional changes are not expected to have only transitory effects
and are not regarded as 'disturbances' with respect to the underlying
forces of 'supply and demand'. For the same reason, institutional fac-
tors determining wage stickiness cannot be seen as the causes of unem-
ployment (the latter depending essentially on effective demand and
technical change, given labour supply).
The structure of the chapter is as follows. In the next section we
describe the trends of contractual wages and earnings, both gross and
net of taxes and employees' contributions, and discuss some regression
results as a means of assessing the major influences on wage trends. In
this connection, we also comment upon the observed scarce influence
of labour productivity and terms of trade on wages. In the following two
sections we first analyze in greater detail labour market conditions at
particular junctures and then broaden the analysis to look at the influ-
ence of institutional and economic changes. In the final section we sum-
marize our results and discuss some open questions in connection with
the literature.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Enrico Sergio Levrero and Antonella Stirati 95

Description of trends and some quantitative analysis

Although the rest of the chapter focuses solely on the period 1970-2000,
we shall look here at the data available on contractual wages since 1956
so as to provide some longer-term perspective and ground for compari-

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
son. We are primarily interested in the real wages of production workers in
the business sector. Within the latter, we focus on Industry and Trade,
since they account together for about 70 per cent of all employees and
appear to be reasonably homogeneous in terms of types of activities and
of employees.3
Examination of Figure 5.1 clearly shows the drive towards higher
wage increases in industry and trade in 1962-65, a subsequent slow-
down caused by a cyclical downturn, and then sustained growth in
1970-77 at an annual average rate of about 7 per cent. The average
growth rates then decline sharply, approaching zero in the 1990s
(1.3 per cent per annum in 1978-88 and 0.3 per cent in 1989-2000).
They were thus similar in the 1980s and lower in the 1990s with respect
to the 1955-61 period, usually regarded as one of trade-union weakness
and large labour reserves in backward agriculture and in the southern
regions.
Changes in gross earnings (which are an average of the earnings of all
employees, including white collars) in both Industry and Trade closely

Trade -•— Industry

Figure 5.1 Annual rates of change of real hourly wages of production workers in
trade and industry, 1956-2000
Sources: ISTAT, Sommario di statistiche storiche 1926-1985, and Lavoro e Retribuzioni (various
years).

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
96 Unemployment, Productivity and Institutions: Italy

180
160 H
140
120

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
100 H

80
60 H

40
Real gross wages, industry
20 Real net wages, industry
0 n—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i
C M ^ C O C O O C M ^ C O O O O C M ^ - C D C O O
r ^ h « . i ^ i ^ c o o o o o c o c o a > a > o > a ) o > o

Figure 5.2 Real gross and net wages in industry (1972 = 100), 1972-2000
Sources: OECD, The Tax-Benefit Position of Production Workers (various years), and ISTAT,
Contabilita Nazionale (2002b) (see also note 4).

follow those in contractual wages. Considering Industry alone, the


change in the trend occurring after 1977 is, however, even more marked
if we look at earnings net of income taxes and employees' contributions,
as shown in Figure 5.2.4 Net earnings rose about six points less than gross
earnings between 1976 and 1979 owing to the increasing incidence of
employee contributions and income-tax rates and to the phenomenon of
fiscal drag. Subsequently they even fell (10 points in 1979-83), regaining
their 1979 value only in 1988. After a very slight recovery, net earnings
have again performed worse than gross since 1992. They fell three per-
centage points more than gross earnings in 1992-95 (the aim of fulfilling
the 'Maastricht parameters' led to an increase in taxes on labour at a time
when real gross earnings were falling) and, unlike the latter, had not yet
regained the 1990 level in 2000 (see also Banca d'ltalia, 2001a).5
In order to assess possible explanations of the trends described above,
we have explored the correlation between (gross) wages and earnings in
Industry and Trade and a number of variables that could be expected to
prove influential, particularly productivity growth and labour-market
conditions. The latter in turn may be general, reflected in indicators
such as the unemployment rate,6 or specific to given groups of workers,
reflected in the unemployment rates of sub-sets of employees or in
sector employment trends.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Enrico Sergio Levrero and Antonella Stirati 97

There are two aspects to be noted at the start of this empirical


exploration. First, since simple correlation analysis shows that the mov-
ing averages of actual earning rates of change in each sector are strongly
correlated with those of contractual wages (the Pearson correlation coef-
ficients are R = 0.92 and R = 0.89 in Industry and Trade respectively),

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
we have chosen to discuss here the main factors affecting the rates of
change of contractual wages. Second, as can be gathered from Figure 5.1
above, there is a very close correlation between changes in contractual
wages in Industry and Trade (R = 0.96 for moving averages). This corre-
lation is greater than, and cannot therefore be simply attributed to, cor-
relation with a third factor (such as the general unemployment rate)
influencing both variables, and suggests a direct link between the two.
Since both are the result of national contracts, this link indicates that
there is harmonization between wage demands or wage concessions
made by the parties involved in the two sectors. Both data analysis and
historical experience suggest, however, that the industrial sector has
played a dominant role in wage determination. As regards the data, the
Pearson correlation coefficient between contractual wages and rates of
change in the sector's employees7 is in fact relatively high in Industry
(R = 0.7), while it is almost zero in the Trade sector (R = 0.03).8 Together
with the very high correlation between wages in Industry and Trade,
this suggests a leading role played by wage setting in Industry (see also
Brunello 1996). We can therefore regard industrial wages as our pivot
variable, affecting wages and earnings of employees in the business
sector as a whole.
Focusing thus on Industry, Table 5.1 shows the results obtained by
regressing the rates of change in real wages in Industry on the rate of
change in productivity and employees in the same sector, the general
unemployment rate, and two dummies for the years 1979 and 1993-95,
three-year moving averages being taken for all the variables.9 The over-
all unemployment rate appears to have played the major role in affect-
ing real wages in the 1970-2000 period (a similar influence being found
also in the previous decade).10 As discussed above, changes in industrial
employment also play a less important but still significant role. By contrast,
changes in productivity have a small and statistically non-significant
coefficient, while the 1979 dummy has a coefficient of -1.8 and is sta-
tistically significant: in that year there was a downward movement in
the rate of change in wages not fully accounted for by unemployment
and industrial employment trends. After 1977 wages also began to
grow less than industrial sector productivity, having grown (on average)
more than productivity as from 1969. Another change with respect to

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
98 Unemployment, Productivity and Institutions: Italy

Table 5.1 Determinants of rates of change in industrial wages

Dependent variable: rates of change in industrial real wagesa


Adjusted R square 0.86
Standard error 1.00
Observations 28

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
F 33.10
D.W. 2.61

Coefficients Standard error t-stat P-value

Intercept 9.71 1.03 9.41 0.00


Productivity5 0.10 0.10 0.99 0.34
1993-95d -1.51 0.61 -2.47 0.02
1979d -1.80 0.99 -1.81 0.08
Unemployment0 -0.78 0.10 -7.73 0.00
Employees5 0.44 0.13 3.26 0.00

Notes: a three years moving average; b three years moving averages of annual rates of change:
Employees measured in 'heads'; c three years moving average; d assumes value one in
1993-95, zero otherwise; assumes value one in 1979, zero otherwise.
Sources: ISTAT, Contabilitd Nazionale (2002b) for productivity and employees; ISTAT,
Rilevazioni delle Forze di Lavoro (various years) for unemployment; ISTAT, Sommario di
Statistiche storiche 1926-1985 and Lavoro e Retribuzioni (various years) for wages (see also
notes 3 and 16).

previous trends emerges in the early 1990s, namely a fall in contractual


real wages for the first time since the 1950s followed by stagnation.
Again, the coefficient on the 1993-95 dummy is negative (-1.5), and
statistically significant. Such decline in real wages, albeit moderate, is
quite a new phenomenon in the postwar period, and calls for enquiry.
Some of the factors discussed are clearly described by the scatter dia-
gram in Figure 5.3, which illustrates with immediacy the relationship
between (moving averages of) unemployment and real wage growth and
its changes in 1979 and 1993-95. The diagram also shows that the rela-
tion tends to become non-linear as growth in real wages approaches zero
for the well-known reason that decreases in real wages are generally dif-
ficult to bring about and, if they take place, tend to be very gradual.11
In the next section we shall look more closely at labour-market condi-
tions at those particular junctures. Before that, however, we shall discuss
the question of the influence of productivity on real wage trends a little
further. In this connection, it is advisable to extend the analysis to the
business sector as a whole. Actually, wage increases might not match
productivity growth in individual sectors, but productivity gains might

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Enrico Sergio Levrero and Antonella Stirati 99

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Figure 5.3 Unemployment and rate of growth of real wages in industry, 1960-99
(moving averages)
Sources: ISTAT, Rilevazioni delle Forze di Lavoro (various years) for the unemployment rate;
ISTAT, Sommario di Statistiche storiche 1926-1985 and Lavoro e Retribuzioni (various years) for
wages (see also notes 3 and 16).

be redistributed from workers in sectors with high productivity growth


to workers in other sectors, for example owing to a centralized bargain-
ing system ensuring equal wage increases across industries. It is worth
examining this point. Productivity growth evidently 'makes room' for
wage increases at a given rate of profit, and it has often been argued, on
different grounds, that changes in productivity tend to determine
changes in real wages.12 However, since the same effect upon the
'margins' for wage increases arises also from improvements in the terms
of trade, we should in fact ascertain whether productivity and terms of
trade have played a joint direct role in determining wage trends. 13
Table 5.2 shows the results of a regression similar to the one estimated
for Industry but referring to the Business sector as a whole and with rates
of change in actual real earnings as the dependent variable and terms of
trade as one of the independent variables. Changes in terms of trade
have a very small and statistically non-significant coefficient. The coef-
ficient of productivity is now statistically significant, albeit not very
large (one percentage point change in productivity is on average
associated with a half-percentage-point increase in real wages, other
things remaining constant). Indeed, in the Business sector as a whole, as
in Industry, until 1977, with a comparatively low unemployment rate,
wages tended to rise systematically more than productivity, despite a

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
100 Unemployment, Productivity and Institutions: Italy

Table 5.2 Determinants of rates of change in real earnings in the business sector

Dependent variable: rates of change in real earnings in the business sector a


Regression statistics
Adjusted R square 0.80

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Standard error 0.93
Observations 28
F 19.05
D.W. 1.55

Coefficients Standard error t-stat P-value

Intercept 7.39 1.43 5.15 0.00


1993-95 d -0.14 1.03 -0.14 0.89
Unemployment 3 -0.70 0.14 -4.86 0.00
1979 d -2.42 0.97 -2.49 0.02
Productivity 15 0.47 0.17 2.85 0.01
Terms of trade c 0.05 0.12 0.39 0.70
Employment (economy) 3 0.82 0.38 2.18 0.04

Notes: a three years moving average; b three years moving averages of annual rates of change;
c
three years moving average of annual rates of change of terms of trade index; d assumes
value one in 1979, zero otherwise; assumes value one in 1993-95, zero otherwise.
Sources: ISTAT, Contabilita Nazionale (2002b) for productivity, employees and earnings; ISTAT,
Rilevazioni delle Forze di Lavoro (various years) for unemployment; European Commission
(2002) for terms of trade (see also notes 3 and 16).

worsening of the terms of trade, while they rose systematically less


afterwards, when the unemployment rates were rising, notwithstanding
the improvement in the terms of trade during the 1980s.14 This does not
mean, of course, that the trends in labour productivity and the terms of
trade are irrelevant in accounting for the movements in money and real
wages. It does mean, however, that no mechanical ox a priori link can be
claimed between those variables and that the distribution of gains from
productivity growth or changes in the terms of trade actually depends
on the workers' bargaining position (Levrero 1999).

Broader view of labour-market conditions

As we have seen, the overall unemployment rate is the most relevant


variable affecting real wage changes in Industry as well as in Trade
and the private sector as a whole. We have also seen that in 1979 and
1993-94 there were changes in real wages that must be explained by

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Enrico Sergio Levrero and Antonella Stirati 101

other circumstances besides the variables included in the regression. For


reasons that will become clear in this and the next sections, these
circumstances had also an influence on subsequent wage and (particu-
larly after 1979) unemployment trends. In this section we shall take into
consideration the possible influence of changes in aspects of the labour

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
market not captured by the variables included in the regression, while
the following section will examine the broader economic and institu-
tional situation.
There is one aspect of the labour market situation in particular that
may have played a role in the changes in industrial relations that took
place around 1979, namely, the process of restructuring underway in the
large firms, which were at the core of union militancy and action, and
where employment was declining (in contrast with the still positive
trend for the sector as a whole) and many workers were made temporarily
redundant under the CIG (Cassa integrazione guadagui) (see note 8).
Employment fell, albeit moderately, in firms with more than 200
employees in the period 1978-80 (-1.2 per cent), continuing the
slightly negative trend of the period 1974-77 (-0.4). Similarly, employ-
ment fell in firms with more than 500 employees at annual rates of
about 1 per cent between 1975 and 1978, while total hours worked (net
of temporary redundancies) dropped by 4 per cent between 1977 and
1979 is This was due to technical innovations and the development of
what was called the strategy of 'decentralization of production', which
assigned some phases of production to smaller units that were only
nominally independent of the larger firms and characterized by greater
labour flexibility and a lower degree of conflict. Changes in employ-
ment and redundancies in large firms after the mid-1970s therefore
probably contributed to the changes in wage trends and industrial rela-
tions observed after 1977. Nor should we overlook the fact that while
employment growth was still positive, general unemployment did
increase between 1975 and 1977. While the figure involved may appear
moderate nowadays, it caused alarm at the time and may have contributed
to the changes in trade-union attitudes described in the next section.
As regards the fall in real wages in the 1990s following the devaluation
of the lira in 1992, it should be noted that the unemployment rate in
1992-93 and the subsequent years may not be a good indicator of what
was happening in the labour market for two reasons. One is a break in
the statistical series caused by a restrictive change in the definition of
the unemployed;16 the other is the very significant decrease in the num-
ber of employees for the economy as a whole in 1993-94 associated with
a drop of two percentage points in both male and female activity rates.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
102 Unemployment, Productivity and Institutions: Italy

It should also be noted that the Italian labour market underwent other
important changes in the 1990s that may have played a role in the
general weakening of the workers' position and the stagnation of real
wages and earnings during this period. First, there was an increase in the
number of jobs characterized by limited duration, made possible by

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
changes in legislation and accounting at the end of the 1990s for about
18 per cent of all employees in Trade and Industry. Second, there were
increasing flows of immigrant workers, accounting for about 4.8 per
cent of the total labour force, including irregular jobs, in the year 2000
(ISTAT 2002a). While these changes in the composition of the labour
force have a direct effect on average actual labour costs and earnings,17
it is not easy to evaluate their impact on union bargaining and contrac-
tual wages, especially in Industry and Trade. Combined with high juve-
nile unemployment rates, the particular high incidence of temporary
jobs among the youngest segment of the labour force has very probably
inhibited the young workers' potential for union militancy. On the
other hand, a union report suggests that the availability of immigrant
labour during the 1990s had no negative effect on contractual wages
except in Agriculture and, to a lesser extent, Construction (Olini 2003).
All in all, given that the changes described intensified in the second part
of the decade, it appears that they may have played more of a role in the
slow growth of wages in this latter period than in the fall in wages
between 1993 and 1995.
To sum up, labour market conditions not captured by changes in rates
of unemployment and employment growth appear to have contributed
to triggering the fall in real wage growth taking place in the late 1970s
and early 1990s, thus confirming the influence of the labour market sit-
uation on real wages. Their downward movement between (moving
averages centred in) 1978 and 1979 appears to have been favoured by
negative trends in employment in large firms, while the fall in real wages
after 1992 by a fall in the overall employment level and activity rates.
Both in 1977-79 and in 1992-95, however, broader changes in indus-
trial relations and the general institutional and economic framework
were also taking place, as discussed in the next section.

External constraints, macroeconomic policies


and institutional changes

Let us thus look first at the social and political situation, the pressure of
international competition, and economic policy decisions around the
years 1978-79, which facilitated the emergence of policies that to some

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Enrico Sergio Levrero and Antonella Stirati 103

extent contributed to the increase in unemployment and the reduction


in industrial employment in the subsequent years, and of a dramatic
change in industrial relations.
The fact that wages begin to grow less than productivity as early
as 1978 and that the rate of increase in wages feel sharply at the very

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
beginning of the recession in 1980, despite the good economic perform-
ance of the previous year, suggest that the roots of the fall in real wage
growth are to be sought before 1979. In particular, it appears possible to
trace them back to the period of the so-called 'consensual stabilization'
(Salvati 2000) in 1977-79, that is to the years that saw the end of the
previous phase beginning in 1969 characterized by real wages increas-
ing more than productivity and a sharp fall in the lira's nominal
exchange rate.
As regards wage bargaining, the 'consensual stabilization' was charac-
terized by the trade unions' acceptance of restraints in wage claims. An
initial reduction of the coverage of the wage-indexation clauses set up in
1975 was introduced in 1977 (severance pay was excluded from wage
indexation and some wage-goods were eliminated from the basket of the
cost of living index). Furthermore, the congress of the major workers'
organization (the CGIL) substantially accepted the idea that wages can-
not be an 'independent variable' and that high wages could conflict
with the aim of full employment and growth. With reference mainly to
the experience of other countries, an 'exchange' was thus proposed by
the CGIL (and accepted in 1978 by all the three major trade unions)
between wage moderation and the workers' participation in private and
public investment decisions.
Whatever the reason for the trade unions' new strategy - whether it
was the political prospect of the Communist Party's full participation in
government after its electoral peak in 1975-76 or the belief that wage
moderation would lead to an increase in the amount of employment 18 -
it substantially failed. On the one hand, owing to the vagueness of the
proposal and Italy's different political and historical context with
respect to countries like Austria, Germany and Sweden, no 'exchange'
actually took place to reward moderation in real wages.19 On the other,
by reducing conflicts at the workplace level, that new strategy permitted
a faster increase in productivity and employment fell in the large firms
forming the 'core' of the workers' organizations. Together with an
increase in the number of small firms (to which the larger firms increas-
ingly subcontracted certain phases of production) and in outsourcing
(see Heimler and Milana 1986), these processes undermined the trade
unions' strength precisely when a government plan ('Documento

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
104 Unemployment, Productivity and Institutions: Italy

Pandolfi') explicitly advocated restrictive monetary and fiscal policies.


Moreover, as in the case of income policy in the United Kingdom (Tarling
and Wilkinson 1977), wage restraint undermined worker militancy at the
very time of mounting dissatisfaction with the trade unions on the part
of the social groups and workers less protected against price inflation.20

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
One consequence of this phase of 'consensual stabilization' was to
facilitate Italian adherence to the Franco-German architecture of the
European Monetary System (EMS), which helped in turn to place organ-
ized labour under further pressure.
First, the worsening of the real exchange rate in the years 1979-80 in
terms of export prices (whose index rose from 89.8 to 93.5) and the
second 'oil shock' in 1979 contributed greatly to the fall in exports and
industrial production in 1980. This fall and the new exchange rate
regime set up in 1979 were regarded by Confindustria (the national asso-
ciation of entrepreneurs) as calling for a sharp change in industrial rela-
tions. Its first significant manifestation was in Fiat's new industrial
policy and anti-union attitude (Romiti and Pansa 1988; Lama 1987),
which led to the defeat of the metalworkers' trade unions (the leading
force in wage bargaining) in 1980 during the strike at the Fiat factories
against the dismissal of 14,000 workers. After the defeat, which marked
a turning point in industrial relations and highlighted the above-
mentioned difficulties of organized labour, the membership of the met-
alworking trade unions fell sharply,21 as did the numbers of working
hours lost through strikes (ISTAT 2002a).
Second, the monetary regime established by the EMS and the rise in
the real rates of interest in the United States necessitated an abrupt halt
in the rate of inflation, which was higher than in other countries
(12 percentage points higher than Germany in 1978 and 20 in 1979). In
line with suggestions put forward in OECD documents, 22 the Governor
of the Bank of Italy thus stressed the need for a change in the 'monetary
constitution' (Banca d'ltalia 1981 and 1982) and implemented restric-
tive monetary and credit policies associated with high nominal interest
rates. Together with the substantial fall in exports in 1980, this led to a
sharp decrease in gross private fixed investment in the years 1981-83
and a standstill in real GDP in the period 1980-83. In spite of this, not
only were the deflationary monetary policies continued but restrictive
fiscal measures were also implemented after 1981 and to a greater extent
than in 1976. The growth rate of public expenditure dropped and
reached a minimum in 1984 and 1985 (Giarda 1986), thus contributing
to the continuing decrease in industrial employment and the rise in
unemployment (see Figure 5.4).23

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Enrico Sergio Levrero and Antonella Stirati 105

-+- Unemployment rate - * - Male unemployment rate


- • - Industrial employees (heads), - A - Total employees (heads),
annual rate of change annual rates of change
12
10
8 ( X ) ^<

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
CD

CD

I I I fc-+»-U^>* I v \ J II I ^ * £ + * I I iW*
I

-4
CO LO CD CO LO CD CO LO CD
CO CO CO CO CO CD CD CD
CD CD CD CD CD CD CD CD CD CD a>
CD CD a>
CD CD CD

Figure 5.4 Unemployment rates* and annual rates of change in industrial and
total employees, 1971-2000
Notes: * Since 1993 the new definition of unemployment gives rise to a lower estimated
unemployment rate (see note 16).
Sources: ISTAT, Rilevazioni delle Forze di Lavoro (various years) for the unemployment rates;
ISTAT Contabilita Nazionale (2002b), for employees.

In addition to these, other measures were also introduced with the


same aim of slowing down the rate of inflation and helping firms to
cope with the pressure of international competition (the real exchange
rates worsened again in 1982-84). In order to increase labour productiv-
ity, public subsidies to firms were reduced and recourse was facilitated
to early retirement and temporary redundancy under the CIG. More
importantly, an agreement reducing wage-indexation by 15 per cent and
linking it during the first half of 1984 to a government-targeted infla-
tion rate was signed in 1983 by Confindustria and some of the unions,
but not by the CGIL, thus causing a deep rift in the trade-union move-
ment. The CGIL and the Communist Party called a referendum against
that agreement, and were defeated in 1985.
Anti-union practices and individual benefits and premiums became
widespread in the following years,24 while the divisions among the trade
unions persisted (as shown by several 'separate agreements' signed by
them). As a result of this situation and the rise in unemployment, high
real rates of interest and a change in distribution favourable to profits
were able to persist. After being negative for most of the 1970s and
becoming positive in 1981, the long-term real rate of interest rose to
3 per cent in 1983 and 6.5 per cent in 1987.
As regards the period 1992-96 (which marked another change in the
trend of real wages as shown in Figure 5.1), it can be said that the fall in

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
106 Unemployment, Productivity and Institutions: Italy

real wages during these years reflects the worsening of the workers'
bargaining position during the 1980s as a result of the social and politi-
cal factors mentioned above and the increase in the average rate of
unemployment (which rose to 10 per cent). Other factors also appear to
have been relevant, however.

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
The 1992 devaluation of the lira occurred after the fixed exchange rate
regime in 1988-92 (Italy's adherence to the 'narrow band' of the
European Monetary System) designed to discipline wage bargaining
(Pivetti 1999). As is known, the regime proved only partially successful
in this regard. On the one hand, there was a growing gap with respect to
France and Germany in the rate of change in the unit labour costs of the
private sector, arising both from a recovery of money wages in 1989-91
(fostered by a cyclical increase in industrial employment) and from
lower productivity growth than in other countries (see Table 5.3). On
the other, the money costs of production were increased also by the rise
in nominal interest rates decided upon by the Bank of Italy in order to
guarantee a surplus in the overall balance of payments after the full lib-
eralization of capital movements in 1988. There was thus a real appreci-
ation of the lira (in terms both of unit labour costs and of export prices)
and the balance of trade became negative in 1987-91. Manufacturing
firms were accordingly under increasing pressure not to raise their prices
and the situation led to an increase in the wage-income share of indus-
trial value-added and to Confindustria's unilateral suspension of the
price-indexation clauses.
The stringent fiscal measures introduced by the Amato government in
1992 to reduce the domestic product failed to prevent the currency cri-
sis brought about by these causes and fuelled by Germany's monetary
policy. (The Bundesbank had raised the interest rates and did not inter-
vene to defend the lira, which depreciated by 58 per cent with respect to
the mark and 30 per cent with respect to sterling.) Those measures
favoured, however, the fall in the unit labour cost even in terms of
the national currency, which was greater than in other countries. The
real exchange rates thus improved sharply, albeit more in terms of
unit-labour-cost rate than of export prices. Firms used the depreciation
to some extent to increase their profit margins (see Banca d'ltalia
1999a).
It can be stated that, as in other periods (see for example Graziani and
Meloni 1980), currency depreciation and deflationary policies were the
tools used to boost profits and improve the competitiveness of Italian
industry. Unlike the period 1971-76,25 both these aims were now

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Enrico Sergio Levrero and Antonella Stirati 107

Table 5.3 Rates of change (%) in hourly productivity, hourly labour costs, real
exchanges rates in manufacturing in the main industrialized countries
1979-2001 1979-85 1985-90 1990-95 1995-2000

Productivity per manhour

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
USA 3.4 3.5 2.4 3.3 4.7
FRA 3.7 3.0 3.4 4.0 4.6
GER 2.5 2.1 2.1 3.3 2.6
ITA 2.2 3.5 1.9 2.4 0.9
UK 3.6 4.4 4.6 3.3 2.4

Hourly labour costs


USA 4.6 7.2 3.9 3.5 4.0
FRA 5.9 12.8 4.5 3.9 1.9
GER 5.0 6.0 5.0 6.4 2.6
ITA 7.7 15.9 6.8 4.9 2.8
UK 7.9 12.2 9.4 5.4 4.6

Unit labour cost in national currency


USA 1.2 3.6 1.4 0.2 -0.7
FRA 2.2 9.5 1.0 -0.1 -2.6
GER 2.4 3.8 2.8 3.1 -0.1
ITA 5.4 12.0 4.8 2.4 1.9
UK 4.1 7.5 4.5 2.1 2.2

Real exchange rates


FRA -0.3 -3.3 11.6 1.7 -9.2
GER 1.6 -4.1 15.9 5.6 -7.6
ITA 0.9 -2.5 15.1 -3.7 -3.1
UK 2.3 -1.0 11.4 -0.4 1.3

Source: US Department of Labor (2002).

achieved. The depreciation of the currency was not accompanied by a


wage-price spiral; on the contrary, the rate of inflation decreased
slightly in 1992-94, and money wages rose less than prices.
Although the decrease in industrial and total employment during
the period 1992-96 certainly helped to keep money wages in check, the
inability of the latter to respond to price increases resulted also from the
reforms of wage settlements introduced by Amato in 1992 and 1993.
With the new wage-setting procedures, nominal wages were linked to
the target inflation rate, which was usually fixed at a level lower than
the actual (see Banca d'ltalia 2000a), while the task of linking wages to
productivity was assigned to firm-level bargaining, which never

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
108 Unemployment, Productivity and Institutions: Italy

developed extensively (it covered only 39 per cent of employees in firms


with more than 10 employees and 62 per cent of employees in firms
with more than 500 employees in the period 1995-96). Thus, the fact
that trade unions accepted the need to restrain wage claims on the
grounds of the Maastricht Treaty signed in 1992 appears to have played

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
a role in bringing about the fall in real wages in the years 1992-96, as
well as in keeping their rise at a lower rate than productivity in the years
1996-2000.26

Some final remarks

We shall now summarize our main results, comment briefly on some of


these, and indicate some open questions.
Our empirical analysis shows that the primary factor affecting real
wage trends in Italy in the 1970-2000 period appears to have been
unemployment, while productivity growth played no role in directly
affecting the rates of change of industrial wages and a minor one in the
trends of earnings in the business sector as a whole. There were, how-
ever, also broader factors involved in affecting wage trends. In particular,
the fall in wage growth after 1977 precedes the dramatic decline in
industrial production and industrial employment (and the parallel rise
in the unemployment rate) beginning respectively in 1980 and 1981. Its
roots lie in a weakening of the trade unions related both to the decrease
in employment in large industrial firms between 1975 and 1979 and to
the broader social and political context of the phase of 'consensual sta-
bilization'. After 1979 the decline in wage growth reflects, besides the
worsening of labour market conditions, the changes in the institutional
and socio-political situation in the previous years as well as the increas-
ing pressures to reduce inflation generated by the exchange rate
agreements in Europe and the rise in US interest rates.
In the 1990s we observe an unprecedented (in the postwar period)
vulnerability of real wages to the exogenous increase in prices after the
lira devaluation in 1992. This also must be seen in connection with
institutional changes, namely the new wage-setting procedures agreed
upon in the same year, combined with the unions' inability to ensure
the development of firm-level bargaining. This new wage-setting proce-
dures and the overall economic situation, characterized in particular
by the government aim of fulfilling the 'Maastricht parameters', con-
tributed to the stagnation of wages during the 1990s.
It may be worth commenting on some of the above points in the light
of the existing literature, namely the changes in wage behaviour in the

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Enrico Sergio Levrero and Antonella Stirati 109

1990s, the relationship between unemployment and real wages, and


the role of institutional factors.

• Changes in the characteristic behaviour of real wages. According to Boyer


(1979) the characteristic feature of the period from the end of the

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Second World War to the mid-1970s (but already beginning to estab-
lish itself in the first decades of the century) was the absence of
even transitory downward movements in real wage levels and their
tendency to grow at a fairly stable annual rate. In this respect, the
Italian experience in the 1990s may suggest the appearance of a new
scenario - or rather perhaps the return to an older one - in which real
wages, or their annual rates of change, may decrease during the neg-
ative phases of the economic cycle or as a consequence of changes in
the price level. Further investigation might establish whether this
pattern is common to other industrialised economies.
• A real-wage Phillips curve? The scatter diagram in Figure 5.3 suggests a
'real wage Phillips curve'. There are in fact some obvious analogies
with the original Phillips curve, namely the fact that it was also con-
cerned with average values of the variables (albeit derived with a
different procedure) rather then cyclical changes, and the non-linear
nature of the relationship, associated with a downward stickiness
of wages even with high unemployment. A further analogy is that
the relationship between unemployment and real wages, like that
between unemployment and money wages, can be interpreted as a
consequence of the influence of the former on the bargaining posi-
tion of the parties involved (Rothschild 1993: 129-30). It is, however,
interesting that the relationship holds for real wages, since it suggests
that changes in the bargaining position of workers related to
labour market conditions affected distribution, despite the ability of
employers to respond to money wage increases with price inflation,
at least in some phases.27 In general, however, we do not believe that
quantitative relations should be sought for predictive purposes (that is,
to predict the rate of change of wages associated with particular val-
ues of the unemployment rate or other variables). The shape and
position of such a relation actually depend on the broader institu-
tional and economic setting and are bound to change when changes
take place in the latter. Moreover, such changes in the broader con-
text may be triggered by changes in unemployment itself. Generally
speaking, one would not expect the effect of the unemployment rate
on the growth of real wages to be independent of the previous path of
these and other relevant variables.28 In addition to this, great care

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
110 Unemployment, Productivity and Institutions: Italy

should of course be taken not to generalize from a relation found to


hold in Italy over a certain period, which may not hold in different
countries or different periods. Further investigation in this direction
should contribute to a better understanding of the channels through
which unemployment may actually affect wage determination in

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
specific institutional settings.
Having said that, we believe that the strong correlation we have
found between average values of the unemployment rate and real
wage growth is an interesting result of our work. As regards Italy, it
actually contradicts the apparently widespread view that wages are
not affected by unemployment levels, which is probably based on the
results of some cross-regional studies.29
• An independent role for political and institutional factors? It has been
maintained that institutional factors play no role in real wage trends
in the very long run because the latter must reflect the underlying
forces of 'supply and demand' (see for example Phelps-Brown 1968
and Levrero 1999, for critical discussion). Apart from the theoretical
difference between the traditional view (shared by Phelps-Brown)
and our approach as regards the way in which labour market condi-
tions and institutional factors affect wages, our study suggests that
such an independent role was played in Italy, for example, by the
broad political and union situation in the years between 1976-79 and
that the effects of the changed institutional context and political
climate were largely felt in the following years through the high
unemployment generated by unconditional adherence to the EMS,
the more restrictive approach in fiscal and monetary policy, and
industrial restructuring. This is in agreement with a stream of contri-
butions, mainly from political scientists, maintaining that political
factors affect unemployment levels and income distribution (Hibbs
1977 and Alt 1985, among others). But the case of Italy also appears
to lend support to Korpi's (1991) warning that, while the political
context is a very relevant factor, a number of political and institu-
tional circumstances must be taken into account in addition to the
political orientation of government parties.

Notes
1 For recent critical discussion of neoclassical substitution mechanisms, see
Garegnani (2003); for analytical criticism of the tendency to the NAIRU,
see Petri (2003) and Petri (2004: chapter 7 and appendix 2).
2 Worsening labour-market conditions, for example, may trigger broader
changes in the institutional framework, such as a weakening of trade unions,
while on the other hand labour market conditions themselves are not

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Enrico Sergio Levrero and Antonella Stirati 111

independent of economic policies, which are influenced by the social and


political climate. In some instances, labour market conditions can also be
affected by the employers' direct response to conflict, the large-firm restruc-
turing and 'decentralization of production' carried out in Italy in the 1970s
being a case in point (see pp. 100-8).
3 Industry includes Manufacturing and Energy production but not

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Construction; Trade includes Catering and Hotels. Wages are contractual
hourly wages based on national collective agreements between unions and
employers. We shall use the series of contractual wages for production workers
only (not including white-collars). The source for aggregate earnings is
national account data on gross incomes of employees. Earnings are then
divided by 'standard labour units' defined as the normal working hours of a
full-time employee. Wages and earnings are deflated by the cost-of-living
index based on the typical consumption bundle of employees.
4 The real net earnings per standard labour unit in industry are derived by
applying to gross earnings the percentage of taxes and contributions paid by an
average Italian production worker as estimated by the OECD (various years).
Similar results can be obtained from the data of the Bureau of Labour Statistics
and also from the Bank of Italy's J Bilanci delle Famiglie Italiane (1989-2002).
5 Notice also that the increasing weight of contributions and income tax
on labour incomes was not compensated by greater social expenditure or
structural changes in the Italian fiscal system towards higher de facto progres-
siveness. Social expenditure in real terms has stagnated in the last decade in
particular, and there has been a worsening of the tax-benefit position for
workers. In spite of rising fiscal pressure (attenuated after 1998, but more to
the advantage of employers than employees) there has been a tendency to
reduce the progressive nature of the tax system, and the real social expendi-
ture index increased only from 100 in 1990 to 117.2 in 1999, having doubled
every decade during the 1960s and the 1970s (OECD 1985; EUROSTAT 2000).
6 Activity rates and employment rates are also interesting indicators of labour-
market conditions given their ability to pinpoint changes in disguised rather
than explicit unemployment. Systematic use has not been made of these
indicators due to some breaks in the statistical series.
7 We have chosen to measure employment in 'heads' here because the relevant
point for the bargaining strength of the parties involved appears to be the
number of actual people getting or losing jobs (rather than the total hours
worked). The drawback of this measurement is that it is gross of workers
temporarily made redundant under the CIG (a national fund financed by the
state and by contributions of employers and employees: the workers
made redundant maintain their employment relationship and receive a
subsidy).
8 Similar indications can be obtained from regression analyses, which show that
along with unemployment the other statistically significant influence on real
wage rates of change in the Trade sector comes from the rate of change in
industrial employment. Employment and productivity in the Trade sector are
instead not statistically significant.
9 We use moving averages because we expect the impact of economic factors on
wages to be greater if they persist over time. On the other hand, we are not
interested in very short-term variations in wage growth.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
112 Unemployment, Productivity and Institutions: Italy

10 Data not reported here show that wage changes are more correlated with the
overall unemployment rate than with the unemployment rates for specific
groups (male workers, north-central regions) and the rates of growth of
employment for the economy as a whole. It is worth noting, however, that
the correlation with the latter becomes very strong in the last decade. This
role of unemployment is almost surprising, as it might have been expected

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
that the high female and southern component of unemployment would
have a limited influence on union action, largely carried out in large firms
concentrated in the north of Italy. It is possible that this influence was indi-
rect, at least to some extent, that is related to a tendency on the part of
unions to regard unemployment as the result of the country's economic
difficulties and wage moderation as a factor that could help to overcome
those difficulties (see also pp. 102-8).
11 We have kept to the linear specification of our regressions for the sake of
greater simplicity and easier interpretation of the coefficients. The fit of the
equations is good also in this form, and the improvements to be obtained by
considering non-linearity are modest.
12 According to the traditional Marginalist theory, for example, an increase in
labour productivity due to technical innovation, with a given labour supply,
would be associated with an increase in the equilibrium wage level (unless it
were due to pervasive 'very labour-saving' technical change, according to
Hicks' classification). Again according to this theory, an increase in labour
productivity due to the change in the proportion of capital to labour, with
given technical knowledge, would determine an increase in real wages.
This would be the same size as that in productivity if the elasticity of substi-
tution between labour and capital were equal to one (see Blanchard 1997 for
a recent discussion of the latter properties). The same conclusions would
derive from standard current models of wage determination, combining
a 'wage curve' with a price equation, while in a different analytical con-
text, a tendency of wages to grow in step with productivity has been attrib-
uted to specific institutional arrangements, characterizing the period
from the end of the Second World War to the mid-1970s (Boyer 1979, among
others).
13 To avoid misunderstandings, it should be noted that those 'margins' set no
rigid limit to an increase in real wages, since the rate of profit can fall.
Technological progress and improvements in the terms of trade only 'make
room' for those increases by determining a rise in the rate of profit for a given
wage (or vice versa).
14 See Levrero and Stirati (2004: 71-4). For the 1970s and early 1980s, see also
OECD (1983).
15 Barca and Magnani (1989: 115), Banca d'ltalia (1977a, 1980a). In the
Industrial sector as a whole (firms of all sizes), the number of workers made
redundant under the CIG peaked in 1972, 1975 and 1978, and again, more
markedly, in 1980, while in 1979 there was a strong rise in employment.
16 The unemployment rate figures are taken from historical series reconstructed
by ISTAT for the period until 1985 and between 1994 and 2000. The annual
estimates from the Rilevazioni della Forze di Lavoro are used for the other years.
Because of a restrictive change in the definition of unemployment, since
October 1992 the statistical figures for the unemployment rate tend to be

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Enrico Sergio Levrero and Antonella Stirati 113

lower than with the previous definition. The difference between the
unemployment rate statistics based on the new and old definition was
estimated to be of 3.3 percentage points in 1993 (Banca d'ltalia, 1994a).
17 Limited duration contracts (including collaborations) entail as a rule lower
contributions and in some instances also lower pay. Immigrant workers tend
to be paid between 5 and 10 per cent less than the Italians, after controlling

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
for worker and job characteristics (Brandolini et al. 2003).
18 In a difficult political climate, the Communist Party put pressure on the
CGIL between 1977 and the first part of 1979 to accept wage moderation and
gave its 'external' backing to the Christian Democrat government. The CGIL
in turn regarded the experience of 'national solidarity' as a 'political' result
capable (at least in the following years) of opening up a phase of social
improvements for workers (CGIL 1981: 96-7). With respect to opinions as
regards the effects of wage moderation on employment, see Trentin (1980)
according to whom those years had already seen a 'cultural and political
hegemony' of the 'laissez-faire' tendencies influencing the policies of the
workers' organisations. The trade unions did in fact agree on the need for a
ceiling to the public deficit and for a wage policy 'coherent with the aim of
full employment' (CGIL 1981: 183 and 186-7).
19 It could be argued that wage restraint actually brought about an increase in
employment thanks to the rise in exports. This was, however, partly offset by
restrictive domestic policy (de Vivo and Pivetti 1980). Moreover, the trade
unions even failed in their attempt to make fiscal and financial provisions for
firms conditional upon their acceptance of an industrial policy based on
sector investment plans.
20 Apart from the effects of inflation on the real wealth of the middle class, the
'egalitarian' wage-indexation clauses of 1975 combined with non-uniformity
in the timing of wage negotiations and the different bargaining strength of
workers in the various sectors of the economy to generate dissatisfaction with
the three main trade unions (CGIL, CISL and UIL) amongst skilled workers,
white-collar workers and public-sector employees (whose real wages had
fallen in the period 1975-77).
21 Similarly, the overall rate of worker unionization fell from 49.3 in 1980 to
39.7 in 1986, 38.8 in 1990 and 35.8 in 1997.
22 See for example OECD (1977). See also Kaldor (1984: 113-69), who shows
that the explicit aim of the deflationary policies in the United Kingdom was
to weaken the workers' bargaining position.
23 The character of fiscal policies was instead not restrictive in 1979-81, due
among other things to the increase in family subsidies and the salaries of
public employees.
24 This is reflected by the emergence of a 'wage drift' in industry between 1983
and 1988 and its disappearance in the 1990s, when the trade unions were
already weakened.
25 The fall in the lira's nominal exchange rate between 1971 and 1976 helped to
avoid deterioration of the competitive position of Italian industry. The real
exchange rate thus decreased slightly despite an increase in unit labour costs
in terms of national currency that was greater than in the other industrial-
ized countries (see Aquino 1986; European Commission 2002). The depreci-
ation of the currency and the worsening of labour-market conditions did not

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
114 Unemployment, Productivity and Institutions: Italy

stop the growth of real wages, however, owing to the ever-increasing


strength and militancy of the trade unions and to price indexation.
26 It should be noted that no increase in employment or social expenditure
could have been expected by the trade unions. The Maastricht parameters
imposed a primary budget surplus and overall budget corrections amounted
to 430 trillion lira between 1992 and 1998. The trade unions' strategy might

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
thus have been guided by the idea that it was the 'lesser evil' and that lower
interest rates might then help to change the restrictive character of fiscal
policies.
27 See Stirati (2001), for a more general perspective.
28 It is very common in the literature to describe 'shifts' that remain largely
unexplained in some functional relations (for example the traditional
Phillips curve, or the pseudo demand and supply curves). Our position may
be regarded as assuming that 'shifts' in empirical relations are only to be
expected because wages are affected by changes in the institutional frame-
work, but the causes of the 'shifts' should be explained as much as possible
and not left outside the realm of economic analysis.
29 See Lucifora and Origo (1999), who also provide a survey of previous results.

References
Alt, J.E. (1985) 'Political Parties, World Demand and Unemployment: Domestic
and International Sources of Economic Activity', American Political Science
Review, 79: 1016-40.
Aquino, A. (1986) Dinamica dei prezzi relativi nell'economia internazionale:
Lineamenti teorici ed esperienze deipaesi industriali (Bologna: II Mulino).
Banca d'ltalia (1978-2001a) Relazione Annuale (Rome: Banca d'ltalia).
Banca d'ltalia (1989-2002b) I Bilanci delle Famiglie Italiane, Supplemento al
Bollettino Statistico (Rome: Banca d'ltalia).
Barca, F. and Magnani, M. (1989) L'Industria fra Capitale e Lavoro (Bologna: II
Mulino).
Blanchard, OJ. (1997) 'The Medium Run', Brookings Papers on Economic Activity, 2:
89-158.
Blanchflower, D.G. and Oswald, J. (1994) The Wage Curve (Cambridge and
London: MIT Press).
Boyer, R. (1979) 'Wage Formation in Historical Perspective: The French
Experience', Cambridge Journal of Economics, 3(2): 99-118.
Brandolini, A., Cipollone, P. and Rosolia, A. (2003) 'Immigrazione e mercato del
lavoro in Italia', paper presented at the conference 'L'incidenza economica
dell'immigrazione' Florence, December.
Brunello, G. (1996) 'L'interazione tra salari pubblici e privati nell'economia
Italiana del dopoguerra', Politica Economica, 12(1): 85-114.
CGIL (1981), La CGIL dal 9th al 10th Congresso. Atti e Documenti CGIL e documenti
unitari, in L. Frisino (ed.) (Rome: Editrice Sindacale Italiana).
de Vivo, G. and Pivetti, M. (1980) 'International Integration and the Balance of
Payments Constraint: The Case of Italy', Cambridge Journal ofEconomics, 4:1-22.
European Commission (2002) European Economy, 6 (Brussels: European
Commission).
Eurostat (2000) Annuaire. Le guide statistique de VEurope (Luxembourg: Eurostat).

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Enrico Sergio Levrero and Antonella Stirati 115

Garegnani, P. (2003) 'Saving, Investment and the Quantity of Capital in General


Intertemporal Equilibrium', in F. Hahn and F. Petri (eds), General Equilibrium.
Problems and Prospect (London: Routledge).
Giarda, P.D. (1986) 'L'evoluzione della spesa pubblica in Italia: Alcuni fatti e
qualche proposta', in Ente per gli Studi Monetari, Bancari e Finanziari Luigi
Einaudi, Oltre la crisi. Le prospettive dell'economia italiana e il contributo del sistema

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
finanziario (Bologna: II Mulino).
Graziani, A. and Meloni, F. (1980) 'Inflazione e fluttuazione della lira', in
G. Nardozzi (ed.), I difpcili Anni '70, (Milan: Etas Libri).
Heimler, A. and Milana, C. (1986) Prezzi Relativi Ristrutturazione e Produttivita: le
Trasformazioni dell'Industria Italiana (Bologna: II Mulino).
Hibbs, D.A. Jr (1977) 'Political Parties and Macroeconomic Policy', American
Political Science Review, 71(4): 1467-87.
IMF (2002) International Financial Statistics Year Book (Washington: IMF).
Instituto Nazionale di Statistica (ISTAT) (1989) Sommario di statistiche storiche,
1926-1985 (Rome: ISTAT).
ISTAT (various years)(a) Rilevazione delle Forze di Lavoro (Rome: ISTAT).
ISTAT (various years)(b) Lavoro e Retribuzioni (Rome: ISTAT).
ISTAT (2002a) Rapporto annuale sulla situazione del paese (Rome: ISTAT).
ISTAT (2002b) Contabilitd Nazionale. Conti Economici Nazionali Anni 1970-2000
(Rome: ISTAT).
Kaldor, N. (1984) // flagello del monetarismo (Turin: Loescher editore).
Korpi, W. (1991) 'Political and Economic Explanations for Unemployment: A Cross-
national and Long-term Analysis', British Journal ofPolitical Science, 21 (3): 315-48.
Lama, L. (1987) Intervista sul mio partito (Rome, Bari: Laterza).
Levrero, E.S. (1999) 'Worker Bargaining Power and Real Wages from 1870 to 1913:
Phelps Brown Reconsidered', Review of Political Economy, 11(2): 183-203 (with
Errata, Review of Political Economy, 11(4): 484).
Levrero, E.S. (2006) 'Some Notes on Wages and Competition in the Labour
Market', in R. Ciccone, C. Gehrke and G. Mongiovi (eds), Sraffa and Modern
Economics (London: Routledge), forthcoming.
Levrero, E.S. and Stirati, A. (2004) 'Real Wages in Italy 1970-2000: Elements for
an Interpretation', Economia & Lavoro, xxxvm (1): 65-89.
Lucifora, C. and Origo, F. (1999) 'Alia ricerca della flessibilita: un'analisi della
curva dei salari in Italia', Rivista italiana degli economisti, iv (1): 3-35.
Organisation for Economic Co-operation and Development (OECD) (1977)
Towards Full Employment and Price Stability (Paris: OECD).
OECD (1981, 1983) Economic Outlook (Paris: OECD).
OECD (1985) Social Expenditures: 1960-1990. Problems of Growth and Control
(Paris: OECD).
OECD (2002) Revenue Statistics 1965-2001 (Paris: OECD).
OECD (various years) The Tax-Benefit Position of Production Workers (Paris: OECD).
Olini, G. (2003) Retribuzioni e produttivita in Italia dal 1993 al 2001, Studi e ricerche
CISL, Rome, October.
Petri, F. (2003) 'Should the Endogenous Theory of Growth be Based on Say's Law
and the Full Employment of Resources?', in N. Salvadori (ed.), The Theory of
Economic Growth-A Classical Perspective (Cheltenham: Edward Elgar).
Petri, F. (2004) General Equilibrium, Capital and Macroeconomics (Cheltenham:
Edward Elgar).

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
116 Unemployment, Productivity and Institutions: Italy

Phelps-Brown, E.H. (1968) A Century of Pay (London: Macmillan - now Palgrave


Macmillan).
Pivetti, M. (1999) 'High Public Debt and Inflation: On the Disciplinary View
of European Monetary Union', in G. Mongiovi and F. Petri (eds), Value,
Distribution and Capital. Essays in Honour of Pierangelo Garegnani (London:
Routledge).

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Romiti, C. and Pansa, G. (1988) Questi anni alia Fiat (Milan: Rizzoli).
Rothschild, K.W. (1993) 'The Phillips Curve and All That', in K.W Rothschild (ed.),
Employment, Wages and Income Distribution (London: Routledge), 125-61.
Salvati, M. (2000) Occasioni Mancate. Economia e Politica in Italia dagli Anni 60 a
Oggi (Rome-Bari: Laterza).
Stirati, A. (1992) 'Unemployment, Institutions and the Living Standard in the
Classical Theory of Wages', Contributions to Political Economy, 11: 41-66.
Stirati, A. (2001), 'Inflation, Unemployment and Hysteresis. An Alternative View',
Review of Political Economy, 13(4): 427-51.
Tarling, R. and Wilkinson, F. (1977) 'The Social Contract', Cambridge Journal of
Economics, 1(4): 395-414.
Trentin, B. (1980), II Sindacato dei Consigli, interview with Bruno Ugolini (Rome:
Editori Riuniti).
US Department of Labor (2000) Comparative Real Gross Domestic Product per Capita
and per Employed Person. Fourteen Countries 1960-1998 (Washington, DC: US
Department of Labor).
US Department of Labor (2002) International Comparisons of Hourly Compensation
Costs for Production Workers in Manufacturing, 1975-2001 (Washington, DC: US
Department of Labor).

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
6
Unequal Fortunes, Unstable
Households: Has Rising Inequality

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Contributed to Economic Troubles
for Households in the USA?
Heather Boushey and Christian E. Weller

Introduction
For the better part of the past three decades, the US economy has been
characterized by a growing income disparity. Rising inequality has taken
two forms. The share of national income that has gone to employee
compensation has tended to decline, and the distribution of this shrinking
share of the pie has become more unequal.
Rising inequality may also have impacted the economy in the
aggregate. While the link between rising inequality and productivity
growth is hard to substantiate, the connection between greater inequal-
ity and aggregate demand finds more support. However, despite rising
inequality demand growth has remained relatively strong. One resolu-
tion to this ambiguity may have been a rise in more consumer borrow-
ing, especially for low-income households. In turn, this may have
increased economic distress for households, particularly if economic
mobility was limited.
The rest of the chapter proceeds as follows. In the next section we
present background data on inequality as well as on household indebt-
edness and household economic distress. This is followed by a discus-
sion of the causes of the rising inequality in the USA, followed by an
overview over the macro economic consequences of rising inequality.
We then discuss the evidence on the link between inequality and
innovation and aggregate demand, and present our evidence on the link
between inequality, consumer indebtedness and economic distress. The
final section contains concluding remarks.

117

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
118 Rising Inequality in the USA

Inequality trends

Inequality between capital and labour and within labour has been on
the rise in the USA for the past decades. A striking characteristic of the
last economic recovery in the USA was a rapidly growing gap between

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
supply and demand; economic growth and employment diverged
because historically weak demand growth did not keep pace with
productivity growth,1 resulting in rising profits and laggard employee
compensation (Bivens and Weller 2005; Weller 2004a). In response,
households borrowed more to maintain their consumption (Weller
2004a) and businesses, not expecting strong sales growth, did not
ratchet up investment quickly (Weller et al. 2004; Weller 2004b).
Instead, much of the additional resources went for uses other than pro-
ductive investments, such as share repurchases and dividends (Bivens
and Weller 2005).
The 'job-loss' recovery, though, was the culmination of long-run
trends. Since the mid-1970s, productivity growth has outpaced com-
pensation growth. Over the period from 1947 to the middle of 1975,
productivity and real hour compensation grew apart by a total of 6.0 per
cent. From the middle of 1975 to early 2004, productivity grew 25.7 per
cent faster than real compensation. Consequently, profits rose, too. The
profit share grew more rapidly after 1975 than before (Table 6.1) (Bivens
and Weller 2004, Wolff 2003).2 The trends of profit shares are reflected
in the trends of labour shares. Although the averages do not differ or
are even greater in the latter period, the trends tell a different story.
Regardless of how the labour share is measured, it declined in the latter
period, whereas it grew in the earlier period. The declines were especially
pronounced after 2001 (Table 6.1).
Since the 1970s, inequality within the distribution of labour income
has also increased, regardless of how it is measured, relatively or
absolutely. We focus first on wage inequality because wages and salaries
make up the bulk of family income. However, research has found
that shifts in the distribution of earnings are not necessarily the most
important factor in explaining changes in the income distribution. Over
the past 25 years, changes in the distribution of other kinds of income
(interest, dividends and rent) have played just as an important role. The
severity of the inequality problem is heightened by decreases in eco-
nomic mobility at the same time.
Overall, the past few decades have seen a dramatic shift of income
away from labour and away from low-wage earners and low-income
families. The sharpest increases occurred during the 1980s: from the

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect -
Table 6.1 Levels and changes of profit shares and labour shares of national income

After tax profit After tax profit Before tax profit Before tax profit Compensation Compensation
share with net share without share with net share without share and proprietors'
interest net interest interest net interest income share

Average - earlier period


(1947-75) 9.3(1.4) 6.7(1.0) 14.2(1.1) 11.6(1.4) 62.9(1.82) 73.2(1.14)
Average monthly rate
of change (percentage
points) - earlier period 0.05 0.01 0.03 -0.01 0.05 0.00

2014-03-15
Average - later period
(1975-2004) 13.6 (1.2) 6.4 (1.0) 16.7 (1.0) 9.5 (1.2) 65.6 (0.88) 73.3 (0.88)
Average monthly rate
of change (percentage
points) - later period 0.04 0.04 0.03 0.03 -0.02 -0.01
Average - 2002-04 14.4 (0.7) 8.3 (0.9) 16.7 (0.8) 10.6(1.0) 65.2 (0.71) 73.7 (0.58)
Average monthly rate
of change (percentage
points) - 2002-04 0.17 0.23 0.23 0.29 -0.18 -0.17

Notes: All figures are in per cent. Figures in parentheses are standard errors. The entire sample for profit shares spans from 1947 to the first quarter of
2004 and is split at the second quarter of 1975.
Source: Bivens and Weller (2005).

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim Truger
120 Rising Inequality in the USA

economic peak in 1979 to the next peak in 1989, workers in the bottom
10th percentile saw their wages fall by 14.1 per cent while those in the
95th percentile saw their wages rise by 8.1 per cent. Inequality also
increased between workers at the median and those at the top, as wages
for the median worker grew by 0.0 per cent over that same period

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
(Mishel et al. 2003). The overall trend was for the top and bottom to pull
away from each other, while the middle saw little change.
Over the 1990s, the growth in wage inequality slowed. Between 1989
and 2000, wages increased by 13.1 per cent for those in the bottom
10th percentile and increased by 16.6 per cent for those in the 95th per-
centile. Inequality between the top and the middle increased more so
than between those at the top and bottom, as the median worker saw
their wages rise by only 5.9 per cent. Much of the slowing of inequality
occurred during the latter half of the 1990s, as unemployment dipped
below 5 per cent. Between 1995 and 2000, wages for those at the bottom
of the wage distribution actually saw their wages rise faster than those at
the top for the first time in decades (Mishel et al. 2003).
If overall inequality rose primarily because of across-group inequality,
then it would be due to factors specific to particular groups, such as
discrimination or increasing returns to skills. But if inequality has been
primarily driven by within-group differences, then the problems are
more generalized and cannot be evaluated with microeconomic, human
capital based models. The general consensus is that there has been an
increase in both across- and within-group inequality. While across-
group inequality has increased in terms of educational attainment, it
has not increased substantially across African Americans and whites,
and gender inequality has actually attenuated.
Across-group wage inequality declined among male and female work-
ers over the past three decades, mostly because male wages fell during
the 1970s and 1980s. Over the period from 1975 to 1996, average male
wages stagnated, while average female wages rose by about one-fifth.
Even once wages are regression-adjusted to control for education and
experience, the gender gap closed from 47 per cent in 1975 to 27 per
cent in 1993. However, since 1993, the gender pay gap has remained
unchanged, hovering at around 25 per cent (Gottschalk and Danziger
2003).
The inequality gap between African-American and white workers
stayed relatively constant over the past three decades. Controlling for
personal characteristics, among women, the black/white gap was virtu-
ally non-existent in the late 1970s, however it increased to 4 per cent by
2001. Among men, the black/white gap was 14 per cent in 1975, rising

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Heather Boushey and Christian E. Weller 121

to about 20 per cent in the early 1980s. It came back down to 15 per cent
by 2001.
Inequality has increased, however, across educational attainment
levels. Economists think that human capital, that is the education
and training experiences of the worker, is correlated with a worker's pro-

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
ductivity which, in turn, determines their wage. Thus, there is an expec-
tation of inequality across educational attainment levels. However,
there is not an expectation of an increase in this inequality over time,
which is what has happened. In 2003, college-educated men earned
41.5 per cent more than high-school-educated men, compared to a col-
lege premium of only 25.3 per cent in 1973. For women, the increase
was less dramatic: in 2003, college-educated women earned 46.1 per
cent more than high-school-educated women, compared to a 37.7 per
cent premium in 1973 (Mishel et al. 2005). Most of the increase in edu-
cational inequality occurred between the late 1970s and the early 1990s.
Since 1992, the gap between high-school and college-educated workers
has been relatively flat among both men and women, although there is
slight upward trend, more so for men than women (Gottschalk and
Danziger 2003).
Another way of evaluating wage inequality is to examine the within-
group differences. Changes in within-group inequality indicate that
something has fundamentally changed about the economy that affects
all kinds of workers, rather than only affecting some workers based on
their educational attainment level or other identifying characteristics.
Over the past few decades, this kind of inequality has increased along-
side across-group inequality. As with across-group inequality, the largest
increases in within-group inequality occurred during the 1970s and
1980s. Among both men and women, the residual wage inequality
in hourly wages at the 10th percentile fell precipitously between 1975
and the late 1990s. A person at the 10th percentile of the distribution -
where 10 per cent of households have income below that level and
90 per cent have income above that level-in the late 1980s had
earnings roughly 25 per cent lower than a similar person in 1975. For
individuals at the 90th percentile, wages were about 10 per cent higher
in the mid-1980s, compared to 1975, but stopped increasing during the
1990s (Gottschalk and Danziger 2003).
Since most American families derive the majority of their income
from wages, trends in income inequality closely mirror wage inequality.
Over the period from 1947 to 1979, annual growth in inflation-adjusted
family income was similar across quintiles. However, during the period
from 1979 to 2003, those in the top quintile, and in particular the top

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
122 Rising Inequality in the USA

5 per cent, saw their family income grow significantly faster than those
in the lower income quintiles.
As with wage inequality, the largest increases in income inequality
occurred during the 1980s. Between 1979 and 1989, rising inequality
was the result of large increases in income among high-income families

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
and declining income among lower-income families. During the 1990s,
income among lower-income families started to rise, but higher-income
families saw their incomes rise even faster. Between 1989 and 2000,
among married-couple families with children, income of those in the
bottom fifth of the income distribution rose by 8.8 per cent, compared
to 14.1 per cent among those in the top fifth (Mishel et al. 2005).
Analysis of percentiles does not allow us to see just how much those
at the very top of the income distribution have gained over the past
three decades. To see this, we need to look at data on tax returns that
does not top-code the incomes of those at the very top. Top-coded data
has all values over a certain threshold recoded as a common number in
order to protect the privacy of the very wealthy. Recent research using
tax returns has found that between 1973 and 2000, the average real
income of the bottom 90 per cent of American taxpayers fell by 7 per
cent while the income of the top 1 per cent grew by 148 per cent (Piketty
and Saez 2001).
While the top has pulled away from the bottom, the top has also been
pulling away from the middle. Households at the 80th percentile now
have over twice the income (2.01 times) of households at the 50th per-
centile. Back in 1967, this ratio was only 1.66. The gap between house-
holds at the bottom and middle has remained relatively constant, at
0.42 in both 1967 and 2003. Similarly, the ratio of the income limit for
the top 5 per cent of income earners to the bottom 20 per cent rose from
6.3 in 1967 to 8.6 in 2003. Also, the annual increases almost doubled
after 1980. From 1967 to 1980, the ratio of income cut-offs increased
annually by 0.6 per cent, whereas it rose by 1.0 per cent on average each
year from 1980 to 2003. In comparison, the ratio of the income limits
for the bottom 20 per cent to the middle percentile stayed fairly
constant at about 0.42 (Census 2004a).
As a result of unequal growth in incomes, inequality among American
households is currently at its highest level since the US Census Bureau
began conducting its annual survey of household income. In the aggre-
gate, this meant that more income became concentrated at the top of
the income scale and less income at the bottom. The share of total
income accruing to the 20 per cent of households at the top of the
income scale rose from 43.8 per cent in 1967 to 49.8 per cent in 2003.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Heather Boushey and Christian E. Weller 123

In comparison, the share of income going to the 20 per cent of house-


holds at the bottom of the income scale fell from 4.0 per cent to 3.4 per
cent over the same period. Interestingly, the aggregate income share of
the bottom 20 per cent of households actually rose until 1980, when it
reached 4.3 per cent, whereas the aggregate share of income accruing to

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
the top 20 per cent fell slightly until 1980, when it reached 43.7 per cent
(Census 2004a).
This sharp rise in income inequality over the past three decades
occurred even as families put more family members into the workforce.
Back in 1973, most mothers did not work outside the home, while by
2000 most mothers did hold a paying job. The increase in hours worked
by wives was greater among families in the bottom fifth of the income
distribution. Between 1979 and 2000, wives in families with children
increased their hours by 43.9 per cent, compared to a 27.4 per cent
increase in hours among wives in families in the top fifth. Within
married-couple families with children, inequality would have increased
even more without the contribution of wives' income. Between 1979
and 2000, among families in the bottom fifth, income would have fallen
by 13.9 per cent without the earnings of wives, rather than rising by
7.5 percent; among families in the top fifth, income would have
increased by 51.5 per cent, rather than 63.0 per cent (Mishel etal. 2005).
If wage and income inequality were counterbalanced by the potential
for economic mobility, then greater inequality would not require that
some stay at the bottom (or at the top). This would be especially true
if inequality was the result of immigration as new immigrants enter at
the bottom and then move up. However, this is not the case. In the
1970s, 50.7 per cent of families who began the decade in the bottom
quintile and 49.1 per cent of families who began the decade in the
second bottom quintile moved into a higher quintile over the decade.
However, in the 1990s, only 46.8 per cent of families who began the
decade in the bottom quintile and 37.9 per cent of families who began
the decade in the second bottom quintile moved into a higher quintile
(Bradbury and Katz 2002).
Wysong et al. (2003) found that class matters now more than ever.
Sons from the bottom three-quarters of the socio-economic scale were
less likely to move up in the 1990s than in the 1960s. By 1998, only
10 per cent of sons of fathers in the bottom quarter (defined by income,
education, and occupation) had moved into the top quarter, whereas by
comparison, by 1973, 23 per cent of lower-class sons had moved up to
the top. Thus, there is a smaller chance that a low-income family will
move up the income ladder over time.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
124 Rising Inequality in the USA

The labour force participation of wives has been critical to the story of
mobility. Recent research has found that families where wives had high
and rising employment rates, work hours, and pay were more likely to
move up the income ladder or maintain their position, rather than fall
down the ladder (Bradbury and Katz 2004). But, even the large increase

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
in labour supply of women (and mothers, in particular) has not been
sufficient to counterbalance declining mobility overall.
Even though wages account for most of family income, other factors
are important, too. Capital income as a share of personal income dou-
bled from 7.1 per cent in early 1947 to 14.1 per cent in the second quar-
ter of 2004 (BEA 2004). However, the assets underlying these income
streams are fairly unequally distributed. For instance, Wolff (2002a)
reported that the bottom 40 per cent of households had negative finan-
cial net worth in 1998, and that they had little total net worth. Further,
Wolff (2004) reported that wealth inequality rose from 1998 to 2001 for
total net worth.

Macroeconomic background

With income growth lagging, households increased their consumption


by saving less and borrowing more.3 The personal savings rate declined
from an average high of 9.7 per cent in the early 1970s to 4.7 per cent in
the 1990s and to 2.1 per cent in the latest business cycle (Table 6.2).

Table 6.2 Savings and consumer debt, business cycle averages

Business cycle Personal Total consumer Mortgages Consumer


dates savings debt as share as share of credit as share
rate of disposable disposable of disposable
income income income

1949:IV-1953:II 7A 38.2 23.1 11.6


1953:111-1957:111 7.9 47.3 29.1 13.9
1957:IV-1960:II 7.9 55.0 35.0 15.0
1960:III-1969:IV 8.4 65.4 41.0 17.7
1970:I-1973:IV 9.7 63.7 38.2 18.1
1974:1-1980:1 9.5 65.2 40.1 17.6
1980:11-1990:111 8.9 73.4 47.0 17.9
1990:IV-2001:I 4.7 91.7 61.3 20.0
2001:11-2004:1 2.1 109.0 74.3 23.9

Notes: All figures are in per cent. Mortgages comprise both traditional mortgages and home
equity loans. Consumer credit refers to revolving consumer credit, such as credit-card debt,
and non-revolving credit-card debt, such as car loans.
Source: Bivens and Weller (2005).

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Heather Boushey and Christian E. Weller 125

Simultaneously, consumer debt, especially mortgage debt, rose from


63.7 per cent of disposable income in the early 1970s to over 100 per cent
in the most recent business cycle.
At the same time, the economic distress of households increased.
From 2001 through mid-2004, households dedicated at least 13 per cent

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
of their disposable income to service their debts - the largest share in
twenty years (BOG 2004b). Charge-off rates have also been high with,
for instance, credit card charge-off rates above 5 per cent of all loans
since 2001 (BOG 2004a). In addition, personal bankruptcies have risen
so that the share of households that declared bankruptcy reached an
estimated record 1.5 per cent in 2003.
While households borrowed more to make ends meet, firms used
their additional profits for uses other than productive investments.
Specifically, firms increased their share repurchases and dividend pay-outs.
While in the 1970s, corporations used 10.7 per cent of their resources for
these purposes, this share grew to more than 30 per cent since the 1980s
as capital expenditures simultaneously declined (Table 6.3).

Table 6.3 Selected use of non-financial corporate resources, business cycle


averages
Productive
Financial uses as share of uses as share
total internal resources of total internal
resources
Business cycle Total Dividend Net equity capital
dates payouts issues expenditures

1953:111-1957:111 17.9 23.3 -5.4 79.1


1957:IV-1960:II 17.9 22.3 -4.5 74.6
1960:III-1969:IV 20.1 21.8 -1.7 83.1
1970:I-1973:IV 10.7 20.0 -9.3 101.2
1974:1-1980:1 14.8 17.2 -2.5 106.4
1980:11-1990:111 31.4 18.5 12.9 95.1
1990:IV-2001:I 32.0 25.2 6.8 88.3
2001:11-2004:1 30.8 25.4 5.4 76.7

Notes: All figures are in per cent. Total internal resources are defined as after-tax profits plus
inventory valuation adjustments and capital consumption allowance. Net equity issues, orig-
inally a source of funds, are multiplied by minus one to make them comparable with other
uses of funds. Figures do not add to 100% since other sources, especially borrowing of funds,
are excluded.
Source: Bivens and Weller (2005).

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect -
Table 6.4 Selected macroeconomic measures, business cycle averages, 1948-2003
Growth
Ratio of contribution
Share of income consumption Consumption
income Share of limits of Real & residential Consumption as share of
Business of bottom income of 95th to 50th Productivity Real GDP consumption fixed as share of disposable
cycle dates 20% top 20% percentile growth growth growth investment GDP income

2014-03-15
1948-52 n.a. n.a. n.a. 3.55 4.75 4.77 -37.49 63.73 91.42
1953-57 n.a. n.a. n.a. 1.96 2.90 2.92 6.91 62.25 90.34
1958-59 n.a. n.a. n.a. 2.92 2.93 2.96 6.02 63.05 90.12
1960-69 4.10 43.20 2.61 2.67 4.33 4.33 64.41 61.83 89.49
1970-79 4.13 43.58 2.69 1.15 2.93 2.92 205.69 62.15 87.87
1980-90 4.35 43.48 2.82 1.44 2.90 2.89 25.65 62.63 88.22
1991-2000 3.66 48.67 3.36 2.00 3.23 3.22 83.01 67.26 91.82
2001-03 3.47 49.87 3.56 3.68 1.88 1.86 144.96 70.17 94.52

Notes: All figures are in per cent. Starting date for inequality data is 1967.
Sources: BEA (2004), Census (2004a) and BLS (2004a).

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim Truger
Heather Boushey and Christian E. Weller 127

Despite laggard investment, productivity accelerated. Compared to


other early recovery periods, this was the first time during which
productivity growth actually outpaced aggregate demand growth in the
first two years of a recovery. Measured as business cycle averages, pro-
ductivity growth in 2002 and 2003 was higher than in previous business

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
cycles, although the economy sustained productivity growth that was
almost as high for a period of four years in the late 1940s (Table 6.4).
Reflecting the weakness in demand and the divergence between sup-
ply and demand, consumption growth slowed down since 2001
(Table 6.4). For 2001 until 2003, average real consumption growth was
1.9 per cent, compared to 3.2 per cent in the 1990s-the slowest con-
sumption increase of any postwar business cycle. However, consump-
tion never slowed during the most recent recession and consumer
spending on new homes and home renovations accelerated more than
it had in prior recoveries. Thus, consumer spending contributed more to
growth than the actual growth rate, reflecting the continued trade
deficits and the slowdown in investment (Table 6.4) (Weller etal. 2004).

Explaining the rise in inequality

A number of factors contributed to the rise in inequality between capi-


tal and labour. The two most convincing explanations are increasing
trade intensity and changes in the institutional make-up of US corpora-
tions that gave more power to institutional investors and managers and
reduced the influence of labour unions (Bivens and Weller 2005;
Lazonick and O'Sullivan 2000).
Increased trade, especially in manufactured goods, is a non-negligible
factor in explaining the rise in inequality. About a third of the loss of
manufacturing jobs that occurred from the end of 2000 to the end of
2003 - the largest driving force of the 'job-loss' recovery - can be attrib-
uted to rising trade deficits in manufactured goods (Bivens 2004;
Atkinson 2003). This result is consistent with standard trade theory. The
predicted structure of trade for a nation like the USA - labour-intensive
imports and capital-intensive exports - implies that trade should lead to
a decline in the demand for labour and rise in the demand for capital,
moving wages (down) and profit rates (up) accordingly.
During the 1970s and 1980s, the main distributional impact of trade
seemed to be on the distribution of wages; as trade raised the return to
skilled labour and lowered the return to unskilled labour (Cline 1997).
The relative lack of a capital/labour income dimension of trade was con-
sistent with what economists dubbed 'Leontiefs Paradox': the finding

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
128 Rising Inequality in the USA

that US exports were not notably capital-intensive, nor were US imports


labour-intensive. What really seemed to distinguish US trade in the
1970s and 1980s was that it was particularly biased against blue-collar
workers (Borjas and Ramey 1995). That is, for most of the period that we
are interested in, trade can explain an increase in the within labour

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
inequality, but not a rise in the inequality between labour and capital
income.
Another factor that explains the divergent trends between capital and
labour is a rising imbalance in the corporate governance realm.
Specifically, a growing concentration among institutional shareholders
and rising power of managers in deciding corporate resource allocations
was juxtaposed by a declining unionization rate.
A greater reliance on managed assets in household savings led to a
rising concentration of shares in the hands of mutual funds, brokers,
public and private pension funds and insurance companies. In 1952,
institutional investors owned less than 10 per cent of outstanding equi-
ties, and in 2004 it surpassed 50 per cent for the first time (BOG 2004c).
Also, institutional investors had growing incentives to use this oppor-
tunity to allocate resources towards capital. Starting in the 1970s, insti-
tutional changes, such as the Employee Retirement Income Security Act
(ERISA) of 1974 and the introduction of 401 (k) retirement saving plans,
gave fund managers and households a common interest in maximizing
asset returns (Bivens and Weller 2005). This altered the way corporations
were run, creating a new class of professional managers that enjoyed
greater freedom in allocating corporate resources towards a strategy of
rent extraction, including downsizing, outsourcing and restructuring, as
well as a reorientation towards financial service activities, away from
actual production (Lazonick and O'Sullivan 2000; O'Sullivan 2000).
Consequently, labour compensation declined, along with union repre-
sentation, as jobs especially in manufacturing were lost. Thus, the allo-
cation of corporate resources towards faster profit growth proceeded
with less opposition than in the past. Profits in turn were increasingly
used for dividend pay-outs and share repurchases, which directly bene-
fited executives, whose compensation was dependent on the perform-
ance of a company's share value.
Corporate governance changes explain the rise in income inequality
in important ways. For one, they explain a growing emphasis of profit
generation, thus shrinking the allocation of corporate resources towards
labour. They also explain the growing inequality within labour as a
result of more rent-extracting activities and executive compensation
contingent on share price performance. Finally, the latter aspect also

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Heather Boushey and Christian E. Weller 129

explains the growing share of personal income that is derived from


assets, which in turn contributes to the rise in income inequality due to
the unequal distribution of household wealth.
Because inequality has risen along so many dimensions, the reasons
for its expansion are not easy to pinpoint on a single source or in a sin-

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
gle regression. The concentration of financial market power, combined
with declining bargaining power - the decline in unionization, deregu-
lation and privatization, assaults on social programmes, a lack of
increase in the minimum wage, and an inequitable distribution of
healthcare - all contribute the power imbalance of workers with respect
to employers and the state.
The level of unemployment is critical in explaining changes in
inequality - higher unemployment leads to inequality growth and
inequality tends to shrink during periods of tight labour markets. This is
because unemployment affects the fortunes of those at the bottom of
the labour market more than those at the top. Individuals with limited
education or who earn in the bottom of the wage distribution are more
likely to lose a job when unemployment rises. The mechanisms through
which unemployment affects wages are not always direct, but, at their
core, they are related to the relative power of labour and capital within
the US economy. If workers are fearful of losing their job, because of
high levels of unemployment, they will be less likely to bargain hard for
higher wages and less able to search for a new position at a higher wage
(or benefit) level.
Structural changes in the US economy are also implicated in inequal-
ity's growth. The decline in unionization, the lessening of labour market
regulation, changes in the industrial and occupation mix of jobs, and
globalization have all contributed to growing inequality. A declining
real minimum wage and de-unionization can explain about one-third of
the growth in wage inequality, while globalization - immigration, trade
and capital mobility - can explain another one-third of inequality's rise
(DiNardo etal. 1996; Gottschalk 1997; Lee 1999; Card etal. 2003).
Growing inequality is also the result of a lack of a broad-based social
insurance system. Once an American becomes poor, it is exceedingly
difficult to rise back up into the middle class. The OECD has found
that in the USA there are not only more poor families, but they are less
likely than the poor in other countries to 'exit' from poverty. For exam-
ple, while 41.1 per cent of poor Germans exit poverty each year, only
29.5 per cent of poor Americans do so. Because income transfers are so
small, the only way out of poverty is earnings or marriage; this has left
US poor people more likely to exit poverty through earnings than in

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
130 Rising Inequality in the USA

other OECD nations. Yet, because of limited growth in wages among


low-wage workers over the past few decades (up until the late 1990s),
this led to lower poverty exit rates in the USA. Thus, the US social wel-
fare state does not help to reduce inequality through helping families
back into the middle class; further, the USA has more wage and income

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
inequality than any other OECD country (Mishel et al. 2003).
In general, the US labour market has become an increasingly insecure
place for workers. As they emerged from the stagflation of the 1970s,
US firms made the case to government and to workers that they needed
more control over workers and that could not afford to compete inter-
nationally while paying decent wages and benefits. Much of the blame
for stagflation was placed on labour's 'unreasonable' demands. The
plant closings and outsourcing that happened over the next decade -
the decimation of manufacturing and the unions - scared policy-makers
as well as labour unions. For example, one trend in the USA has been for
cities and states to compete with one another to reduce taxes to lure and
keep employers in their locality, often without requiring anything of
the firms in exchange. At the same time, the responsibility for ensuring
a living wage has fallen more to the government than to low-wage
employers, as low wages are supplemented by the Earned Income Tax
Credit, even as the inflation-adjusted minimum wage declined
markedly. Further, no longer can a worker assume that they will stay in
a job for their career. Internal labour markets are increasingly a thing of
the past.

Inequality and the macro-economy

If the rise in the profit share translates into an increase in the profit rate,
it may result in more investment and potentially more growth since
greater profits provide firms with more incentives to invest, particularly
if tax policy supports the increase in the profit rate (Bivens and Weller
2004; Palomba 2002). However, a greater profit rate may also attract
more entrants into an industry and reduce the rate of profit below
what was originally expected (Peretto 1995). In addition, profit rate
gains may have been generated by reducing wage growth, which may
suppress demand growth and thus profit growth over time (Palley 1996).
Furthermore, through rent extraction, many core activities of a firm are
abandoned. This may impede organizational learning and thus innova-
tion and growth (Lazonick and O'Sullivan 2000; O'Sullivan 2000). These
concerns, though, do not necessarily contradict the original notion
that a higher profit rate will lead to more investments. They do suggest,

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Heather Boushey and Christian E. Weller 131

however, that over time countervailing forces will gain ground and
weaken the link. This may explain why there is little empirical evidence for
profit led growth (Stockhammer and Onaran 2004, and in this volume).
A positive link between inequality and innovation has been suspected,
particularly because inequality may imply a greater wage premium on

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
skills. The opposing view, though, contends that rising inequality
leads to growing political instabilities, and thus to disincentives for
accumulation, followed by reduced innovation (Alesina and Perotti
1996; Larrain and Vergara 1997; Rodriguez 2000). There may be a
threshold below which inequality fosters growth, but above which polit-
ical instability considerations outweigh skill development effects
(Benhabib 2003; Chen 2003).
Patterns of wage growth do not suggest a connection between
inequality and skill development (Mishel et al. 2001, 2005; Appelbaum
and Weller 2001). From the late 1970s to the mid-1990s, wages declined,
especially for young high-school graduates. By the late 1990s they had
not recovered to the levels of the late 1970s. Also, entry-level wages
among male college graduates were stagnant from 1973 to 1989, and fell
9.9 per cent from 1989 to 1995. However, between 1995 and 1999
among young college graduates real wages rose 14.9 per cent for men
and 9.4 per cent for women. Further, wage inequality rose fastest in the
1980s, when productivity growth was not faster than during the 1970s,
when wage inequality did not rise, and when productivity growth accel-
erated in the 1990s there was no matching inequality increase (Mishel
et al. 2001, 2005). Furthermore, most of what explains the rise in wage
inequality is a pulling away of the top 10 per cent of wage earners in the
1990s, while the differential between low and middle wage earners was
stagnant (ibid.). Finally, the occupations that account for the largest edu-
cation wage differentials were managers and sales workers, not technical
professions (Mishel etal. 2001).
Additional findings are largely inconclusive as to the effect of inequal-
ity on innovation. Some researchers have found a positive, albeit
small effect of inequality on growth (Scully 2002), while others found
a negative relationship between the two (Alesina and Perotti 1996;
Panizza 2002; Rodriguez 2000; Rupasingha etal. 2002). The findings of a
link between inequality and growth, though, appear to be sensitive to
the empirical model's specification (Crafts 1992; Panizza 2002). Also, the
contention that income inequality in the USA has resulted in greater
productivity growth due to better skill development does not seem to
enjoy empirical support (Appelbaum and Weller 2001; Osberg 2003;
Mishel eta/. 2001, 2005).

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
132 Rising Inequality in the USA

Even though there does not seem to exist a link between inequality
and aggregate supply, there may be a connection between inequality
and demand. If the labour share of national income declines, aggregate
demand could fall as consumption growth slows. This effect could be
exacerbated if inequality within labour rises, too. A greater concentra-

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
tion of total income among higher income earners may reduce aggre-
gate demand growth as higher income earners have a lower marginal
propensity to consume and thus are more prone to save than lower
income households (Keynes 1936).
On the face of it, the figures do not necessarily support the notion
that rising inequality can lead to declining consumption. For instance,
the share of consumption relative to disposable income, one of the
measures that should decline with inequality, actually increases as
inequality rises (Table 6.4). Also, research based on micro data appears to
be somewhat ambivalent. Specifically, savings incentives should be
relatively most effective for high-income earners and least effective for
low-income earners. An analysis using micro data to study the effective-
ness of savings incentives, such as 401 (k)s and individual retirement
account, tend to find that savings incentives seem to be more effective
in raising savings for low-income earners and not for high-income ones
(Engen and Gale 2000).
A number of studies, though, based on macro data suggest that there is
a negative connection between inequality and aggregate demand (Arestis
and Howell 1995; Brown 2004; Pressman 1997). It is possible that the gap
between income and consumption has been filled by consumer debt.

Inequality, consumer debt and financial distress

As inequality rose, consumption should have declined without com-


pensating increases in consumer debt (Brown 2004). Some researchers
have argued that in fact the rise in inequality has given way to an
endogenous development of credit markets, increasing the credit supply
in response to rising inequality (Kruger and Perri 2002).
The credit supply rose for a number of reasons and its expansion was
most notable among low-income households. First, the standardization
of mortgages and the introduction of mortgage-backed securities
took shape in the 1960s with the creation of Ginnie Mae (Government
National Mortgage Association, GNMA) under the Housing and Urban
Development Act of 1968, the creation of Freddie Mac (Federal Home
Loan Mortgage Corporation), the engagement of Fannie Mae (Federal
National Mortgage Association, FNMA)in the pass-through market under

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Heather Boushey and Christian E. Weller 133

the Emergency Home Finance Act of 1970, and tax advantages for
mortgages under the 1986 Tax Reform Act (Vandell 2000). These inno-
vations helped to reduce the risks for mortgage lenders and lowered the
costs of mortgages (Van Order 2001). Pearce and Miller (2001) estimated
that the costs savings to consumers amounted to somewhere between

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
$8.4 billion to $23.5 billion.
Second, financial innovation increased credit supply. The Tax Reform
Act of 1986 phased out the deductibility of most non-mortgage interest
and introduced new marginal tax rates that reduced the tax advantage
of all types of debt. This led to a shift of consumer debt towards mort-
gages and home equity lines (Dunsky and Follain 2000, Stango 1999).
Stango (1999) estimates that by 1991 aggregate mortgage debt was over
1 percent higher, credit-card debt approximately 14 per cent lower, and
auto loan debt approximately 9 per cent lower than they would have
been without these changes.
Third, increased financial competition raised the credit supply.
Specifically, competition among credit-card providers gave financial
institutions incentives to offer credit cards to clients that were previ-
ously underserved (Manning 2000). The relative increases of credit-card
debt appear to be larger among lower-income households than among
higher-income households (Manning 2000; Yoo 1996). In addition,
non-bank credit also expanded in response to rising inequality. These
types of loans, including payday loans, pawnbroking, rent-to-own and
appliance title loans, and tax-refund anticipation loans grew and were
concentrated among low-income customers (Barr 2001; CFA 1998, 1999;
Stegman and Faris 2003).
Greater credit supply should offset the adverse effects of rising
inequality on consumption. Higher debt-service costs, though, could
outweigh the added impulse to consumption from more debt. In partic-
ular, household sources have to equal their uses:

ACt + ATrt + AIt + AAt = AYt + ADt (6.1)

The change in consumption, C, is thus equal to the change in disposable


income, Y, plus the change in debt, D, minus the change in net transfers,
Tr, minus the change in assets, A, minus the change in interest payments,
I. Assuming that borrowing is the only thing that keeps consumption
going, consumption can increase as long as new debt is larger than the
increases in interest payments. Increases in interest payments are:

AIt = ArtDt.l (6.2)

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
134 Rising Inequality in the USA

Table 6.5 Sources and uses of household finances


Change in
disposable Change in Change in Change in
income financial interest transfer Change in
relative to assets to payments payments to debt to

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Business consumer consumer to consumer consumer consumer
cycle dates spending spending spending spending spending

1954:3-1957:3 0.74 -7.93 0.31 0.00 -1.28


1957:3-1960:2 -1.78 2.79 0.16 -0.13 1.68
1960:2-1969:4 3.91 -0.55 0.49 0.13 -2.14
1969:4-1973:4 2.76 4.11 -0.05 0.11 3.11
1973:4-1980:1 -1.74 -0.40 0.22 -0.11 1.56
1980:1-1990:3 -0.94 -0.18 0.47 0.41 -2.44
1990:3-2001:1 -5.91 -6.91 0.00 0.39 0.22
2001:1-2004:2 -2.18 2.97 -0.77 0.13 5.14

Notes: All figures are in per cent. Totals do not add to zero due to statistical discrepancies.
Disposable income refers to personal income minus taxes plus net investments in consumer
durables and consumption of fixed capital minus all new spending on consumer durables,
government insurance and pension reserves, and net capital transfers. Consumer spending
refers to personal consumption expenditures plus capital expenditures on real estate.
Sources: BOG (2004c) and BEA (2004).

such that the growth of household debt has to be greater than the
percentage point increase in the interest rate. The figures from past
business cycles show that debt has played a larger role in financing
consumption in this business cycle than in any previous ones, and that
debt changes were indeed greater than changes in interest payments
(Table 6.5).4 Specifically, personal income became a negative contributor
to changes in consumption, meaning that consumption growth was
larger than personal income growth. Also, since the 1980s, debt growth
has outpaced consumption growth helping to finance the gap left by
slower income growth. Yet, despite faster debt growth, interest pay-
ments were unchanged and even declined relative to consumption
spending since the 1980s.
For debt and consumption to rise simultaneously, income has to go
up, interest rates have to go down, or both. Debt and consumption
grew at the same time, while income growth has been lagging at least
behind consumption growth (Table 6.5). At the same time, though,
interest rates have been falling in nominal and real terms. Thus,
declining interest rates have allowed households to sustain their
consumption.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Heather Boushey and Christian E. Weller 135

Inequality and consumer debt


We can estimate the effect of inequality on the credit supply since under
credit rationing, realized credit is equal to the credit supply (Stiglitz
and Weiss 1981). Credit supply is measured as credit relative to dis-
posable income. Its explanatory variables are standard measures of

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
collateral - expected income gains-and interest that we supplement
with measures of inequality. Specifically, we use real disposable income
lagged once and the real mortgage rate. In addition, we use the labour
share of national income as a measure for inequality between capital
and labour. To account for within-labour inequality we use two meas-
ures, one for wage inequality and a proxy for wealth inequality and thus
capital income inequality. Our wage inequality measure is Atkinson's
inequality measure with a calibration factor of 0.5. As proxy of wealth
inequality, we use the ratio of the stockmarket index to the housing-
price index (Wolff 2002b):

+ ft(. 2 4 ^ § | p ) / 4 + PM-i + fat + *t (6.3)

where LI is labour income, NI is national income, Atkinson05 is


Atkinson's inequality measure calibrated with a parameter of 0.5, SP500
refers to the S&P 500 index and HPI to the housing price index, y is real
disposable income, r is the real mortgage rate, and s is a randomly dis-
tributed error term. Data for labour income, national income and dis-
posable income come from BEA (2004); data for the Atkinson's
inequality measure is from CEPR (2004); data for the S&P 500 is from
Yahoo! Finance (2004); data for the Housing Price Index is from OFHEO
(2004); and interest rate data are from BOG (2004d). Income inequality
is likely to affect debt over the course of some time, so that we use the
average of the four quarters ending in the current quarter for all of our
inequality measures.5 In each case the natural logarithm is used.
Table 6.6 presents our results for the determinants of total debt. The
first regression presents our baseline results with the expected positive
signs for the explanatory variables. In regression (2), we add our meas-
ure for inequality between capital and labour. We would expect this
measure to have a negative sign indicating that a greater distribution of
national economic resources towards labour is less likely to give rise
to endogenous credit expansions and thus less likely to lead to an

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
136 Rising Inequality in the USA

Table 6.6 ]Regression results for determinants of household debt, 1980-2003

(4) (5) (6)


(2) (3) Alternative Alternative Combined
Explanatory (1) Between Within between within labour
variables Baseline inequality inequality inequality inequality inequality

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
yt-i 0.47*** 0.56*** 0.62*** 0.64*** 0.61*** 0.59***
(0.10) (0.10) (0.10) (0.09) (0.11) (0.11)
rt -0.05** -0.04 -0.04 -0.04 -0.04 -0.04
(0.02) (0.03) (0.03) (0.03) (0.03) (0.03)
LI/NIt 0.72 0.81 0.78 0.78
(0.51) (0.52) (0.52) (0.52)
LPI/NIt 1.01**
(0.51)
Atkinson05t -0.14 -0.11
(0.14) (0.13)
AtkinsonlOt -0.13
(0.16)
SP500/HPIt 0.01 0.01 0.01
(0.03) (0.03) (0.03)
LIneqt -0.01
(0.02)
Constant —4 14*** -4.56*** -5.42*** -5.60*** -5.23*** -4.80***
(0.90) (0.85) (1.00) (0.89) (0.98) (0.91)
N 93 93 93 93 93 93
Adj. 0.23 0.25 0.45 0.65 0.44 0.33
JR-squared
Durbin- 1.86 1.86 1.88 1.86 1.87 1.88
Watson

Notes: In each case, a Prais-Winsten regression is used. LPI refers to labour income
plus proprietors' income and AtkinsonlO refers to the Atkinson inequality measure with
1.0 parameter instead of 0.5. LIneq refers to the combined labour inequality measure derived
by using the first factor. All inequality measures refer to the four-quarter average ending
in the current quarter. Figures in parentheses are standard deviations. ** denotes significance
at the 5% level, and *** denotes significance at the 1% level.

increase in the credit supply, ceteris paribus. This measure is statistically


insignificant. In regression (3), we add our two measures for income
inequality. Again, neither one is statistically significant. Regression
(4) introduces an alternative measure for inequality between capital and
labour. Now, we consider proprietors' income as part of total labour
compensation; this generates a statistically significant, yet positive cor-
relation between inequality and the amount of debt. One explanation
for this unexpected sign may be that more labour income may also con-
stitute more collateral for households to borrow against. In regressions
(5) and (6) we test the robustness of our results with respect to within

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Heather Boushey and Christian E. Weller 137

labour inequality. In regression (5), we replace the Atkinson measure


with the 0/£ calibration factor with the Atkinson measure with a 1.0 cal-
ibration factor. The results essentially remain the same. One problem
with our labour income inequality measures may arise from the fact that
both are highly correlated. To circumvent this problem, we combine

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
both measures using factor analysis. We first standardize both variable
and then calculate the principal factors. We use only the first factor to
generate a new variable 'labour inequality', which is a linear combina-
tion of the two separate variables and explains 89 per cent of the vari-
ance of both variables. Using this new variable instead of two separate
measures for labour inequality generates regression (6). The results are
again largely robust.
So far, we do not find a link between inequality and household debt.
One explanation may be that total household debt is too broad a cate-
gory to be affected by inequality. Borrowing by high and low-income
households in response to rising inequality may have had offsetting
effects. Specifically, higher-income earners are more likely than lower-
income households to own a home and thus be able to borrow against
their real estate (Wolff 2002b, 2004). Income inequality rose because
incomes of higher-income earners pulled away from the middle. Thus,
households that were more likely to own their residence also had more
collateral to borrow against, but less need to borrow additional money.
In comparison, lower-income households saw below-average income
gains as inequality rose, but also had fewer opportunities to borrow
against their own homes. As they had a greater need to borrow, but less
collateral, the literature suggests that a cycle of endogenous credit
expansion took place, which may have manifested itself in a dispropor-
tionate increase in credit-card debt among low-income households.
Thus, we estimate our results separately for mortgages (Table 6.7) and
credit-card debt (Table 6.8).6
Table 6.7 presents our results for mortgage debt. We find a consistent
positive and statistically significant relationship between the labour
share of national income and mortgage debt, suggesting that less labour
income allowed fewer households to buy a home and borrow against the
value of their real estate than otherwise would have been the case. A 1 per
cent decline in the four-quarter average of labour's share of national
income translated into a 1 per cent decrease in the ratio of mortgage
debt to disposable income. The results also show that the rising inequal-
ity in the distribution of wage earnings had an adverse effect on mort-
gage debt. A rise in earnings inequality translated into less mortgage
debt as well. Thus, our results suggest that the increasing unequal

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
138 Rising Inequality in the USA

Table 6.7 1Regression results for determinants of mortgage; debt, 1980--2003

(4) (5) * (6)


(2) (3) Alternative Alternative Combined
Explanatory (1) Between Within between within labour
variables Baseline inequality inequality inequality inequality inequality

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Yt-i 0.51*** 0.59*** 0.66*** 0.63 0.65*** 0.66***
(0.11) (0.16) (0.12) (0.11) (0.12) (0.12)
h -0.05* -0.03 -0.03 -0.03 -0.03 -0.03
(0.03) (0.03) (0.03) (0.03) (0.03) (0.03)
LI/NIt 0.95* 1.13** 1.09** 1.13**
(0.51) (0.51) (0.51) (0.51)
LPI/NIt 1.30**
(0.53)
Atkinson05t -0.27* -0.24*
(0.14) (0.14)
AtkinsonlO t -0.27*
(0.16)
SP500IHPIt -0.03 -0.03 -0.03
(0.03) (0.03) (0.03)
LIneqt -0.03**
(0.02)
Constant -4.82*** -5.12*** -6.31*** -6.10*** -6.12*** -5.68***
(0.98) (0.97) (1.09) (1.05) (1.08) (0.98)
N 93 93 93 93 93 93
Adj. 0.35 0.41 0.45 0.50 0.46 0.25
i?-squared
Durbin- 1.37 1.39 1.50 1.51 1.48 1.98
Watson

Notes: In each case, a Prais-Winsten regression is used. LPI refers to labour income plus pro-
prietors' income and AtkinsonlO refers to the Atkinson inequality measure with 1.0 parameter
instead of 0.5. LIneq refers to the combined labour inequality measure derived by using the
first factor. All inequality measures refer to the four-quarter average ending in the current
quarter. Figures in parentheses are standard deviations. * denotes significance at the 10%
level, ** denotes significance at the 5% level, and *** denotes significance at the 1% level.

distribution between capital and labour away from labour and the rising
inequality of labour income lowered the amount of mortgage below
where it otherwise would have been.
Our results on mortgage debt stand in contrast to our results on credit-
card debt (Table 6.8). We find no connection between an increasingly
unequal distribution between capital and labour and the amount
of credit-card debt, but we find that greater inequality within labour
results in more credit-card debt, regardless of which measure is used. For
instance, a 1 per cent increase in wage inequality has typically resulted
in a 0.5 per cent increase in credit-card debt relative to disposable

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Heather Boushey and Christian E. Weller 139

Table 6.8 Regression results for determinants of credit-card debt, 1980-2003

(4) (5) (6)


(2) (3) Alternative Alternative Combined
Explanatory (1) Between Within between within labour
variables Baseline inequality inequality inequality inequality inequality

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
yt-i 1.19*** 1.18*** 0.99*** 1.06*** 0.98*** 0.99***
(0.22) (0.24) (0.23) (0.22) (0.23) (0.23)
prt 0.001 -0.01 -0.02 -0.01 -0.02 -0.02
(0.04) (0.04) (0.04) (0.04) (0.04) (0.04)
LI/NIt -0.81 -1.28 -1.23 -1.28
(1.05) (1.04) (1.04) (1.04)
LPI/NIt -0.86
(1.08)
Atkinson05t 0.51* 0.46*
(0.28) (0.28)
Atkinson 101 0.61*
0.12* 0.11* (0.33)
SP500/HPIt (0.06) (0.06) 0.11*
(0.06) 0.08**
LIneqt (0.03)
Constant -13.21*** -12.91*** -10.78*** -11.22 -10.83*** -11.97***
(1.92) (1.96) (2.15) (2.11) (2.11) (1.92)
N 93 93 93 93 93 93
Adj. 0.78 0.85 0.85 0.85 0.85 0.84
fl-squared
Durbin- 1.48 1.93 1.93 1.92 1.93 1.93
Watson

Notes: In each case, a Prais-Winsten regression is used. Credit-card debt refers to the share of
revolving credit out of total household debt. LPI refers to labour income plus proprietors'
income and AtkinsonlO refers to the Atkinson inequality measure with 1.0 parameter instead
of 0.5. LIneq refers to the combined labour inequality measure derived by using the first
factor. All inequality measures refer to the four-quarter average ending in the current quarter.
Figures in parentheses are standard deviations. * denotes significance at the 10% level,
** denotes significance at the 5% level, and *** denotes significance at the 1% level.

income. For the period from 1980 to 2003, our inequality measure had
a standard deviation that was 6.5 per cent of its mean. Thus, a one stan-
dard deviation increase would explain a 3.2 per cent increase in debt
relative to disposable income. This is a small fraction of the total
increase in credit-card debt relative to disposable income as it rose
almost fourfold over the period from 1980 to 2003. In comparison, a
1 per cent increase in our capital income proxy translated into an
increase of 0.1 per cent in the ratio of credit-card debt to disposable
income. The standard deviation of our capital income inequality proxy
was 56 per cent of its average, which would have meant a 5.6 per cent
increase in the ratio of credit-card debt to disposable income.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
140 Rising Inequality in the USA

One way to combine the results on mortgage debt and credit-card


debt is to use the share of credit-card debt out of total debt as the
dependent variable (Table 6.9). The results show that the trend towards
rising inequality between labour and capital and within labour has
resulted in a larger share of credit-card debt relative to total household

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
debt over the period from 1980 to 2003. As the labour share of national
income has trended downward, households increasingly shifted their
borrowing towards credit-card debt, possibly because it was more
accessible than mortgage debt. A 1 per cent decrease in the labour
share of national income resulted in a 2 per cent increase in the relative

Table 6.9 )Regression results for determinants of debt composition, 1980-2003

(4) (5) (6)


(2) (3) Alternative Alternative Combined
Explanatory (1) Between Within between within labour
variables Baseline inequality inequality inequality inequality inequality

yu 0.34*** 0.59*** 0.44*** 0.53*** 0.44*** 0.44***


(0.23) (0.17) (0.16) (0.16) (0.16) (0.16)
prt 0.04 0.01 0.0005 0.01 0.004 0.001
(0.03) (0.03) (0.03) (0.03) (0.03) (0.03)
LI/Mt -1.53** -2.08*** -2.00*** -2.08***
(0.76) (0.72) (0.71) (0.71)
LPI/NIt -1.79**
(0.75)
Atkinson05t 0.66*** 0.60***
(0.20) (0.20)
AtkinsonlO t 0.76***
(0.22)
SP500/HPIt 0.11** 0.10** 0.10**
(0.04) (0.04) (0.04)
LIneqt 0.10***
(0.02)
Constant -687.39*** -8.60*** -5.86*** -6.52*** -6.06*** -7.43***
(317.39) (1.46) (1.49) (1.48) (1.46) (1.32)
N 93 93 93 93 93 93
Adj. 0.03 0.78 0.90 0.89 0.90 0.90
i?-squared
Durbin- 1.50 1.45 1.75 1.73 1.74 1.75
Watson

Notes: In each case, a Prais-Winsten regression is used. LPI refers to labour income plus
proprietors' income and AtkinsonlO refers to the Atkinson inequality measure with 1.0
parameter instead of 0.5. All inequality measures refer to the four-quarter average ending in
the current quarter. Figures in parentheses are standard deviations. * denotes significance
at the 10% level, ** denotes significance at the 5% level, and *** denotes significance at the
1% level.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Heather Boushey and Christian E. Weller 141

borrowing from credit cards. Also, increasing inequality within labour


led to a growing share of credit-card debt relative to total debt. A 1 per
cent increase in wage inequality, averaged over four quarters, translated
into an increase of 0.6 per cent in the relative size of credit-card debt and
a 1 per cent increase in capital income inequality translated into a

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
0.1 per cent increase in the share of credit-card debt. The size of the
effects is comparable to their impact on the volume of credit.

Inequality and household economic distress


The rise in consumer debt could also have given rise to household eco-
nomic distress if it was more pronounced for lower-income households
(Iyigun and Owen 1997). The increase in debt levels, caused by a grow-
ing disparity between labour and capital, should increase financial dis-
tress. So should the rise in labour inequality as more costly credit-card
debt disproportionately increases among lower-income earners. Already,
the largest increases in consumer default exist with respect to credit-card
debt. The charge-off rate on credit-card loans increased almost threefold
from 2.0 per cent in 1985 to 5.9 per cent at the end of 2003, while it only
doubled on other consumer loans and declined on real-estate loans
(residential and commercial combined) (BOG 2004a).
The rise in economic distress measures is somewhat surprising given
the fundamental characteristics of household finances. Importantly, the
debt service burden rose from an average of 11.9 per cent in the early
1980s to 13.0 per cent since 1999, a 9 per cent increase. At the time,
though, household debt relative to disposable income rose from
78.0 per cent to 105.8 per cent, or a 36 per cent increase. Moreover, cal-
culations based on data from the Fed (BOG 2004c) show that household
wealth relative to disposable income grew at the same time, too. Thus,
as households borrowed more, their assets increased even faster, while
the burden of repaying the debt rose much more slowly.
Based on these figures, one would not necessarily expect a rise in
economic distress, but that is exactly what has happened. Personal
bankruptcies, as a share of households rose fourfold from 0.4 per cent in
1980 to 1.5 per cent in 2003. Further, default rates on consumer credit
almost doubled from the early 1980s to the period since 1999, while
consumer credit relative to disposable income increased only by 23.4 per
cent. Thus, household economic distress clearly seems to have grown
faster than one would expect by merely looking at household debt
measures.
Increasing inequality may explain this divergence. Rising inequality
may have given rise to an endogenous credit expansion and thus a larger

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
142 Rising Inequality in the USA

credit supply, especially to those households that saw the smallest


income increases, that is low and moderate-income households (Bird
et al. 1998; Black and Morgan 1999). Additionally, lenders may have
screened their customers carefully and offered worse terms to customers
that were more likely to become delinquent on their payments than to

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
others (Ausubel 1999; Stavins 2000). Combined with the lack of upward
mobility, low and moderate-income households may have been caught
in an increasing cycle of economic distress, caused by the combination
of low income-growth, high debt-growth, and high debt-costs. Thus,
both types of inequality may have contributed to a rise in economic dis-
tress due to the ensuing demand for credit and the extension of rising,
and more costly credit, such as credit-card debt (Chatterjee et al. 2002,
Gross and Souleles 1998; Stavins 2000).
Our regression model to estimate the link between income inequality
and economic distress builds on the existing literature and extends it by
including our three income inequality measures. We use three measures of
economic distress, personal bankruptcy, charge-off rates on all consumer
credit and charge-off rates on credit-card debt. The inclusion of inequality
measures in addition to debt service and debt composition essentially con-
trols for the fact that credit may have expanded in forms not captured here,
for example payday loans and pawnshops. Economic distress is typically a
function of income growth, debt composition, out-of-pocket medical
expenditures, debt service and unemployment, in addition to demographic
characteristics (Ausubel 1997; Chaterjee et al. 2002; Gross and Souleles
1998; Stavins 2000). We supplement this model with our measure for
between-labour and capital inequality and within-labour inequality:

(BR) _ R , a T m , R MEX , R Yt ~ Yt-\ , aCCR , R DS

+ &(. 2 J r ) / 4 + P\. tAtkinsonOSy*

where BR refers to the total number of quarterly bankruptcy cases, at an


annualized rate, HH refers to the total number of households,7 MEX to
total medical expenditures, CCR to credit-card debt, TC to total house-
hold credit, and DS to debt services. The source for medical expenditures
is BEA (2004), the source for the unemployment rate is BLS (2004c),

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Heather Boushey and Christian E. Weller 143

and the source for debt service is BOG (2004b). To make the results
comparable for both our economic distress variables, we estimate the
regression for the period from 1985 to 2003.8
Table 6.10 summarizes our regression results for all of our economic
distress variables, personal bankruptcies and charge-off rates.9 All

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
explanatory variables either have the expected sign or are statistically
insignificant. Regression (1) presents our baseline results for personal
bankruptcies. The share of households declaring bankruptcy in a given
year is negatively related to income growth, positively to medical expen-
ditures, to the composition of debt and to debt service. Also, inequality
between labour and profits does not affect the bankruptcy rate as regres-
sion (2) shows. However, inequality within labour, especially arising
from wealth inequality, results in a higher personal bankruptcy rate as
our results in regression (3) suggest. Using our combined labour inequal-
ity measure in regression (4) does not change the results materially. A
rise in labour inequality by 1 per cent results in a rise in the personal
bankruptcy rate by 0.2 per cent. Thus, a one-standard deviation increase
in labour inequality, which equals 217 per cent of the average labour
inequality, would result in a 39 per cent increase in the personal bank-
ruptcy rate. In comparison, though, the personal bankruptcy rate rose
by 473 per cent from 1985 to 2003. Thus, rising inequality within labour
has significantly contributed to the increase in personal bankruptcies,
but it does not explain the majority of the increase. Again, it is impor-
tant to keep in mind that this is an average effect that ignores the fact
that slow income growth at the bottom of the income scale went hand-
in-hand with a disproportionate increase of credit-card debt, non-bank
loans and economic distress among lower-income households. Thus,
the effect is likely more pronounced by income.
The results differ somewhat for the regressions on charge-off rates on
consumer credit and credit-card debt (Table 6.10). For instance, the
unemployment rate, instead of personal income growth, is a significant
indicator of default. Higher unemployment rates raise the rate of default
which may be a reflection of the fact that unemployment rates are more
volatile among low and middle-income households, where the expan-
sion of credit-card debt has been more pronounced. However, health-
care expenditures did not play a significant role in determining default
rates, which may reflect the fact that healthcare coverage and thus out-
of-pocket medical expenditures are unequally distributed. Further, the
debt-service burden has a much more pronounced effect on default rates
than on personal bankruptcies as the size of the effect almost doubles.
A 1 per cent increase in the debt-service burden raises the default rate

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect -
Table 6.10 Regression results for economic distress measures
Explanatory (1) (2) (3) (4) (5) (6) (7) (8)
variables Baseline Between Within Combined Within Combined Within Combined
inequality inequality labour inequality inequality labour inequality inequality labour inequality

Personal bankruptcy rate Charge-offi "ate, consumer credit Charge-off rate, credit cards

(y-yt-i)/yt-i -0.02* -0.02* -0.02* -0.02* -0.01 -0.01 -0.01 -0.01


(0.01) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01)
URt -0.03 -0.03 0.24 0.13 0.49** 0.35 0.72*** 0.67**
(0.02) (0.17) (0.19) (0.19) (0.24) (0.23) (0.24) (0.25)
(MEXJY)t 1.80*** 1.81*** ^ yg*** 1.84*** 0.66 0.80 0.13 0.80
(0.43) (0.44) (0.41) (0.44) (0.58) (0.59) (0.49) (0.63)
(CCR/TC)t 0.64** 0.64** -0.02 0.17 -0.08 -0.25 0.77 -0.09

2014-03-15
(0.30) (0.31) (0.40) (0.40) (0.60) (0.61) (0.51) (0.68)
(DS/Y)t 0.96* 0.96* 0.95* 0.87 1.76** 1.92** 1.69*** 1.73**
(0.52) (0.53) (0.48) (0.53) (0.75) (0.77) (0.57) (0.79)
LI/NIt -0.08 -2.47 -2.47 -0.64 -1.70 1.76 -2.16
(2.69) (2.85) (2.99) (3.42) (3.47) (3.51) (3.63)
Atkinson05t 0.60 0.38 -0.70
(0.78) (0.87) (0.99)
SP500/HPIt 0.40*** 0.44 0.31*
(0.15) (0.15) (0.18)
LIneqt 0.18* 0.21** 0.21**
(0.10) (0.09) (0.10)
N 76 76 76 76 76 76 76 76
fl-squared 0.96 0.96 0.96 0.96 0.13 0.08 0.25 0.11
Durbin-Watson 1.78 1.78 1.81 1.80 2.08 1.96 1.97 1.96

Notes: In each case, a Prais-Winsten regression is used. All inequality measures refer to the four-quarter average ending in the current quarter. Figures
in parentheses are standard deviations. * denotes significance at the 10% level, ** denotes significance at the 5% level, and *** denotes significance at
the 1% level.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim Truger
Heather Boushey and Christian E. Weller 145

by 1.7-1.9 per cent as regressions (5) through (8) show. The effect of the
within-labour inequality on charge-off rates is similar to that on per-
sonal bankruptcy rates as they also rose fourfold from 1985 to 2003.

Conclusion

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
In this chapter we have looked at the macro-economic effects of rising
inequality in the USA. The distribution of national income between
capital and labour has become more unequal as has the distribution
within labour. At the same time that inequality rose, consumer debt and
household economic distress grew, too. The evidence on a positive link
between rising inequality and innovation is not supported by the data.
The data either suggest no connection or a potential negative link.
In comparison, the link between growing inequality and aggregate
demand is somewhat ambiguous since macro economic data show a
negative connection that is not supported by the micro data. One way
to clarify this ambiguity is the possibility that debt has increased and
that it has increased more among low-income households. This seems to
be the effect, especially when considering more costly forms of debt
such as credit cards and non-bank credit. We find that credit-card debt is
especially sensitive to changes in inequality and that rising inequality
may thus have contributed to the stark increases in personal bankrupt-
cies in the USA. However, while our results show that rising inequality has
played a non-trivial role in the expansion of some forms of credit, and
that it has contributed to the large increases in economic distress over the
past two decades, it explains only a small part of the overall increases in
debt levels and in economic distress. Yet, a look at aggregate data ignores
the fact that changes in inequality, consumer credit and economic distress
likely affected lower-income households more than others.

Notes
1 All comparisons are for nine quarters or 28 months after the start of the
recovery.
2 All differences were statistically significant at least at the 5% level.
3 Simultaneously, labour income became more unequally distributed, exacer-
bating the trends discussed here.
4 For ease of comparison, all figures are divided by consumer spending.
5 All series are non-stationary, integrated at the first degree, and cointegrated at
the 1% level.
6 Credit-card debt refers to revolving consumer credit (BOG 2004e).
7 The total number of households is calculated by dividing the number of
housing units by 1.1.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
146 Rising Inequality in the USA

8 Our previous results on debt levels are unaffected by the choice of time
period.
9 The results remain robust if the bankruptcy rate is calculated separately for
Chapter 7 and Chapter 13 filings. They also are robust if the between-inequality
measure includes proprietors' income and the AtkinsonlO measure instead of
the Atkinson05 measure used for wage inequality.

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
References
Alesina, A. and Perotti, R. (1996) 'Income Distribution, Political Instability, and
Investment', European Economic Review, 40(6): 1203-28.
American Bankruptcy Institute (2004) Non-Business Bankruptcy Filings by Chapter,
1990-2004, per Quarter (Alexandria: ABI).
Appelbaum, E. and Weller, C. (2001) 'Structural Change, Employment, and
Inequality: Impacts on Productivity Growth', in A. Heise (ed.), USA - Modellfall
der New Economy? (Marburg: Metropolis).
Arestis, P. and Howell, P. (1995) 'Changes in Income Distribution and Aggregate
Spending: Constraints on Full-Employment?', Review of Political Economy, 7(2):
150-63.
Atkinson, R. (2003) The Bush Manufacturing Crisis, Policy Report (Washington DC:
Progressive Policy Institute).
Ausubel, L. (1997) 'Credit Card Defaults, Credit Card Profits, and Bankruptcy',
American Bankruptcy Law Journal, 71 (Spring): 249-70.
Barr, M. (2001) Five Opportunities for the Bush Administration and the 107th Congress
(Washington, DC: The Brookings Institution Center on Urban and
Metropolitan Policy).
Benhabib, J. (2003) 'The Tradeoff Between Inequality and Growth', Annals of
Economics and Finance, 4(2): 491-507.
Bird, E. J., Hagstrom, P.A. and Wild, R. (1999) 'Credit Card Debts of the Poor: High
and Rising', Journal of Policy Analysis and Management, 18(1): 125-33.
Bivens, J. and Weller, C. (2004) Institutional Shareholder Concentration, Corporate
Governance Changes, and Diverging Fortunes of Capital and Labour, paper pre-
sented at the conference 'Pension Fund Capitalism and the Crisis of Old-Age
Security in the United States', September, held by the Center for Economic
Policy Analysis at the New School University, New York, NY.
Bivens, J. and Weller, C. (2005) 'Causes of the "Job Loss" Recovery', Challenge,
48(2): 23-48.
Bivens, L. (2004) 'Shifting Blame for Manufacturing Job-loss: Effects of Rising Trade
Deficits can't be Ignored', Economic Policy Institute Briefing Paper (Washington,
DC: EPI).
Black, S. and Morgan, D. (1999) 'Meet the New Borrowers', Current Issues in
Economics and Finance, Federal Reserve Bank of New York, February (New York:
FRBNY).
Board of Governors and Federal Reserve (2004a) Charge-Off and Delinquency Rates
on Loans and Leases at Commercial Banks (Washington, DC: BOG).
Board of Governors and Federal Reserve (2004b) Household Debt Service and
Financial Obligations Ratios (Washington, DC: BOG).
Board of Governors and Federal Reserve (2004c) Release Z.l Flow of Funds Accounts
of the United States (Washington, DC: BOG).

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Heather Boushey and Christian E. Weller 147

Board of Governors and Federal Reserve (2004d) Release H. 15 Selected Interest Rates
(Washington, DC: BOG).
Board of Governors and Federal Reserve (2004e) Release G.19 Consumer Credit
(Washington, DC: BOG).
Borjas, G. and Ramey, V. (1995) 'Foreign Competition, Market Power, and Wage
Inequality', Quarterly Journal of Economics (November), 110(4): 1075-110.

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Bradbury, K. and Katz, J. (2002) 'Are Lifetime Incomes Growing More Unequal?
Looking at New Evidence on Family Income Mobility', Regional Review, 4: 3-5.
Bradbury, K. and Katz, J. (2004) Wives' Work and Family Income Mobility, Federal
Reserve Bank of Boston Public Policy Discussion Paper, No.(04-3) (Boston,
MA: FRBB).
Brown, C. (2004) 'Does Income Distribution Matter for Effective Demand?
Evidence from the United States', Review of Political Economy, 16(3): 291-307.
Bureau of Economic Analysis (2004) National Income and Product Accounts
(Washington, DC: BEA).
Bureau of Labor Statistics (2004a) Productivity (Washington, DC: BLS).
Bureau of Labor Statistics (2004b) Consumer Price Index (Washington, DC: BLS).
Bureau of Labor Statistics (2004c) Unemployment Rate (Washington, DC: BLS).
Card, D., Lemieux, T. and Riddell, W.C. (2003) 'In Unionization and Wage
Inequality: A Comparative Study of the U.S., the U.K., and Canada', NBER Working
Paper No. 9473 (Cambridge MA: National Bureau of Economic Research).
Census Bureau (2004a) Income, Poverty, and Health Insurance Coverage in the United
States: 2003 (Washington, DC: Census).
Census Bureau (2004b) Households and Housing Units Estimates (Washington, DC:
Census).
Center for Economic and Policy Research (CEPR) (2004) CPS ORG Uniform Data
Files Version 0.9.3 (Washington, DC: CEPR).
Chaterjee, S., Corbae, D., Nakajima, M. and Rios-Rull, J. (2002) 'A Quantitative
Theory of Unsecured Consumer Credit Risk with Default, Federal Reserve Bank of
Philadelphia Working Paper No. 02-6 (Philadelphia, PA: FRBP).
Chen, B. (2003) 'An Inverted-U Relationship Between Inequality and Long-run
Growth', Economics Letters, 78(2): 205-12.
Cline, W. (1997) Trade and Income Distribution (Washington, DC: Institute for
International Economics).
Consumer Federation of America (1998) The Growth of Legal Loan Sharking: A
Report on the Payday Loan Industry (Washington, DC: CFA).
Consumer Federation of America (1999) Safe Harbor for Usury: Recent Developments
in Payday Lending (Washington, DC: CFA).
Crafts, N. (1992) Was the Thatcher Experiment Worth It? British Economic Growth
in a European Context, CEPR Discussion Paper (710) (London, UK: Center for
Economic and Policy Research).
DiNardo, J., Fortin, N.M. and Lemieux, T. (1996) 'Labor Market Institutions and
the Distribution of Wages, 1973-1992: A Semiparametric Approach', Econometrica,
64(5): 1001-44.
Dunsky, R. and Follain, J. (2000) 'Tax-induced Portfolio Reshuffling: The Case of
the Mortgage Interest Deduction', Real Estate Economics, 28(4): 683-718.
Engen, E. and Gale, W (2000) The Effects of 401 (k) Plans on Household Wealth:
Differences Across Earnings Groups, NBER Working Paper No. 8032 (Cambridge
MA: National Bureau of Economic Research).

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
148 Rising Inequality in the USA

Gottschalk, P. (1997) 'Inequality, Income Growth, and Mobility: The Basic Facts',
Journal of Economic Perspectives, 11(2): 21-40.
Gottschalk, P. and Danziger, S. (2003) Wage Inequality, Earnings Inequality, and
Poverty in the US. Over the Last Quarter of the Twentieth Century (Boston, MA:
Boston College).
Gross, D. and Souleles, N.S. (1998) An Empirical Analysis of Personal Bankruptcy

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
and Delinquency, Financial Institutions Center Working Paper no. 98-28-B
(Philadelphia, PA: The Wharton School, University of Pennsylvania).
Iyigun, M. and Owen, A. (1997) Income Inequality and Macroeconomic Fluctuations,
FEDS Working Paper (1997-586) (Washington, DC: Board of Governors of the
Federal Reserve System).
Keynes, J. M. (1936) The General Theory of Employment, Interest and Money
(Cambridge, UK: Cambridge University Press).
Krueger, D. and Perri, F. (2002) Does Income Inequality Lead to Consumption
Inequality? Evidence and Theory, NBER Working Paper no. 9,202 (Cambridge,
MA: National Bureau of Economic Research).
Larrain, F.B. and Vergara, R.M. (1997) Income Distribution, Investment, and Growth,
Harvard Institute for International Development Working Paper (Cambridge,
MA: Harvard University).
Lazonick, W. and O'Sullivan, M. (2000) 'Maximizing Shareholder Value: A New
Ideology for Corporate Governance', Economy and Society, 29(1): 13-35.
Lee, David S. (1999) 'Wage Inequality in the United States during the 1980s:
Rising Dispersion of Falling Minimum Wage?' Quarterly Journal of Economics,
114(3): 977-1023.
Manning, R. (2000) Credit Card Nation: The Consequences of America's Addiction to
Credit (New York, NY: Basic Books).
Mishel, L., Bernstein, J. and Allegretto, S. (2005) State of Working America 2004-05
(Ithaca, NY: Cornell University Press).
Mishel, L., Bernstein, J. and Schmitt, J. (2001) The State of Working America
2000-01 (Ithaca, NY: Cornell University Press).
Mishel, Lawrence, Bernstein, J. and Boushey, H. (2003) The State of Working
America 2002-03 (Ithaca, NY: Cornell University Press).
O'Sullivan, M. (2000) Contests for Corporate Control (Oxford UK et al: Oxford
University Press).
Office of Federal Housing Enterprise Oversight (2004) Housing Price Index
(Washington, DC: OFHEO).
Osberg, L. (2003) 'Understanding Growth and Inequality Trends: The Role of
Labour Supply in the US and Germany', Canadian Public Policy (January
Supplement): 163-83.
Palley, T. I. (1996) 'Inside Debt, Aggregate Demand, and the Cambridge Theory of
Distribution', Cambridge Journal of Economics, 20(4): 465-474.
Palomba, G. (2002) Firm Investment, Corporate Finance, and Taxation, IMF Working
Paper no. 02/237 (Washington, DC: International Monetary Fund).
Panizza, U. (2002) 'Income Inequality and Economic Growth: Evidence from
American Data', Journal of Economic Growth, 7(1): 25-41.
Pearce, J. and Miller, J. (2003) Freddie Mac and Fannie Mae: Their Funding Advantage
and Benefits to Consumers. Federal Reserve Bank of Chicago Proceedings (March
2003) (Chicago: Federal Reserve Bank of Chicago).

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Heather Boushey and Christian E. Weller 149

Peretto, P. (1995) Sunk Costs, Market Structure, and Growth, Duke University,
Department of Economics Working Paper no. 95-34 (Durham, NC: Duke
University).
Piketty, T. and Saez, E. (2001) Income Inequality in the United States, 1913-1998,
NBER Working Paper no. 8467 (Cambridge MA: National Bureau of Economic
Research).

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Pressman, S. (1997) 'Consumption, Income Distribution and Taxation: Keynes'
Fiscal Policy', Journal of Income Distribution, 7(1): 29-44.
Rodriguez, C. (2000) 'An Empirical Test of the Institutionalist View on Income
Inequality: Economic Growth Within the United States', American Journal of
Economics and Sociology, 59(2): 303-13.
Rupasingha, A., Goetz, S. and Freshwater, D. (2002) 'Social and Institutional
Factors as Determinants of Economic Growth: Evidence from United States
Counties', Papers in Regional Science, 81(2): 139-55.
Scully, G.W. (2002) 'Economic Freedom, Government Policy and the Trade-off
Between Equity and Economic Growth', Public Choice, 113(1-2): 77-96.
Stango, V. (1999) 'The Tax Reform Act of 1986 and the Composition of Consumer
Debt', National Tax Journal, 52: 717-39.
Stavins, J. (2000) 'Credit Card Borrowing, Delinquency, and Personal
Bankruptcy', New England Economic Review (July/August): 15-30.
Stegman, M.A. and Faris, R. (2003) 'Payday Lending: A Business Model that
Encourages Chronic Borrowing', Economic Development Quarterly, 17(1): 8-32.
Stiglitz, J. and Weiss (1981) Credit Rationing in Markets with Imperfect
Information, American Economic Review, 71(3): 393-410.
Stockhammer, E. and Onaran, O. (2004) 'Accumulation, Distribution and
Employment: A Structural VAR Approach to a Kaleckian Macro Model',
Structural Change and Economic Dynamics 15(4): 421-47.
Van Order, R. (2001) 'The Structure and Evolution of American Secondary
Mortgage Markets, with Some Implications for Developing Markets', Housing
Finance International 26(1): 16-31.
Vandell, K. (2000) Securitization of the US Mortgage Market: Progress and Pitfalls, with
Lessons for Japan, presentation at the Seventh Annual International Land Policy
Review.
Weller, C. (2004a) Reversing the Vpside-Down' Economy, CAP Policy Brief
(Washington, DC: Center for American Progress).
Weller, C. (2004b) Lagging Investment: The Cost of the Vpside-Down' Economy, CAP
Policy Brief (Washington, DC: Center for American Progress).
Weller, C, Bivens, J. and Sawicky, M. (2004) 'Macro Policy Lessons from the
Recent Recession', Challenge, 47(3): 42-72.
Wolff, E. (2002a) Is the Equalizing Effect of Retirement Wealth Wearing Off?, unpub-
lished manuscript (New York, NY: New York University).
Wolff, E. (2002b) Top Heavy: A Study of Increasing Inequality of Wealth in America,
newly updated and expanded edition (New York, NY: The New Press).
Wolff, E. (2003) 'What's Behind the Rise of Profitability in the U.S. in the 1980s
and 1990s?', Cambridge Journal of Economics, 27(4): 479-99.
Wolff, E. (2004) 'The Unraveling of the American Pension System, 1983-2001',
unpublished manuscript, Department of Economics, New York, NY: New York
University.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
150 Rising Inequality in the USA

Wysong, E., Wright, D. and Perrucci, R. (2003) 'Organizations, Resources, and


Class Analysis: The Distributional Model and the U.S. Class Structure', working
paper, Department of Sociology (Indiana University, IN: Kokomo).
Yahoo! Finance (2004) S&P 500 index data, www.yahoo.com.
Yoo, P. (1996) Charging up a Mountain of Debt: Households and Their Credit Cards.
Federal Reserve Bank of St. Louis, Working Papers (96-015A) (St Louis, MO:

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
FRBSTL).

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
7
The Effects of Economic
Liberalization on Income

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Distribution: A Panel-Data
Analysis
Gerardo Angeles-Castro

Introduction
The economic liberalization process, undertaken on a global scale since the
early 1980s, has induced support for a set of market-oriented policies
that can be summarized as deregulation, privatization, liberalization of
markets and macroeconomic discipline. This prescription creates the
preconditions for the expansion of trade and the flow of investments
across countries. The theoretical support for this development model is
standard neoclassical theory (Jones 1988: 30-3; Corden 1993), which
argues that trade, investment, and in general the market mechanisms
boost growth and facilitate development. This view also holds that an
important factor affecting growth and the effectiveness of the market
system is the efficient organization of the domestic economy itself
(Gilpin 1987: 265-6).
The implications of this model for income distribution are that high
and sustained rates of growth and the expansion of exports foster employ-
ment, reduce poverty and eventually provide additional resources that
facilitate the distribution of income.
Moreover, economic liberalization facilitates the operation of market
forces and the adjustment to world prices, which allow resources to be
allocated more efficiently. The theoretical formulations explaining this
effect are the orthodox principle of comparative advantages (Jones
1988: 34-5) and the Stolper-Samuelson theorem (FitzGerald 1996: 32;
Litwin 1998: 3). The latter is a neoclassical two-factor model, in which
liberalization of foreign trade increases the use of the cheaper-abundant
factor as exports and imports adjust according to the principle of

151

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
152 Economic Liberalization and Income Distribution

comparative advantages, while the costly-scare factor is used less. This


mechanism increases the income of the factor which is relatively most
used in the export sector and which is also most abundant. For example,
this factor is conventionally assumed to be unskilled labour in develop-
ing countries, by the same token income distribution is assumed to

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
improve.
The economic policies involved in this early stage of the economic lib-
eralization process are often referred to as the Washington Consensus1
or first generation reforms (Ortiz 2003: 14-17). These terms are applied
especially in developing countries.
This chapter is aimed at testing the effect of the variables that are
assumed to improve income distribution over this early process of eco-
nomic liberalization. These variables are mainly trade, investment,
macroeconomic discipline and employment.
Over the last few years, leading globalizers and multilateral institu-
tions have induced support for a set of socio-political norms and have
also recognized the need for a stronger governance dimension. They
have added institution-building, civil society participation, social and
human capital formation, safety nets, transparency and accountability,
among others, to the original economic norms or first-generation policies
enveloped in the Washington Consensus. This new global governance
agenda is a response to the financial crises that have hit emerging markets
since 1995, the increasing perception that liberalization brings with it
inequality, and other subsequent forms of resistance to globalization
(Higgott 2000: 131-40).
This further stage in the economic liberalization process is usually
called Post-Washington Consensus (PWC) and the set of socio-political
norms embedded in this approach are often referred to as second-
generation reforms. The PWC is an attempt to socialize and humanize
the operation of market forces and to legitimize global economic liber-
alization, although there is also a genuine recognition of the importance
of tackling issues of fairness and inequality (Edwards 1999). In this
development paradigm, the re-empowerment of the state plays a central
role for addressing the socioeconomic dislocation that may be generated
by global liberalization. In this context, the understanding of governance
is thus the effective and efficient management of the modern state.
Moreover, from this perspective, domestic efficiency and sound and
disciplined macroeconomic policies accentuate the benefits of global-
ization. These policies are associated with sustainable economic growth
and are assumed to improve equity over the long run. In contrast, those
countries which do not adopt sound policies and show evidence of

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Gerardo Angeles-Castro 153

pronounced macroeconomic disequilibria are likely to fall behind in


relative terms (IMF 1997: 72; Camdessus 1998: xiv; Higgott and Phillips
2000: 363). In this context, the chapter also tests the effects of the PWC
approach on income distribution. For this purpose we use variables such
as government expenditure to proxy the size of the state, and we also

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
explore the effect of first generation policies under different scenarios of
macroeconomic stability and governance. Finally, we test the effect of
human capital formation, represented by secondary school enrolment,
on income distribution. Education is deemed as one of the key elements
in the set of socio-political norms advocated by the PWC.
We find that the variable on trade exerts a weak benefit on income
distribution and the effect is statistically significant only on the samples
which comprise countries associated with good governance or macro-
economic stability, while the effect is significant only on the latter
sample when the variable on trade is represented by changes in trade
volume. The flow of FDI increases inequality under any scenario
although the effect is mitigated in those countries that exhibit domestic
efficiency. Inflation worsens inequality in those countries with domestic
inefficiency, and this variable adversely affects income distribution even
in those countries which exhibit good governance, although in any case
the effect is weak. Not surprisingly, inflation is not significant in those
countries that have traditionally kept macroeconomic discipline.
The export-led growth strategy is able to reduce inequality when the
export sector is oriented toward manufactured production. On the other
hand, the expansion of primary exports does not exert benefit on
income distribution under any scenario. Employment benefits income
distribution; however when we explore the effects of employment by
sector we notice that the expansion of employment in industry reduces
inequality, whereas employment in agriculture is not able to improve
income distribution. Consequently, the analysis of exports and employ-
ment by sector suggests that emphasis on primary production does not
form the basis for redistributional effects. Even those countries with
some form of domestic inefficiency, which are associated with lower
levels of development and also with comparative advantages supported
on natural resources and unskilled labour, do not seem to improve
income distribution with the expansion of primary exports.
These results are in keeping with the theoretical foundations and
expectations that have supported the global liberalization process to
the extent that low inflation, fiscal discipline, larger employment and
domestic efficiency can benefit income distribution. On the other hand,
the results undermine other aspects of these theoretical foundations and

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
154 Economic Liberalization and Income Distribution

expectations, as the benefits of trade on income distribution are weak


and FDI worsens inequality, besides the fact that the export-led growth
strategy and the expansion of employment based on the primary sector
do not improve income distribution. Consequently, the results suggest
that there is room for contesting theories and hypotheses explaining the

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
relationship between trade, investment and income distribution and
the relationship between primary production, industrialization and
income distribution.
As for the set of policies underlined in the second stage of the eco-
nomic liberalization process, we find that domestic efficiency can help
to reduce the adverse effect of FDI on income distribution, while it can
also help to obtain some benefits from trade liberalization in income
distribution. Moreover, a stronger state is important to decrease inequal-
ity, and the set of socio-political norms enclosed in the PWC approach,
and represented in this study by a proxy of human capital formation,
improves income distribution. Accordingly, these findings suggest that
the PWC represents an improvement to mitigate the adverse effects
that the global process of economic liberalization can exert on income
distribution.
Nevertheless, the empirical evidence in this study suggests that the
role of the state is not enough to socialize the operation of trade and
investment. In this context, even under conditions of macroeconomic
stability and high governance, FDI does not benefit income distribution;
on the contrary, it seems to hinder it. Moreover, the beneficial effect of
trade on income distribution is weak in those countries with domestic
efficiency. Hence, this fact suggests that further supranational mecha-
nisms, beyond the scope of the state, are required to socialize the flow of
trade and investment.
The chapter is organized as follows. The next section analyses the
characteristics of the data-sets on income distribution available in the
literature and selects the appropriate option for this study. We then pres-
ent the features of the explanatory variables included in the model, and
explain the econometric method applied in the analysis. The fourth
section gives results, and is followed by concluding remarks.

The data

Data on income inequality


One of the features which characterize the available data-sets on income
inequality is that the coverage is sparse and varies widely across coun-
tries and over time. In the absence of adequate longitudinal data, some

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Gerardo A ngeles-Cas tro 155

studies attempting to assess the trend of inequality worldwide over time


draw general conclusions from cross-sectional data so as to try to over-
come this major drawback (Bourguignon 1994; Milanovic 1995; Jha
1996). However, this type of data does not deal with intertemporal rela-
tionships. In addition, other studies restrict attention to a subset of the

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
data such as five-year intervals (Forbes 2000; De Gregorio and Lee 2002)
or group the data in five-year or ten-year averages (Deininger and Squire
1998; Calderon and Chong 2001), but in these cases there is a risk of
bias in the selection of the subset or in the construction of the average
respectively. Finally, in order to improve coverage across space and
through time, other studies use data-sets measuring a component of
overall income inequality (Galbraith and Kum 2002); however, these
data-sets are not representative samples covering all of the population.
Accordingly, so as to provide an accurate assessment, the data-set on
income inequality must contain a substantial coverage across countries
and over time. It must also be consistent and harmonized, and based on
a representative measure covering all of the population.
The competing options available in the literature have the following
characteristics: the World Bank data-set by Deininger and Squire (hereafter,
D & S) in its 1996 version and the World Income Inequality database are
important compilations of Gini coefficients reported in the literature;
however, they suffer from a substantial degree of heterogeneity and
sparse coverage. The Luxembourg Income Study assembles observations
that are more harmonized than the previous two data-sets, since it obtains
information from standardized macro-level data, but its coverage is con-
strained to 29 countries. The UTIP-INIDO data-set calculates industrial
pay-inequality. It has a large coverage and shows evidence of consistency
and accuracy, but it is not an index of overall inequality.
The Estimated Household Income Inequality (EHII), constructed by
Galbraith and Kum (2003), aims to fill the gaps and correct what they
consider errors in the D & S data-set. This indicator takes advantage of
the information in D & S and the information in the UTIP-UNIDO data-
base. As a matter of fact, they replicate the coverage of the latter with
estimated measures of household income inequality, taking into
account the relationship between industrial pay inequality, household
income inequality and an additional set of variables.
Galbraith and Kum (2003: 9) underline two advantages from their
data-set:

First, the coverage basically matches that of the UTIP-UNIDO exer-


cise, providing substantially annual estimates of household income

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
156 Economic Liberalization and Income Distribution

inequality for most countries, including developing countries that


are badly under-represented in D & S. Second, the data set borrows
accuracy from the UTIP-UNIDO ... with due adjustment for the
different weight of manufacturing in different economies. At the same
time, unexplained variations in the D & S income inequality measures

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
are treated for what they probably are: as inexplicable. They are there-
fore disregarded in the calculations of the UTIP Gini coefficients.

The EHII data-set provides consistency, representation of household


income inequality, and large coverage. In addition, it is a new data-set
available in the literature which can provide further insights for the
study of income distribution. As a consequence, we select it from the
competing data sets as a source of information for this study. Table A7.1
in the Appendix outlines the main features of the data-sets on inequality.

Explanatory variables
The analysis includes four different sets of explanatory variables. The
first set comprises proxies representing markets, and it is aimed at test-
ing the effects of the original Washington Consensus prescription on
inequality. The second set contains macroeconomic stability proxies
and it will explore the role of macroeconomic efficiency on income
inequality. The third set includes governance proxies, and it is aimed at
assessing the influence of the PWC themes on the distribution of
income. Finally, the fourth set contains variables related to employ-
ment; these variables are included because employment is considered
one of the main factors affecting inequality in the neoliberal model. The
variables are analysed on a yearly basis and they are described as follows.
The market proxies are trade volume, which is the sum of exports and
imports of goods and services measured as a share of GDP. Alternatively,
we also use changes in trade volume. So as to test the effect of exports by
sector on income distribution we apply two variables representing the
percentage of manufactured exports and the percentage of primary
exports to merchandise exports. In order to represent the international
flow of investment we use FDI inflow measured as a percentage of GDP.
The source for the variables on trade is the World Development Indicators
(hereafter, WDI) by the World Bank (2002); the source for the variable
on FDI is a compilation of data from the WDI and UNCTAD (2003).
Inflation is included as both fiscal discipline and macroeconomic
stability proxy. It reflects the annual percentage of change in consumer
prices and is taken from the WDI. This analysis also uses standard devi-
ation of inflation for the period 1980-98, calculated from the same

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Gerardo Angeles-Castro 157

source to construct two sub-samples. The first sub-sample comprises


countries with standard deviation lower than 8.5, whereas the second
sub-sample groups the countries with standard deviation greater than
8.5. This indicator allows us to assess the effect of economic liberaliza-
tion on inequality in two different scenarios of macroeconomic stability.

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
The sub-sample classification is carried out through a two-step cluster
analysis.
The variables on governance comprise human capital formation,
which is one of the main elements included in the concept of gover-
nance, and is represented by gross secondary school enrolment. This
indicator is a compilation of data from UNESCO (2003) and WDI. The
analysis also includes an aggregate governance indicator for the year
1996. It is the average of six indicators measuring the following dimen-
sions of governance: voice and accountability, political stability and
absence of violence, government effectiveness, regulatory quality, rule
of law, and control of corruption. Its score lies between -3.0 and 3.0
with a higher score corresponding to better governance. The data-set is
obtained from the World Bank website. The indicator is used to con-
struct low governance and high governance sub-samples that involve
countries with scores lower than zero and greater than zero respectively.
From these two sub-samples it is possible to analyse the trend of inequal-
ity within two different performances of governance.
It has been claimed that the understanding of governance also involves
the reempowerment of the state as a means to tackle issues of inequality.
Bearing this in mind, we include government expenditure as a ratio of
GDP as a variable to represent the size of the state and to explore its effect
on income distribution. Because of constraints on data availability, this
exercise is carried out only with the overall sample. The source is WDI.
The employment proxies comprise an unemployment variable, which
is measured by the share of the labour force that is without work in
the total labour force. Furthermore, in order to assess the effect of
employment by sector on income distribution, we consider variables on
employment in two main sectors, Industry and Agriculture. They are
the proportion of total employment recorded as working in the corre-
sponding sector. As government expenditure, these variables are only
analysed with the overall sample because of constraints on data
availability. The source is WDI.

The dimension of the panel


The original equation includes trade, investment, inflation and educa-
tion variables. After assembling these variables and their different

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
158 Economic Liberalization and Income Distribution

sources of information, it is possible to construct an unbalanced panel


consisting of 1,301 observations across 93 countries over the period
1980-98. The list of countries is presented in Table A7.2 in the
Appendix. After including government expenditure and employment
variables, the number of observations and countries drops. Hence, when

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
these variables are included in the equation we conduct the regression
only for the overall sample. The analysis of primary and manufactured
exports is conducted in a separate equation including these variables
only. In this case the number of observations and countries also drops,
but allows us to conduct regressions across the sub-samples. In every
panel the number of time periods available may vary from country to
country, but the number of variables included in each year is the same.

The model

Initially we explore a general regression model for income inequality as


follows:

EHIIit = a{ + fr TRAGDPit + /32FDIGDPit


+ p3INFLit + /3 4 EDUSECit + uit (7.1)

where EHII is the estimated household income inequality indicator,


TRAGDP is the ratio of trade to GDP, FDIGDP is the inflow of FDI as a
percentage of GDP, INFL shows the annual percentage change of con-
sumer prices, and EDUSEC represents the gross secondary school enrol-
ment as outlined earlier. Subscripts i and t indicate country and year
respectively; the error term uit is assumed to satisfy white-noise assump-
tions; af lets the intercept vary for each country and captures country-
specific differences; and finally, (31 to /34 are parameters to be estimated.

Standard methods
The process of estimation starts with the standard ordinary-least-squares
(OLS) method pooling or combining all the observations, and assuming
that at = a. As column 1 in Table 7.1 illustrates, all the variables are indi-
vidually statistically significant. However, the traditional OLS approach
has two major drawbacks. It assumes that the intercept value of the
countries is the same and it does not control for country-specific
factors. So as to confirm whether these are implausible characteristics
in the model, the Breusch and Pagan Lagrange Multiplier (LM) test
(1980) is conducted. Based on the OLS residuals and under the null

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Gerardo Angeles-Castro 159

Table 7.1 Estimation methods

OLS (1) FEM (2) sys-GMM (3)

EHIIt_j 0.841*
TRAGDP -0.015* -0.003 -0.027*
0.474* 0.218* 0.594*

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
FDIGDP
INFL 0.001* 0.000 0.001*
EDUSEC -0.098* 0.054* -0.102*
Constant 44.702* 36.438* 45.959*
Adjustment coefficient 0.159
Observations 1,302 1,302 1,209
Countries 93 93 93
BP LM test [0.000]
Hausman test [0.000]
Sargan test [0.247]
AR (1) test [0.000] [0.000] [0.000]
AR (2) test [0.074]

Notes: Dependent variable: EHII, P values in parenthesis, * significant at i., column (3) fig-
ures are long-run parameters.

hypothesis: o^ = 0, that is at = a, the LM test is distributed as a \2 with


one degree of freedom (Greene 2000: 572-3). In this case, the result of
the test is to reject the null hypothesis.2 Thus, it is possible to conclude
that the classical regression model with a single constant term is inap-
propriate. We turn therefore to panel estimation methods that may take
into account the specific nature of the countries.
The fixed effect model (FEM) lets the intercept vary for each country
by adding dummy variables that take into account country-specific
effects. In the random effect model (REM), differences across countries
are captured through a disturbance term <%, which follows coit = st + uitf
where e, is an unobservable term that represents the individual specific
error component, and uit is the combined time series and cross-section
error component. The REM assumes that e, is not correlated to any
explanatory variable in the equation.
In order to choose between the FEM and the REM, we apply the
Hausman test for specification (1978). The null hypothesis underlying
this test is that the regressors and the unobservable individual specific
random error are uncorrelated. If the test statistic, based on an asymp-
totic x2 distribution, rejects the null hypothesis, then the random effect

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
160 Economic Liberalization and Income Distribution

estimators are biased and the fixed effect model is preferred. The result
of the Hausman test from equation (7.1) suggests that the REM estimates
are inconsistent and the FEM would be more appropriate.3 Since the LM
test suggests that there are individual effects, and the Hausman test
emphasizes that these effects are correlated with the other variables in

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
the model, it is possible to conclude that of the alternatives previously
considered, the FEM is the better choice. Column 2 in Table 7.1 shows
the results from this model.
Before adopting the FEM as the final estimation procedure, it is impor-
tant to conduct an additional test in equation (7.1). It has been already
contended that the error term uit is assumed to satisfy white-noise
assumptions, that is zero mean, constant variance o2, and serially uncor-
related, which is denoted as M/t~I.I.D.(0, o2) (The error term is inde-
pendently and identically distributed with zero mean and constant
variance). By the same token an AR(1) (autoregressive process of order
one) test should be available. In the presence of autocorrelation, both a2
and the standard errors are likely to be underestimated and biased,
which leads to misleading conclusions about the statistical significance
of the estimated regression coefficients.4 The test for first-order serial
autocorrelation is not satisfied, as it rejects the null: no evidence of
AR(1) or p = 0.5 The P value of the test is also presented in column 2,
Table 7.1. To deal with this problem it is essential to explore the possi-
bility that autocorrelation may arise due to model misspecification; to
be precise, because of omitted lagged dependent variables. In this con-
text, equation (7.1) is extended and transformed into a dynamic panel
data model (DPDM) by adding a lagged dependent variable as follows:

EHIIit = at + yEHIIit^p{TFtAGDPit + p2FDIGDPit + p3INFLit


+ p4EDUSECit + rit + uit (7.2)

However, the inclusion of a lagged dependent variable introduces a


source of persistence over time: correlation between the right hand
regressor yit_x and the error term uit. Furthermore, DPDMs are character-
ized by individual effects r\{ caused by heterogeneity among the individ-
uals.6 As a consequence, it is necessary to adopt different estimation and
testing procedures for this model.

The sys-GMM method


In order to estimate the model, we use a system generalized method
of moment estimation (sys-GMM) for DPDMs, initially proposed by
Arellano and Bover (1995). Firstly, the estimation method eliminates

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Gerardo Angeles-Castro 161

country-effects (17,) by expressing equation (7.2) in first differences as


follows:

EHIIit - EHIIit_x = y(EHIIit^ - EHII^2) + pk(Xit - Xit_,) + (uit - uit_x)


(7.3)

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
where X is the set of explanatory variables outlined earlier. On the basis
of the following standard moment condition:

E(EHIIif^sAuit) =0 for t = 3,...,2V and s > 2

That is, lagged levels of EHIIit are uncorrelated with the error term in
first difference, the method uses lagged endogenous variables as instru-
ments to control for likely endogeneity of the lagged dependent vari-
able, reflected in the correlation between this variable and the error term
in the transformed equation. The resulting GMM estimator is known as
the difference estimator.
Blundell and Bond (1998) contended that the GMM estimator
obtained after first differencing has been found to have large finite sam-
ple bias and poor precision. They attribute the bias and poor precision of
this estimator to the problem of weak instruments, as they assert that
lagged levels of the series provide weak instruments for the first differ-
ence (Blundel and Bond 1998: 115-16). So as to improve the properties
of the standard first-differenced GMM estimator, they justified the use
of an extended GMM estimator, on the basis of the following moment
condition:

ElAEHII^i^ + uit)] = 0

That is, there is no correlation between lagged differences of EHIIit and


the country-specific effect. The method therefore uses lagged differences
of EHIIit (the endogenous variable) as instruments for equations in
levels, in addition to lagged levels of yit as instruments for equations in
first differences. The extended sys-GMM encompasses a regression equa-
tion in both differences and levels, each one with its specific set of
instrumental variables. This type of estimation, called system estimator,
not only improves precision but also reduces finite sample bias.7
The method assumes that the disturbances uit are not serially
correlated. If this were the case, there should be evidence of first-order
serial correlation in differenced residuals (that is, uit— uit-^ and no evi-
dence of second-order serial correlation in the differenced residuals

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
162 Economic Liberalization and Income Distribution

(Doornik et al. 2002: 5-8). It is an important assumption because


the consistency of the GMM estimators hinges upon the fact that
E[AuitAuit_ 2] = 0. Accordingly, tests of autocorrelation up to order 2 in
the first-differenced residuals should be available.
The results of the sys-GMM regression are reported in column 3,

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Table 7.1. The tests of serial correlation in the first-differenced residuals
are in both cases consistent with the maintained assumption of no serial
correlation in uit. The AR(2) test fails to reject the null hypothesis that
the first-differenced error term is not second-order serially correlated,
whereas by construction, the AR(1) tests rejects the null that this process
does not exhibit first-order serial correlation.8
In order to assess the validity of the instruments, a Sargan test of overi-
dentifying restrictions, proposed by Arellano and Bond (1991), is also
reported. Under the null hypothesis that the instruments are not corre-
lated with the error process, the Sargan test is asymptotically distributed
as a chi-square with as many degrees of freedom as overidentifying
restrictions. In this case, the test is unable to reject the validity of the
instruments. 9

Results

Original equation
The results are illustrated in Table 7.2. Column 1 shows the results
initially obtained by regressing equation (7.3) with the whole sample.
Subsequently, in order to assess the effect of economic liberalization on
income distribution, under different scenarios of governance and
macroeconomic stability, we conducted regressions with four different
sub-samples. The first two sub-samples comprise countries with low and
high governance and their results are illustrated in columns 2 and 3
respectively. The last two sub-samples contain countries with high and
low standard deviation of inflation, and their results are shown in
columns 4 and 5 respectively. All the equations are regressed using the
sys-GMM procedure. Bearing in mind that a positive sign in the corre-
sponding coefficient indicates a worsening in the distribution of
income, the results yield the following conclusions.

1 Trade. The overall sample indicates that trade reduces income inequal-
ity. At first glance this finding is in keeping with the expectations sup-
porting trade liberalization and is also consistent with other studies
(Calderon and Chong 2001). However, if we look at the magnitude of the
coefficients, we find that the impact is relatively low. A 37 units increase

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Gerardo Angeles-Castro 163

Table 7.2 Scenarios


Whole Low High High Low
sample governance governance inflation SD inflation SD
(D (2) (3) (4) (5)

Short-run parameters

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
EHII,-! 0.841* 0.684* 0.942* 0.775* 0.899*
TRAGDP -0.004** -0.003 -0.001* -0.012** 0.000*
FDIGDP 0.094** 0.140* 0.021* 0.203* 0.032*
INFL 0.000* 0.000* 0.000* 0.000* 0.001
EDUSEC -0.016* -0.020* -0.002* -0.021* -0.007*
Constant 7.306* 13.886* 2.449* 10.698* 4.332*
Long-run parameters
TRAGDP -0.027* -0.009 -0.017* -0.055 -0.005*
FDIGDP 0.594* 0.442* 0.359* 0.900* 0.317*
INFL 0.001* 0.001* 0.003* 0.001* 0.012
EDUSEC -0.102* -0.065* -0.039* -0.094* -0.072*
Constant 45.959* 43.876* 42.129* 47.532* 42.707*
Adjustment coefficient 0.159 0.316 0.058 0.225 0.101
Observations 1,209 546 663 517 692
Countries 93 47 46 44 49
Sargan test [0.247] [0.617] [0.929] [0.923] [0.565]
AR(1) test [0.000] [0.007] [0.001] [0.007] [0.002]
AR(2) test [0.074] [0.070] [0.385] [0.124] [0.159]

Notes: Dependent variable: EHII, Sargan and serial correlation test are P values, * significant
at 5%, ** significant at 10%, all equations use the sys-GMM estimation method.

in the rate of trade leads to a long-run decline of 1 point in the income


distribution indicator ((0.004/(1-0.841))*37).10 Trade is deemed the
cornerstone supporting distributional effects in the process of economic
liberalization (Bulmer-Thomas 1996: 10). In this sense, we should expect
a larger benefit from trade on income distribution but the weak evidence
above is not supportive of this assumption.
It should also be emphasized that trade is able to exert a long-term
positive impact on income distribution under conditions of macroeco-
nomic balance and high governance. On the other hand, the rate of
trade as a percentage of GDP is not significant in countries with macro-
economic distortions and low governance. Thus, the outcome indicates
that in the presence of macroeconomic disequilibria and failure of good
governance, trade does not exert benefits on income distribution.
To some extent this finding in particular is in accordance with the

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
164 Economic Liberalization and Income Distribution

neoliberal assertion that the overall efficiency within a country is an


essential factor for the appropriate operation of trade policies.
2 Investment. The outcome of the overall regression suggests that FDI
has an adverse effect on income distribution. An upturn of 1.7 points on
the rate of investment as a percentage of GDP raises the inequality indi-

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
cator by 1 point over the long run ((0.094/(1- 0.841))*1.7). This result
undermines the assumptions and expectations outlined in the neolib-
eral postulates.
In addition, Table 7.2 indicates that the effect of FDI is adverse under any
scenario. However, if we compare the coefficients across columns 2 to 5, it
is possible to observe that this effect is worse on countries with macroeco-
nomic disequilibria and low governance. For example, a 1.7 point upturn
on the rate of investment as a percentage of GDP raises the inequality indi-
cator over the long run by 0.75 points ((0.14)/(1- 0.684)*1.7) and by 0.61
points ((0.021)/(1- 0.942)*1.7) in low and high-governance countries
respectively, whereas the effects on countries with high-inflation and
low-inflation standard deviation is 1.53 points ((0.203)/(l- 0.775)*1.7)
and 0.54 points ((0.032)/(l-0.899)*1.7) respectively. Consequently,
inefficiency within countries seems to be a factor that accentuates the
adverse effect of FDI on income distribution.
3 Inflation. Macroeconomic imbalance is represented by the infla-
tion variable. Results from the overall regression reveal that this variable
is a determinant of income inequality. This finding is in accordance
with the orthodoxy supporting liberalization and stabilization policies,
although the magnitude of the effect is low. An increase in inflation of
930 points, which could be considered an episode of hyperinflation,11 is
required to raise inequality by 1 point ((0.0002/(1- 0.841))*930). Hence,
inflation has a negative effect on the distribution of income, but this
effect does not seem to be large.
Table 7.2 reveals that inflation has adverse effects on countries with
low governance and macroeconomic distortions. It is worth noting that
the repercussions of inflation also affect the distribution of income in
countries with high governance. Not surprisingly, this is not significant
in column 5; this fact indicates that the inflation rate is a variable that
does not affect income distribution in countries that have traditionally
kept a stable economy.
4 Education. Our results confirm that education reduces income
inequality. Similar conclusions have also been obtained in previous
studies (Chong and Calderon 2000; De Gregorio and Lee 2002). As for
the overall sample, the education coefficient implies that an increase of
10 points in gross secondary-school enrolment is required to reduce the

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Gerardo Angeles-Castro 165

inequality indicator by 1 point over the long-run ((0.016/(1- 0.841))*10).


Moreover, education exerts a positive impact under any scenario. Hence,
human capital formation can mitigate the negative effect that may be
caused by economic liberalization on income distribution and it is in
keeping with the postulates of the PWC.12

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
5 The adjustment coefficient. The outcome of the regression illustrates
that the adjustment coefficient is larger for those countries that exhibit
macroeconomic mismanagement or a low level of governance. This fact
suggests that these countries are more vulnerable to the effect of the
variables involved in the economic liberalization process, because their
distribution of income adjusts faster to the long-term level, or changes
faster compared to those countries with more domestic efficiency.

Applying changes in trade volume


Some authors contend that trade volume is a variable that reflects coun-
tries' geographical characteristics such as their proximity to major mar-
kets, their size, or whether they are landlocked. As a consequence, this
variable may tell us little about the effect of trade on growth or income
distribution (Dollar and Kraay 2004: 26). With the above in mind,
changes in trade volume is a variable that may eliminate geography or
any other unobserved country characteristic. On the other hand, trade
volume is a variable applied frequently in the empirical literature,
including those studies exploring the relationship between economic
liberalization and income distribution (Calderon and Chong 2001).
In addition, this variable can be more effective in panel data studies,
because they also consider variations over time and not only variations
across countries. Finally, trade volume improves its explanatory power
when it is applied in first difference estimations, such as GMM methods,
because in this way unobserved country characteristics are eliminated.
So as to test if our results can vary depending on the applied variable on
trade liberalization, we regress equation (7.3), replacing trade volume with
changes in trade volume. The outcome of this exercise, including both the
overall sample and the four sub-samples, is illustrated in Table 7.3.
It is worth nothing that the long-term coefficient of the variable
changes in trade volume is not significant in the overall sample. This
outcome suggests that there is no significant long-term relationship
between inequality and changes in trade volume. This finding is consis-
tent with the recent study by Dollar and Kraay (2004). Furthermore,
changes in trade volume is not significant in the sub-samples except in
column 5 where the variable enters negatively and at significant levels.
Consequently, this result illustrates that countries with macroeconomic

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
166 Economic Liberalization and Income Distribution

Table 7.3 Scenarios using changes in trade volume (CTRAGDP)


Whole Low High High Low
sample governance governance inflation SD inflation SD
d) (2) (3) (4) (5)

Short-run parameters

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
EHIIt-! 0.845* 0.677* 0.956* 0.754* 0.900*
CTRAGDP -0.001 -0.003 0.000 -0.003 -0.003*
FDIGDP 0.060* 0.129* 0.006 0.170* 0.026*
INFL 0.000* 0.000* 0.000* 0.000* 0.003
EDUSEC -0.016* -0.021* -0.001 -0.025* -0.007*
Constant 6.889* 14.081* 1.747* 11.060* 4.215*
Long-run parameters
CTRAGDP -0.007 -0.010 -0.007 -0.013 -0.028*
FDIGDP 0.388* 0.398* 0.147 0.688* 0.258*
INFL 0.001* 0.001* 0.004* 0.001* 0.035
EDUSEC -0.101* -0.066* -0.019 -0.101* -0.071*
Constant 44.467* 43.532* 40.015* 44.908* 42.247*
Adjustment coefficient 0.155 0.323 0.044 0.246 0.100
Observations 1,207 546 661 516 691
Countries 93 47 46 44 49
Sargan test [0.177] [0.650] [0.914] [0.816] [0.557]
AR(1) test [0.000] [0.007] [0.001] [0.007] [0.002]
AR(2) test [0.070] [0.074] [0.372] [0.117] [0.156]

Notes: Dependent variable: EHII, Sargan and serial correlation test are P values, * significant
at 5%, all equations use the sys-GMM estimation method.

stability can improve income distribution owing to the expansion of


trade. A 36 per cent increase in trade volume leads to a long-term decline
of 1 point in income inequality ((0.003/(1- 0.90))*36). As for the remain-
ing three variables, they keep the same sign as the results in Table 7.2,
across all the columns, while the magnitude of their coefficients does
not differ substantially.
Only in the case of high-governance countries in Table 7.3, column 3,
there are two variations that deserve highlighting. Firstly, the secondary-
school enrolment variable is not significant. If we consider that high-
governance countries are associated with high levels of secondary-school
enrolment and educational attainment, it is likely that an upturn in this
variable does not impact considerably on income distribution. As a
matter of fact, Table 7.2 shows that the coefficient of secondary-school
enrolment is significant in any scenario, but in the high-governance sub-
sample it is the smallest compared with the corresponding coefficients in

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Gerardo A ngeles-C astro 167

the other sub-samples. In this case, we suggest that higher levels of


education can provide better benefits to income distribution. Secondly
the coefficient of the ratio of FDI to GDP is not significant either. Thus,
when we consider changes in trade volume, FDI does not seem to exert
an adverse effect on income distribution in this kind of economy.

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
In summary, the effect of trade on income distribution seems to be
weak and not robust, because changes in trade volume do not have a sig-
nificant relationship with income distribution in the overall sample and,
of the four sub-samples, only those countries associated with macroeco-
nomic stability can obtain benefits from the expansion of trade, but the
effect is weak as Table 7.3 illustrates. In addition, trade volume does not
have a significant impact on countries that exhibit domestic inefficiency,
and only has a small impact on the remaining scenarios in Table 7.2.
Nevertheless, there is some statistical evidence, emerging from the two
variables on trade, which illustrates that countries with domestic effi-
ciency are more likely to improve income distribution on account of an
upturn in trade.
The conclusions for all other variables practically do not change,
except for the ratio of FDI to GDP and secondary school education,
in high governance countries, as outlined earlier.

The effects of exports by sector


In the economic liberalization process, trade openness plays a prepon-
derant role, as it is assumed to give an unambiguous boost to the
exportable sector and therefore to export-led growth. In this context,
export growth may raise employment directly in the exportable sector
and indirectly by permitting faster GDP growth. This process is expected
to have positive implications for income distribution and longer-term
growth.
We have already contended that the basis for tracing the income-
distribution effects of trade is the Stolper-Samuelson theorem. According
to this neoclassical model, foreign trade increases the income of the fac-
tor which is relatively most used in the export sector and which on the
principle of comparative advantage is most abundant, this factor
being conventionally assumed to be unskilled labour in developing
economies; so income distribution improves (FitzGerald 1996: 32).
In this context, we may expect that primary exports, based on natural
resources and unskilled labour, provide larger benefits to income distri-
bution than manufactured exports in those countries which exhibit low
governance and high-inflation standard deviation, since these countries
are associated with lower levels of development.13

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
168 Economic Liberalization and Income Distribution

Table 7.4 Exports by sector to total exports


Whole Low High High Low
sample governance governance inflation SD inflation SD
(D (2) (3) (4) (5)

Short-run parameters

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
EHIIf_! 0.885* 0.736* 0.925* 0.860* 0.879*
MANEXP -0.013* -0.011* -0.006* -0.012** -0.011**
PRIEXP 0.001 0.009 0.003 0.021 -0.010
Constant 5.039* 11.107* 3.116* 6.039* 5.151*
Long-run parameters
MANEXP -0.112* -0.041** -0.077* -0.084** -0.091*
PRIEXP 0.005 0.033 0.039 0.151 -0.081
Constant 43.850* 42.114* 41.693* 43.154* 42.717*
Adjustment coefficient 0.115 0.264 0.075 0.140 0.121
Observations 1,177 478 699 459 718
Countries 88 43 45 39 49
Sargan test [0.546] [0.757] [0.769] [0.883] [0.556]
AR(1) test [0.000] [0.017] [0.001] [0.024] [0.001]
AR(2) test [0.080] [0.214] [0.193] [0.373] [0.075]

Notes: Dependent variable: EHII, Sargan and serial correlation test are P values, * significant
at 5%, ** significant at 10%, all equations use the sys-GMM estimation method.

In order to test the effect of manufactured exports and primary


exports on income distribution we conduct a regression including two
variables representing the percentage of manufactured exports and pri-
mary exports to merchandise exports. Table 7.4 illustrates that the proxy
of manufactured exports is negative and statistically significant in the
overall sample and in the four sub-samples. An upturn of 8.9 points in
the rate of manufactured exports to merchandise exports is linked with
a long-run decline of 1 point in income inequality in the overall sample
(0.013/(1-0.885)*8.9). On the other hand, the proxy of primary
exports is not significant in any case, even in the low governance and
high inflation variance sub-samples. To some extent, this outcome
undermines theoretical pillars supporting the economic liberalization
process.

The role of employment and government size


It has already been noticed that employment is deemed to be one of the
main factors that can drive a better distribution of income over the eco-
nomic liberalization process. So as to test this assumption, we extend

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Gerardo Angeles-Castro 169

the original equation by adding a proxy for the level of employment.


We include the unemployment rate variable. Results are shown in
Table 7.5, column 2. In addition, so as to test the role of employment by
sector, we include variables of employment by industry and agriculture
to total employment. Results are provided in Table 7.5, columns 3 and 4

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
respectively. Finally, we undertake an additional exercise to explore the

Table 7.5 Level of employment (UNEMP), employment by sector (EMPSEC) and


government size (GOVEXPEN)
Employment by sector
Unem- to total employment
Whole ployment Govern-
sample rate Industry Agriculture ment size
(D (2) (3) (4) (5)

Short-run parameters
EHnVi 0.841* 0.904* 0.982* 0.981* 0.932*
TRAGDP -0.004** 0.000 -0.001* -0.001* -0.001*
FDIGDP 0.094** 0.023* 0.006 0.009 0.024*
INFL 0.000* 0.001* 0.002* 0.002* 0.000*
EDUSEC -0.016* -0.009* 0.001 0.000 -0.004*
UNEMP 0.020*
EMPIND -0.007*
EMPAGR 0.000
GOVEXPEN -0.004*
Constant 7.306* 4.219* 0.947* 0.804** 3.174*
Long-run parameters
TRAGDP -0.027* -0.002 -0.031** -0.032** -0.021*
FDIGDP 0.594* 0.236* 0.356 0.476** 0.347*
INFL 0.001* 0.010* 0.100* 0.089* 0.005*
EDUSEC -0.102* -0.096* 0.030 0.014 -0.064*
UNEMP 0.212*
EMPIND -0.417*
EMPAGR 0.000
GOVEXPEN -0.064*
Constant 45.959* 44.019* 53.995* 43.109* 46.347*
Adjustment coefficient 0.159 0.096 0.018 0.019 0.068
Observations 1,209 722 690 679 974
Countries 93 59 58 57 73
Sargan test: [0.247] [0.437] [0.636] [0.523] [0.189]
AR(1) test [0.000] [0.000] [0.000] [0.000] [0.000]
AR(2) test: [0.074] [0.438] [0.864] [0.940] [0.472]

Notes: Dependent variable: EHII, Sargan and serial correlation test are P values, * significant
at 5%, ** significant at 10%, all equations use the sys-GMM estimation method.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
170 Economic Liberalization and Income Distribution

role of the state and its empowerment in the distribution of income. For
this purpose, we use the government expenditure to GDP variable.
Results are provided in Table 7.5, column 5. In all the equations, the
exercise is conducted only for the whole sample because of constraints
on data availability. That is, after including the unemployment,

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
employment by sector and government expenditure variables, the
number of observations drops.
Firstly, we observe that unemployment enters positively and signifi-
cantly in the equation, indicating that higher levels of employment are
associated with less inequality. A 4.75 points reduction of unemploy-
ment drops the inequality indicator by 1 point over the long run
((0.020)/(l- 0.904)*4.75). What is striking is that the trade volume does
not remain significant. However, this result is not owing to the inclusion
of unemployment in the equation but rather on account of the reduction
of observations in the sample. As a matter of fact, when we conduct the
regression (not reported) with the same sample comprising 781 observa-
tions and dropping the unemployment variable it is worth noting that
the trade volume is neither significant.
When we explore the effect of employment by sector on income dis-
tribution it is interesting to note that the employment in industry to
total employment variable is significant and negative. An increase of
2.4 points in this ratio is required to reduce inequality by one point over
the long-term ((0.007/1- 0.982)*2.4). It should also be added that the
variables FDI to GDP and secondary-school enrolment are no longer sig-
nificant in this equation. However, as in the case of the unemployment
variable this outcome is not due to the inclusion of an additional vari-
able in the equation, in this case employment in industry. This outcome
is rather caused by the reduction of the sample. On the contrary, when
the regression is conducted (not reported) with the same sample (690
observations) but dropping the employment in industry variable both
FDI to GDP and secondary-school enrolment are not significant either.
On the other hand, what is striking is that the employment in agriculture
variable is not significant.
In this sense, results in Tables 7.4 and 7.5 are supportive and comple-
ment each other. In Table 7.4 we show statistical evidence suggesting
that the growth of manufactured exports decreases inequality, while in
Table 7.5 we illustrate that employment in industry is associated with
improvements in income distribution. In contrast, from the two previ-
ous tables there is no statistical evidence suggesting that primary exports
and employment in agriculture can reduce inequality, even in those
countries associated with lower levels of development and abundant

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Gerardo Angeles-Castro 171

unskilled labour such as countries with low governance and high infla-
tion standard deviation. Consequently, results in Table 7.4 and Table 7.5
columns 3 and 4 undermine the theoretical foundations supporting the
economic liberalization process.
As for the proxy for government size, it enters negatively and signifi-

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
cantly, while the other variables remain significant. An upturn of 15 points
on government expenditure drops the inequality indicator by 1 point
over the long run ((0.0044/(1- 0.932))*15). Hence, the reempowerment
of the state is a factor that can improve income distribution. This finding
is in keeping with the prescription advocated by the PWC.

Concluding remarks

From the statistical evidence presented above, we find that a strong


state, associated with higher levels of government expenditure, is
important to reduce inequality. We also find that countries with macro-
economic stability and high governance can mitigate the adverse effect
of FDI on income distribution, while there is some evidence that they
can obtain weak benefits from trade, in terms of income distribution.
In contrast, countries that exhibit domestic inefficiency do not benefit
from trade and the effect of FDI on their distribution of income is worse.
Inflation, which can be a proxy of fiscal discipline, worsens inequality,
but this effect is weak and seems to exert a real impact on income distri-
bution only in episodes of hyperinflation. Not surprisingly, this variable
is not statistically significant in the sample comprising countries with a
stable economy. Furthermore, education socializes the operation of
market forces, since it reduces inequality.
In short, domestic efficiency, the empowerment of the state, and the
set of second generation policies, represented in this study by a proxy of
human capital formation, can mitigate the adverse effect of economic
liberalization on income distribution, and to some extent can reduce
inequality. Consequently, the PWC and second generation reforms rep-
resent an improvement in the global process of economic liberalization.
In this study we show that higher employment reduces inequality.
However, exploring the effects of employment and exports by sector it is
found that emphasis on the primary sector is not a strategy that
improves income distribution, even in those countries that to some
extent can be associated with comparative advantages based on natural
resources and unskilled labour. On the other hand the expansion of
exports and employment through emphasis on the industrial sector can
have better consequences for income distribution.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
172 Economic Liberalization and Income Distribution

Therefore, the effects of exports and employment by sector on income


distribution illustrated in this study are in keeping with what might be
called the 'new Keynesian approach'. According to FitzGerald this
approach stresses that those attempts to exploit comparative advantages
based on natural resources and unskilled labour do not form an appro-

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
priate basis for sustained export growth, since successful export growth
depends on technological innovation and the improvement of skills in
the labour force; furthermore, from this perspective such an alternative
strategy has better consequences for income distribution than the neolib-
eral model. In this concern FitzGerald points out that from the new
Keynesian view industrialization can raise productivity and hence
labour incomes throughout the economy. Moreover, manufactured
products require greater investment in human capital formation which
also leads toward higher labour incomes. In addition, manufactured
exports involve higher rates of both public and private investment. This
set of events strengthens domestic markets and forms the basis for sus-
tained economic growth. He also explains that from this critical view-
point reliance on natural resources and unskilled labour does not solve
the problems of depressed real wages and slow rate of growth, which
leads to weak domestic markets, recessive fiscal retrenchment and struc-
tural unemployment, resulting in a worsening of income dispersion
and absolute poverty. In contrast, with an appropriate industrialization
wage cuts are unnecessary, domestic markets are stronger, faster growth
increases employment and welfare expenditure and eventually inequality
declines (FitzGerald 1996: 34-5).14
The effect of FDI on income distribution tends to be adverse and it
does not provide any distributional effect. This finding undermines
the neoclassical postulate which holds that the international flow of
investments increases efficiency in resource allocation between and
within countries. This assertion faces opposite arguments in the litera-
ture that may be generally supportive of the results that emerge from
this study in terms of FDI. Some of these arguments are summarized as
follows.
The race to attract new inward investment and/or to retain multina-
tional corporations (MNCs) results in subsidy packages, downward pres-
sure on corporation and income taxes, and in general in tax incentives
and tax reductions. Such tendencies have three major adverse conse-
quences. Fiscal policies designed specifically to serve transnationals'
interests lead to an evaporating tax base that constrains the scope
for redistributive and social expenditure by governments (Bailey et al.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Gerardo Angeles-Castro 173

1998: 296). In addition, the perceived capacity of MNCs to shift produc-


tion might reduce countries' ability to tax capital, reduces their tax base
and increasingly moves the burden of taxation on the less mobile factors such
as labour (Held et al. 1999: 277). Finally, preferential tax treatment and
other incentives to induce inward FDI may place domestic industry at a

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
disadvantage and may also introduce a distortion affecting domestic invest-
ment. Such distortion between the return to foreign and domestic capital
could have a strongly negative effect on growth and employment
(Easterly 1993).
Thus, the corporate and capital tax competition between nations that
has emerged on account of capital liberalization and privatization
is likely to result in declining social expenditure because of a reduction
in the tax base, or a rising tax burden on the less mobile factors. As a
matter of fact, Held et al. (1999) notice that corporate tax rates among
countries have tended to fall since the early 1980s. This set of character-
istics, in which FDI operates, can be a justification of our results.
The increasing bargaining power of MNCs is another possible cause of
inequality. Privatization of state-owned firms and liberalization of FDI
encourage a surge of mergers and acquisitions across borders that tend
to create dominant positions and oligopolistic markets. This likely pat-
tern decreases the market power of small and medium-sized enterprises
(SMEs) and leads to a deterioration of the domestic industry and to
capital concentration. 15 In addition, the ability of MNCs to organize
production transnationally and to shift production to reap the benefits
of low wages increases corporate power relative to the power of labour,
putting downward pressure on wages and working conditions. In this
respect the globalization of production may contribute to widening
wage differentials between skilled and unskilled workers within and
between countries.16 Hence, the fact that the balance of power between
multinational capital and social actors (the state, SMEs and labour,
among others), under conditions of liberalization, may shift in favour of
the former, represents an explanation of the adverse effect of FDI on
income distribution.
The operation of MNCs impacts on the effectiveness of the traditional
tools of macroeconomic management and government economic policy
in several ways. Held et al. (1999: 276-7) most especially underline two
ways. The effectiveness of national monetary policy may be compro-
mised since MNCs can borrow abroad when domestic interest rates are
high, and conversely take advantages of low domestic interest rates to
borrow to fund project overseas. In addition, MNCs may also play an

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
174 Economic Liberalization and Income Distribution

important role in exchange rate markets. In this sense, although specu-


lators may initiate an attack on a currency, it is when both MNCs and
institutional investors shift out of that currency, even as a precautionary
measure, that pressure on the exchange rate can become irreversible. As
a consequence, if the monetary and exchange rate policy are aimed at

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
supporting an efficient allocation of resources and redistributive actions,
the apparent erosion in the effectiveness of these policies adversely
affect the distribution of income.
Arguments such as tax competition between nations, the increasing
bargaining power of MNCs, and the erosion of national macroeconomic
policy instruments are neglected in the neoliberal thesis, and provide a
different perspective to explain the adverse effect on income distribu-
tion that may arise owing to the flow of FDI.
As for the variables on trade, we find that there is a weak benefit of
the expansion of trade on income distribution. On the other hand,
we have pointed out that international trade is traditionally considered
the corner-stone supporting distributional effect in the process of
economic liberalization (Bulmer-Thomas 1996:10). Therefore, we should
expect a larger benefit from this variable on income distribution, but the
weak evidence above is not supportive of this assumption.
Consequently, the empirical evidence obtained from this study is in
keeping with the assumptions and expectations supporting the set of first-
generation reforms to the extent that employment and low inflation
can benefit income distribution. In contrast, the results undermine these
assumptions and expectations because the benefit of trade on income dis-
tribution is weak and FDI worsens inequality. In addition, an export-led
growth strategy and the expansion of employment based on the primary
sector do not improve income distribution.
Although second-generation reforms represent an improvement in
the economic liberalization process, the empirical evidence in this study
suggests that the role of the state is not enough to socialize the opera-
tion of trade and investment. In this context, even under conditions of
macroeconomic stability and high governance FDI does not benefit
income distribution; in contrast, it seems to be adverse. Moreover, the
beneficial effect of trade on income distribution is weak in those coun-
tries with domestic efficiency. For these reasons, we argue that further
supranational mechanisms, beyond the scope of the state, are required
to obtain distributional effects from trade liberalization and from the
flow of FDI.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
175

Appendix
Table A7.1 Characteristics of data-sets on income inequality

Data-set Global coverage Assembled from

World Bank data set » Observations: 682 Gini coefficients reported in

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
on Gini coefficients 'high-quality' the literature
(Deininger and • Countries: 108
Squire 1996) • Obs/country avg: 6.31
* Period: concentrated
between 1960 and
early 1990s
WIID (world income • Obs: 603 gross income, Gini coefficients from D&S
inequality data-base) household/family, national (1996), LIS, Central Statistical
(UNU/WIDER) sample) Offices, UNICEF, and
• Countries: 105 research studies
* Obs/country avg: 5.74
• Period: concentrated
between 1960 and
early 1990s
Luxembourg Income • Observations: 132 Micro-level data and primary
Study (LIS) • Countries: 29 data sets from household
• Obs/country avg: 4.55 income surveys
• Period: 1969-2000
UTIP-UNIDO Theil index • Observations: 3,200 Industrial statistics database
(University of Texas • Countries: 153 by UNIDO. The data set is
inequality project) • Obs/country avg: 20.91 the Theil index of
• Period: 1963-99 manufacturing pay inequality
EHII (estimated • Observations: 3,126 Constructed from the
household • Countries: 153 relationship between the
Income Inequality • Obs/country avg: 20.43 UTIP-UNIDO Theil and the
(UTIP) • Period: 1963-99 Gini coefficients in D&S
(1996) plus a limited a m o u n t
of additional information

Sources: For Deininger and Squire (1996): World Bank (2003). For WIID: United Nations University, World
Institute for Development Economics Research (2003). For LIS: Luxemburg Income Study (2003). For
UTIP-UNIDO: University of Texas Inequality Project (2003). For EHII: University of Texas Inequality Project
(2003a).

Table A7.2 List of countries (93)

Countries

Algeria Greece Nigeria


Argentina Guatemala Norway
Australia Guinea Pakistan
Austria Haiti Panama
Bangladesh Honduras Peru

Continued

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
176 Economic Liberalization and Income Distribution

Table A7.2 Continued

Countries

Barbados Hong Kong Philippines


Belgium Hungary Poland
Bolivia Iceland Portugal

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Botswana India Russia
Brazil Indonesia Senegal
Bulgaria Iran Singapore
Burundi Ireland Slovenia
Cameroon Israel South Africa
Canada Italy Spain
Central African Republic Jamaica Sri Lanka
Chile Japan Suriname
Colombia Jordan Swaziland
Costa Rica Kenya Sweden
Cote D'lvoire Korea Syrian Arab Republic
Cyprus Lesotho Thailand
Denmark Malawi Togo
Dominican Republic Malaysia Trinidad and Tobago
Ecuador Malta Tunisia
Egypt Mauritius Turkey
El Salvador Mexico Uganda
Ethiopia Moroco Ukraine
Fiji Mozambique United Kingdom
Finland Nepal United States
France Netherlands Uruguay
Gabon New Zealand Venezuela
Ghana Nicaragua Zimbabwe

Note: The whole list of countries is used in the regressions described in Tables 7.1 to 7.3. In the rest of the
regression the number of countries drops.

Notes
1 John Williamson (1990) is given credit for first labelling as Washington
Consensus the package of policies that multilateral institutions endorsed in
trade and loan negotiations during the 1980s. This package of policies insisted
on unregulated markets and a reduced role of the governments in economic
activity.
2 We obtain a Lagrange Multiplier test statistic of 4,466, which far exceeds the
5 per cent critical value of x2 with one degree of freedom, 3.84. Since the null
hypothesis is rejected, it is concluded that there are individual effects.
3 The value of the Hausman test statistic is 130.65 with a negligible P value;
hence, the test rejects the null hypothesis. In this case, the key assumption of
the REM 'the unobservable individual specific error st is not correlated to any
explanatory variable' is violated; thus, the FEM is preferred.
4 For an empirical application see Galbraith and Kum (2002: 12).
5 The AR test statistic of order one is equal to 29.99 and the P value is negligi-
ble. Hence, the AR(1) test suggests evidence of first-order autocorrelation.
6 For an elaboration in this point see Baltagi (2001: 129-30).

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Gerardo Angeles-Castro 177

7 A discussion of the method through an empirical application can be seen in


Calderon and Chong (2001: 226-7).
8 The P value in the test statistic for second-order serial correlation based on
residuals from the first-difference equation is equal to 0.074. In this case we
can not reject the null hypothesis if we establish a 95% confidence interval.
Even in these circumstances the sys-GMM model represents a substantial

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
improvement in terms of AR compared with ordinary methods. The P value
in the test statistic for first-order serial correlation is negligible; in this case
the null is rejected at any conventional level of significance.
9 The Sargan test statistic is equal to 55.38 with a P value equal to 0.247. Thus,
the test is unable to reject the hypothesis that the instruments are not corre-
lated with the error process.
10 Dollar and Kraay (2004) stress that between the late 1970s and late 1990s the
rate of trade to GDP in OECD countries rose 29 points, that is from 21 to
50 per cent. While trade volume in the top third of developing countries in
terms of this rate expanded 17 points, that is from 16 per cent to 33 per cent.
Accordingly, the effect of trade volume on income distribution in this study
seems to be weak because the expansion of the rate of trade to GDP required
to improve income distribution by one point far exceeds the average increase
in trade volume over the last two decades. Thus, this statistical evidence sug-
gests that the effect of trade on income distribution has not been as large as
was originally expected.
11 The threshold customarily used to define hyperinflation is an increase of
prices averaging 50% a month or 600% a year (Havrylyshyn et al. 1994).
12 As complementary information, we found that a lower rate of population
growth leads towards less inequality, as the corresponding coefficient enters
positively and significantly in the overall sample and the four sub-samples
(regression not reported). In this context, Heerink (1994) shows that income
inequality and population growth reinforce each other, because lower levels
of fertility results in a more equal income distribution; he also underlines the
negative relationship between improvements in nutrition, health and
education on the one hand and the fertility levels on the other. Not
surprisingly, our data base also reveals a negative relationship between
education and the rate of population growth. These findings suggest
that education can contribute directly to improve income distribution, but
also contributes indirectly to this effect by reducing the rate of population
growth.
13 Kaufmann et al. (1999) provide evidence of a strong positive relationship
between governance and better development outcomes. In this sense, coun-
tries with low level of governance are associated with low level of develop-
ment and therefore they can have substantial reliance on natural resources
and unskilled labour.
14 The new Keynesian view evolved partially as a response and as a critique to
the new classical approach. For a review see for example Gordon (1990).
15 For an elaboration of the expansion and challenges of cross-border mergers
and acquisitions see UNCTAD World Investment Report (2000: 15-28).
16 For a discussion about the balance of power among labour and capital see
Held et al. (1999: 278-80).

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
178 Economic Liberalization and Income Distribution

References
Arellano, M. and Bond, S. (1991) 'Some Tests of Specification for Panel Data:
Monte Carlo Evidence and an Application to Employment Equations', The
Review of Economic Studies, 58(2): 277-97.
Arellano, M. and Bover, O. (1995) 'Another Look at the Instrumental Variable
Estimation of Error-Component Models', Journal of Econometrics, 68: 29-51.

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Bailey, D., Harte, G. and Sugden, R. (1998) 'Debate: Transnational Corporations.
The Case for a Monitoring Policy Across Europe' New Political Economy, 3(2):
279-300.
Baltagi, B.H. (2001) Econometric Analysis of Panel Data, 2nd edn (London: John
Wiley & Sons).
Blundell, R. and Bond, S. (1998) 'Initial Conditions and Moment restrictions in
Dynamic Panel Data Models', loumal of Econometrics, 87: 115-43.
Bourguignon, F. (1994) 'Growth, Distribution and Human Resources', in G. Rains
(ed.), In Route to Modern Growth, Essays in Honour of Carlos Diaz-Alejandro
(Washington, DC: Johns Hopkins University Press).
Breusch, T. and Pagan, A. (1980) 'The LM Test and its Applications to Model
Specification in Econometrics', The Review of Economic Studies, 47: 239-54.
Buhner-Thomas, V. (1996) 'Introduction', in V. Buhner-Thomas (ed.), The New
Economic Model in Latin America and its Impact on Income Distribution and Poverty
(London: Palgrave Macmillan).
Calderon, C. and Chong, A. (2001) 'External Sector and Income Inequality in
Interdependent Economies Using a Dynamic Panel Data Approach', Economics
Letters, 71:225-31.
Camdessus, M. (1998) 'Income Distribution and Sustainable Growth: The
Perspective from the IMF at Fifty', in V. Tanzi and K.-Y. Chu (eds), Income
Distribution and High-Quality Growth (Boston, Mass.: MIT Press).
Chong, A. and Calderon, C. (2000) 'Institutional Quality and Income Distribution',
Economic Development and Cultural Change, 48(4): 761-86.
Corden, W.M. (1993) Protection and Liberalization: a Review of the Analytical issues,
IMF Occasional Paper no. 54 (Washington, DC: IMF).
De Gregorio, J. and Lee, J.-W. (2002) 'Education and Income Inequality: New
Evidence from Cross-Country Data', Review of Income and Wealth, 48(3): 395-416.
Deininger, K. and Squire, L. (1996) 'A New Data Set Measuring Income
Inequality', The World Bank Economic Review, 10(3): 565-91.
Deininger, K. and Squire, L. (1998) 'New Ways of Looking at Old Issues:
Inequality and Growth', loumal of Development Economics, 57: 259-87.
Dollar, D. and Kraay, A. (2004) 'Trade, Growth, and Poverty', The Economic
loumal, 114:22-49.
Doornik, J.A., Arellano, M. and Bond, S. (2002) Panel Data Estimation Using DPD
for OX, Centro de Estudios Financieros y Monetarios, http://www.cemfi.es/-
arellano/
Easterly, W. (1993) 'How Much do Distortions Affect Growth', loumal of Monetary
Economics, 32: 187-212.
Edwards, M. (1999) Future Positive: International Cooperation in the 21st Century
(London: Earthscan).
FitzGerald E.V.K. (1996) 'The New Trade Regime: Macroeconomic behaviour and
Income Distribution in Latin America', in V. Bulmer-Thomas (ed.) The

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Gerardo Angeles-Castro 179

New Economic Model in Latin America and its Impact on Income Distribution and
Poverty (London: Palgrave Macmillan).
Forbes, K. (2000) 'A Reassessment of the Relationship Between Inequality and
Growth', American Economic Review, 90: 869-87.
Galbraith, J.K. and Kum, H. (2002) Inequality and Economic Growth: Data Comparison
and Econometric Tests, University of Texas Inequality Project, UTIP Working

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
paper no. 21, http://utip.gov.utexas.edu/web/workingpaper/UTIP21rv.pdf
Galbraith, J.K. and Kum, H. (2003) Estimating the Inequality of Household Incomes:
Filling Gaps and Correcting Errors in Deininger & Squire, University of Texas
Inequality Project, UTIP Working Paper no. 22.
Gilpin, R. (1987) The Political Economy of International Relations (Princeton:
Princeton University Press).
Gordon, R.J. (1990) 'What Is New-Keynesian Economics?', loumal of Economic
Literature, 28(3): 1115-71.
Greene, W.H. (2000) Econometric Analysis, 4th edn (Engle Wood Cliffs, NJ:
Prentice-Hall).
Hausman, J.A. (1978) 'Specification Test in Econometrics', Econometrica, 46:1251-71.
Havrylyshyn, O., Miller, M. and Perraudin, W (1994) 'Deficits, Inflation and the
Political Economy of Ukraine', Economic Policy, 19: 353-401.
Heerink, N. (1994) Population Growth, Income Distribution, and Economic Development:
Theory, Methodology, and Empirical Results (New York and Berlin: Springer).
Held, D. et al. (1999) Global Transformations: Politics, Economics and Culture
(Cambridge: Polity Press).
Higgott, R. (2000) 'Contested Globalization: The Changing Context and
Normative Challenges', Review of International Studies, 26: 131-53.
Higgott, R. and Phillips, N. (2000) 'Challenging Triumphalism and Convergence:
The Limits of Global Liberalization in Asia and Latin America', Review of
International Studies, 26: 359-79.
International Monetary Found (1997) World Economic Outlook: Globalization and
the Opportunities for Developing Countries, (May), chapter 4.
Jha, S. (1996) 'The Kuznets Curve: A Reassessment', World Development, 24(4):
773-80.
Jones, R. and Barry, J. (1988) 'Liberal Political Economy', in R. Jones (ed.), The
Worlds of Political Economy (London: Pinter).
Kaufmann, D., Kraay, A. and Zoido-Lobaton, P. (1999) Governance Matters, World
Bank Policy Research Working Paper no. 2,196, http://www.worldbank.org/
wbi/governance/pubs/govmatters.html
Litwin, C. (1998) Trade and Income Distribution in Developing Countries, Working
Papers in Economics, Goteborg University, 9: 1-42.
Luxembourg Income Study (LIS) (2003) The LIS Database, http://www.lisproject.org/
Milanovic, B. (1995) Poverty, Inequality and Social Policy in Transition Economies,
World Bank, Transition Economics Division, Research Papers 9.
Ortiz, G. (2003) 'Latin America and the Washington Consensus: Overcoming
Reform Fatigue', Finance and Development, 40(3): 14-17.
UNESCO Institute for Statistics (2003) Statistical Tables, http://portal.unesco.org/uis/
ev.php? URLJD = 5275&URL_DO = DO_TOPIC&URL_SECTION = 201
United Nations Conference on Trade and Development (2000) World Investment
Report: Cross-Border Mergers and Acquisitions and Development (New York: United
Nations Publication).

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
180 Economic Liberalization and Income Distribution

United Nations Conference on Trade and Development (2003) Foreign Direct


Investment Database, http://www.unctad.org/Templates/Page.asp? intltemID =
1923&lang; = 1
United Nations University and World Institute for Development Economics
Research (2003) World Income Inequality Data Base, http://www.wider.unu.edu/
wiid/wiid.htm

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
University of Texas Inequality Project (2003) UTIP - UNIDO Database, http://utip.
gov.utexas.edu/
University of Texas Inequality Project (2003a) An Estimated Household Income
Inequality Data Set for the Global Economy, http://utip.gov.utexas.edu
Williamson, J. (1990) 'What Washington Means by Policy Reform', in
J. Williamson (ed.), Latin American Adjustment: How Much Has Happened?
(Washington, DC: Institute for International Economics).
World Bank (2002) World Development Indicators 2002, CD-ROM (Washington, DC:
World Bank), described in http://www.worldbank.org/data/wdi2002/cdrom/
World Bank (2003) Deininger and Squire Data Set, http://www.worldbank.org/
research/growth/dddeisqu.htm

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
8
Pensions and Distribution
in an Ageing Society:

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
A Non-Conventional View*
Sergio Cesaratto

This chapter explores from a Classical point of view the economic


impact of ageing with a particular concern for the future of Social
Security organized along pay-as-you-go (PAYG) lines. Elsewhere I have
shown the shortcomings of the two major pension reforms under dis-
cussion. Cesaratto (2002, 2005: chapter 2, 2006a) pointed out the pros
and cons of the Notional Defined Contribution (NDC) reforms that,
although in principle (and not without delay), provide the financial sta-
bility of PAYG, attain it in the most obvious way of cutting pension ben-
efits, thus leaving aside the social sustainability of PAYG. Cesaratto
(2002, 2005: chapters 3 and 4, 2006b) put forward the various short-
comings of the transition plans aimed at the creation of fully-funded
(FF) schemes. Cesaratto (2005: chapter 6) showed that, according to the
theory of effective demand, the welfare state, and PAYG in particular, are
not necessarily detrimental to economic growth, in fact they may foster
it. I also regarded the welfare state and PAYG as part of a Classical view
of income distribution in which there is no natural distribution setting
associated with factors' full employment as in the Neoclassical theory.
According to this view, it is the political opposition to changes in
income distribution favourable to labour, rather than the Neoclassical
mechanical association between distribution and output, that may
determine a negative influence of social spending on growth.
In this chapter I shall use a Classical-Keynesian approach to explore
the sustainability of an ageing society and of PAYG within it. The

* This chapter is heavily based on chapter 8 of Cesaratto (2005).

181

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
182 Pensions and Distribution in an Ageing Society

concepts of working age and retirement age are eminently cultural, and
only loosely associated, respectively, with physical strength and decline.
In particular, in recent decades retirement and old age have been par-
tially detached from the idea of physical decline, but times are changing
(the OECD has extended the working lifespan from ' 15-64' to '15 and

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
over'). The concept of ageing may instead be considered as a demo-
graphic concept relative to the changing age structure of the population.
Contingent on the retirement of the baby-boom generation and,
more persistently, on the combination of lower fertility rates and higher
longevity, developed and increasingly also intermediate economies are
facing a process of ageing. Conventional wisdom tells us that a forth-
coming labour scarcity plus the ageing process will place the economies
under an intolerable strain, jeopardizing the sustainability of PAYG. We
shall argue that two effects worry the ruling classes: (1) a progressive
shrinking of the industrial reserve army, and (2) an increased tax burden
to sustain the elder generations.
We shall begin in the next section by presenting some demographic
scenarios that put ageing in the context of the possible evolution of
the human population, also allowing it to be considered in a different,
more positive, perspective. This evolution can indeed be interpreted as
the result of the world population approaching a stabilization phase.
Migration flows appear not to be able to reverse the ageing process.
We shall then assess the impact of the demographic developments on
the level of the working-age population, which is the potential labour
supply, and on the proportion of old-age population over the popula-
tion in working-age. The economic impact of ageing, however, cannot
be assessed by mere demographic ratios and we must move from popula-
tion to political economy.1 The later sections of the chapter are thus
mainly devoted to discussing the impact of demographic development
on the labour market, and to exploring the financial versus the real sus-
tainability of PAYG viewed through the lens of the alternative economic
theories.

Stylized demographic trends

Some impressive global scenarios


Both developed and less-developed countries are undergoing major
demographic transformations. Current demographic developments
consist of the continuation of a secular process that started in the most
industrialized countries in the nineteenth century as represented by a
progressive fall in fertility and a lengthening of the average lifespan.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Sergio Cesaratto 183

Declining infant mortality, among other causes, might have represented


the inception of the progressive fall in fertility, which has been the
dominant demographic factor at work so far (Bloom and Canning 2004:
4-10). The lengthening of the expected lifespan, which was quite rapid
in the recent past, is becoming progressively more important on the

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
assumption that, in most regions of the world, fertility stabilizes at the
reproduction level. By definition this complex process determines a pro-
gressive ageing of the population. The baby-boom burst in fertility in
the developed countries after the Second World War appears as an iso-
lated episode within this long period trend, although its effects will be
still felt in the next few decades. Secular demographic transformation is
called 'demographic transition', a process 'in which mortality and then
fertility decline from higher to lower levels' (UN 2002a: 5).
In developed countries the two persistent factors, lower fertility and
increasing longevity, have been at work for many decades. In these
countries the transition of fertility to below the population replacement
level of 2.1 children per woman took place a couple of decades ago (the
fertility rate fell from 2.8 in 1950-55 to 1.5 in 2000-05, cf. UN 2002a: 5).
On top of this, in the first quarter of the century the developed countries
will experience the progressive retirement of the baby-boom generation,
the so-called baby bust. According to the UN Population Division
(2002b), a number of intermediate-fertility developing countries,
roughly those at a relatively high stage of economic development, are
also completing the fertility transition from high fertility rates towards
below-replacement rates. In both cases the initial expectations that the
threshold of the replacement level would not be exceeded proved to be
wrong.
Predicting the future levels and composition of the population is not
so simple. The divergence in the scenarios of the future evolution of the
world population produced by the combination of different conjectures
is striking. The UN experts suggest that these scenarios are not 'projec-
tions' since they represent only 'a few of the many possible future paths
of the world population. The value of the scenarios ... is that they illus-
trate, often dramatically, the implications of small differences in future
fertility levels' (UN 2003: 8).
Before we focus on the effects of the falling fertility rates on ageing
and the working-age population in the more developed countries, let us
first dwell on the effects on total world population trends. Small differ-
ences in the fertility rates of the world population, or of significant
countries or areas - below or above the replacement rates - produce
dramatic divergences in the scenarios. In those prepared by the

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
184 Pensions and Distribution in an Ageing Society

UN Population Division (UN 2003), the fertility rate is assumed to be


below or above the replacement level (2.1) by just a quarter of a child
(Table 8.1). In the below-replacement scenario (fertility 1.85 children
per women) the world population is projected to be 7.4 billion in 2050
and 2.3 billion in 2300, while in the 'high' scenario (fertility 2.35) the

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Table 8.1 Evolution of the world population: alternative UN scenarios by
regions (billions)
World

Year Low Medium Zero-growth High Constant

2000 6.1 6.1 6.1 6.1 6.1


2050 7.4 8.9 8.9 10.6 12.8
2100 5.5 9.1 9.1 14 43.6
2150 3.9 8.5 8.5 16.7 244.4
2200 3.2 8.5 8.3 21.2 1,775
2225 2.7 8.8 8.3 27.8 14,783
2300 2.3 9 8.3 36.4 133,592

More-developed regions*
2000 1.2 1.2 1.2 1.2 1.2
2050 1.1 1.2 1.2 1.4 1.2
2100 0.8 1.1 1.1 1.7 0.9
2150 0.6 1.2 1.1 2.2 0.8
2200 0.6 1.2 1.1 2.8 0.7
2225 0.5 1.2 1.1 3.6 0.7
2300 0.4 1.3 1.1 4.7 0.6

Less-developed regions
2000 4.9 4.9 4.9 4.9 4.9
2050 6.3 7.7 7.7 9.3 11.6
2100 4.7 7.9 7.9 12.4 42.7
2150 3.3 7.3 73 14.6 243.6
2200 2.6 7.3 7.2 18.4 1,775
2225 2.2 7.5 7.2 24.2 14,782
2300 1.9 7.7 7.2 31.8 133,591

Notes: * Excl. Oceania. The scenarios are as follows:


Low: total fertility rates (slightly) below replacement rate for most of the periods.
Medium: total fertility rates below replacement rates over 2050-75 and then at replacement
level.
High: total fertility rates (slightly) above replacement rates.
Zero-growth: when population reaches a standstill under the medium scenario, then births
exactly balance deaths in order to maintain population constant in spite of the enduring
reduction of mortality.
Constant fertility: each country maintains the fertility it showed in 1995-2000.
Source: UN Population Division (2003).

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Sergio Cesaratto 185

figures are, respectively, 10.6 and 36.4 (the world population was 6.1 billion
in 2000). In the 'medium' scenario, in which fertility remains below the
replacement level up to 2175 and then returns to the replacement level,
the population would reach 9.2 billion in 2075 and then decline to
8.3 billion in 2175. What is striking is that all these scenarios result in a

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
significant ageing of the world population (that is, a small increase in
the fertility rates has dramatic effects on the population levels but has a
negligible effect on the ageing process).
Developing countries are where some of the possible population
developments show the most explosive outcomes and where, of course,
the completion of the demographic transition is most hoped for. In
these countries, even a fertility rate which has remained only slightly
above replacement since 2100 leads, albeit in the long run, to levels of
population that appear, at least so far, to strain sustainability, without
avoiding substantial ageing (UN 2003: 13-14). Developed countries
have completed their transition and are, in a sense, in the opposite situ-
ation in which the continuation of the present trends will determine a
dramatic diminution of total native population. What must be noted,
however, is that although in some developed countries a recovery of the
fertility trends may be hoped for, in addition to migration, in order to
reach a stabilization of the local population, this will not be enough to
slow down the ageing process.
After years of alarm about the explosion of the world population, age-
ing is now drawing much attention in the developed countries, where
the below-replacement rates are felt as a problem. According to many
expert opinions collected by the UN (for example UN 2003b: 9), this has
diverted the attention of the developed world from the consequences on
the world population of too-high worldwide fertility rates. It is impress-
ing that the hypothetical continuation of the current fertility rates
would lead to a population of 244 billion in 2150 and of 134 trillion (sic)
in 2300! What is striking in this case is that only in this scenario - or in
equivalent ones - might we obtain, in the long run, a reversal of the age-
ing process capable of calming the ageing alarmists. To rational minds,
ageing would thus appear as the price that humanity has to pay to
bring population growth under control - so that lower fertility, higher
longevity and ageing appear as a positive result of human development.

Working-age population and ageing in the more


developed countries
Although it is serious in all developed countries, the challenge from
demographic developments is not the same in all regions and countries.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
186 Pensions and Distribution in an Ageing Society

The broadest subdivision is between traditional settlement areas


(typically the western offshoots: the USA, Canada and Oceania), in
which the impact is softer, and non-settlement regions (typically Europe
and Japan), in which the impact is more dramatic. Among developed
European countries, however, there are also marked differences.2

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
The first consideration is that ageing is a process that has been taking
place for many decades at the global level, including the less-developed
countries, due to the demographic transition synthesized by the trend of
the fertility rated and life-expectancy. In all regions the age-group com-
position of the population is changing in favour of the older sections
and demographic dependency ratios are on the rise. These ratios should
be considered with care, since they may tell us little about the economic
sustainability of the inactive population. It may be noted here that the
total demographic dependency ratio rises considerably less than the old-
age ratio (the former actually falls in the less-developed world due to the
expected dramatic fall in the fertility rate).
The second consideration regards the differences between the devel-
oped regions. In short, if we were to enlist the regions according to the
momentum of the ageing process, this is more serious in Japan and
Southern Europe than in Western and Northern Europe, and is less dra-
matic in the traditional immigration or 'settlement' western offshoots.
To sum up the argument so far, in spite of the quantitative uncertainty,
especially in the medium-long range - in the short term the retirement
of the baby boom generation is a certain fact - it can safely be acknowl-
edged that an ageing process is underway, condensed by an increasing
'demographic' old-age dependency ratio. This process cannot be reversed
unless the population is allowed to explode at world and local levels. The
process is occurring, however, at a different pace in different world
regions. Examining the most developed regions, the difference between
them is represented by the fertility rate, closer to the replacement level in
the 'new' countries and below it in the 'old' regions. Past and expected
migration flows to the former, settlement, countries are also higher. The
two aspects are likely also to be interconnected, since migrants tend, at
least initially, to show a fertility rate higher than the native population.
The next question is then the degree to which more robust migration
flows both in settlement and non-settlement developed regions may
slow down, or even reverse, the ageing process.

Replacement migration
The UN Population Division has tried to assess the 'replacement migra-
tion' necessary to retain in 2050 some demographic variables at the

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Sergio Cesaratto 187

present level (UN 2000). The estimates (Table 8.2), based on the UN
projections of 1998 (medium variant), indicate the number of immi-
grants necessary, respectively, to keep constant: (1) the size of the over-
all population (column 3) (2) the size of the working-age population
(15-64) (column 4) and (3) the old-age dependency ratio constant

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
(column 5). The number of migrants is increasing from (1) to (3). For
instance, in the EU (15 countries) between 2000 and 2050 about 47 mil-
lion migrants would be needed to maintain the size of the overall popu-
lation at the present level; about 79 million to maintain the size of the
working-age population; and 700 million migrants to preserve the old-
age dependency ratio. The last target is practically out of reach even for
non-European settlement countries. The first target is the most easily
reached in most cases with robust but plausible migration flows (both
total and per-year) since the required migration is actually in line with
that experienced in the recent past (UN 2000: 93). It is more difficult to
obtain the second result (the population in working age declines faster
than the total population since the former suffers from the smaller size
of the younger generations while the latter benefits from the increasing
longevity). Also in this case, however, the UN argues that while 'some of
these numbers may appear high, they remain within the range of migra-
tion experienced in the recent past in some industrialized countries'
(UN 2000: 94). In some countries, the dramatic impression given by the
migration flow required simply to keep the labour population constant
(for example in Italy, 357 thousand immigrants per year according to
the UN) is diminished if we consider that from the point of view of
economics, the sustainability of ageing depends not on the amount of
the labour-age population, but on the share of it that is in effect used in
production, on its efficiency and on income distribution.
We shall next focus upon the repercussions of ageing: (a) on the
labour market and (b) on PAYG's costs in relation to output levels.

Ageing and the labour market

Scenarios on the evolution of labour supply


and dependency ratios
In a recent study (Burnieaux et al. 2003), the OECD has estimated the
evolution of labour supply over the periods 2000-25 and 2025-50 in the
participant countries. It should not pass unnoticed that in this official
document the potential labour supply is defined as the population aged
15 and over, modifying the hitherto standard definition of working-age

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
188

Table 8.2 Impact of migration flows: alternative UN scenarios (thousands)

Medium
Medium variant Constant Constant Constant
variant with zero total age group ratio
Country or region scenario migration population 15-64 15-64/165

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Hypothetical net number of migrants over
2000-50: scenarios by country or region

France 325 0 1,473 5,459 89,584


Germany 10,200 0 17,187 24,330 181,508
Italy 310 0 12,569 18,596 113,381
Japan 0 0 17,141 32,332 523,543
Republic of Korea -350 0 1,509 6,426 5,128,147
United Kingdom 1,000 0 2,634 6,247 59,722
United States 38,000 0 6,384 17,967 592,572
Europe 18,779 0 95,869 161,346 1,356,932
European Union
(15) 13,489 0 47,456 79,375 673,999

Total population in 1995 and by hypothetical


scenarios in 2050
1995 2050

France 58,020 59,883 59,357 61,121 67,130 187,193


Germany 81,661 73,303 58,812 81,661 92,022 299,272
Italy 57,338 41,197 40,722 57,338 66,395 193,518
Japan 125,472 104,921 104,921 127,457 150,697 817,965
Republic of Korea 44,949 51,275 51,751 53,470 60,125 6,233,275
United Kingdom 58,308 56,667 55,594 58,833 64,354 136,138
United States 267,020 349,318 290,643 297,970 315,644 1,065,174
Europe 727,912 627,691 600,464 727,912 809,399 2,346,459
European
Union (15) 371,937 331,307 310,839 372,440 418,509 1,228,341

Potential support ratio in 1995 and by


hypothetical scenarios in 2050

1995 2050

France 4.36 2.26 2.26 2.33 2.49 4.36


Germany 4.41 2.05 1.75 2.26 2.44 4.41
Italy 4.08 1.52 1.52 2.03 2.25 4.08
Japan 4.77 1.71 1.71 2.07 2.19 4.77

Continued

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Sergio Cesaratto 189

Table 8.2 Continued


Potential support ratio in 1995 and by
hypothetical «scenarios in 2050
1995 2050

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Republic of Korea 12.62 2.4 2.4 2.49 2.76 12.62
United Kingdom 4.09 2.37 2.36 2.49 2.64 4.09
United States 5.21 2.82 2.57 2.63 2.74 5.21
Europe 4.81 2.11 2.04 2.38 2.62 4.81
European
Union (15) 4.31 1.97 1.89 2.21 2.42 4.31

Source: UN Population Division (2000).

population as that aged 15-65, as if retirement had become a residual,


not the last part of the standard life-course. More importantly, it must be
observed that the OECD economists share the Neoclassical view accord-
ing to which labour supply tends to be employed at its 'natural' level
that may be identified for short with full employment. 3 According to a
less-conventional view, labour demand depends on effective demand
and not on the labour supply, so that labour supply cannot be defined
as 'abundant' or 'scarce' without knowing the labour demand (let alone
that, in this view, the secular trends of labour demand are one main
determinant of the long-run evolution of the labour supply). So, we
shall sometimes use the expression 'relative labour scarcity (or abun-
dance)', bearing in mind that the scarcity and abundance must be
defined with respect to given levels of labour demand and not in
absolute terms. With this warning in mind, let us consider the OECD
estimates of the future evolution of labour supply.
In Table 8.3 we have reclassified the OECD data, grouping countries
into Settlement, Newcomers, and Greying (gently or not) nations. 4 This
procedure allows some broad patterns to be identified, with the proviso
that each country has a very marked specificity in its demographic and
economic history.
In their baseline scenario the OECD experts consider the positive
effects on labour market participation due:

1 to the cohort effect (frozen at 2000 levels) which, more signifi-


cantly in the Southern European countries, will positively affect the

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
190

Table 8.3 Evolution of labour supply 2000-50, OECD projections


Baseline scenario Baseline scenario Labour
2000-25 2025-50 supply

With With
Total Total Total Total further further

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
population labour population labour reforms reforms
change supply change supply 2000-25* 2000-50*

Settlement countries
Australia 29.8 18.3 9.8 0 25.2 26
Canada 23.8 12.4 3.6 -1.9 17.1 15
New Zealand 25.1 11.9 5.9 -3.8 17.9 13.7
United States 24.9 14.5 19 17.3 19.7 40.5
Newcomers
Mexico 49.8 60.4 17.8 11.6 63.8 83.4
Turkey 43.4 11.8 17.7 2.4 17.1 21.6
Gently greying Europe
Iceland 21.1 16.6 5.1 -3.1 17.7 14.4
Norway 14.4 10.1 4.4 2.5 15.7 18.7
Sweden 8.9 -6.9 2.8 0.8 -2.6 -1.7
Netherlands 14.5 6.6 0.9 -2.2 17.7 14.9
Luxembourg 18.6 12.6 8.4 5.5 22.4 28.1
Ireland 24.2 34.8 7.5 -0.8 45.3 44.7
United
Kingdom 12 4 1.1 -2.3 9.5 7.2
Portugal 8.1 8 1.6 -5.4 18.1 13.2
Greying world
Denmark 7.8 -3.7 -0.7 -3.5 -1.6 -5.1
Finland 5.6 -8.3 -6.8 -8.4 -3.4 -11.4
Austria 4.7 -9.8 -5.8 -15.4 -2.7 -16.9
Belgium 8 3.5 2.1 -4.1 12.8 8.5
France 10 -3.4 -1 -6.6 9.3 2.8
Germany 2.5 -3.7 -7.8 -13.4 5.7 -8.3
Switzerland 6.4 1.6 -3.5 -7.6 7.8 -0.3
Italy -2.2 -4.2 -14.2 -24.7 4.4 -20.2
Spain 10.7 9.1 -5 -15.2 24.6 7.7
Greece 1.4 9.8 -5.7 -11.1 11.6 -0.6
Czech
Republic -0.5 -12.9 -14.7 -32.5 -7.4 -36.9
Hungary -1.9 -17 -15.6 -34.8 -16 -44.7
Poland 5.7 -9 -11.9 -28.6 -4 -31.1
Slovakia 10 -4.4 -10.3 -31.8 -0.4 -31.3
Korea 11.2 -7.2 -13.1 -24.2 4.7 -18
Japan -1 -11.8 -16.1 -23.1 -2.5 -24.5
OECD
unweighted
average 13.2 4.8 -0.8 -8.8 12.9 7

Note: * Conservative scenario.


Source: Burnieaux et al. (2003).

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Sergio Cesaratto 191

participation rates of the younger generations of women with a


stronger educational background; and
2 to already deliberated pension reforms aimed at increasing the
retirement age.

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
They also postulate lower unemployment rates and higher fertility rates.
All these factors increase the participation rates, but are not able to con-
trast the demographic effect in the long run. Over the period 2000-25,
in spite of the fact that total population is still rising (column 1), the
change in its composition due to the ageing process is the main factor
behind the fall of the labour supply in the Greying countries (column 2).
The labour supply is still rising in the settlement countries, latecomers
and in most gently-greying countries in which the more marked rise in
total population helps to offset the ageing process. On average, labour
supply in the OECD area is still on the rise. Over the period 2025-50 the
fall in total labour supply is dramatic in most countries, with the notable
exception of the USA (column 3). The OECD next assumes that addi-
tional reforms are enacted in order to provide further inducement to
the female participation rate, through various incentives, and to later
retirement, so as to obtain a fast convergence to a standard retirement
age of 67 in 2025 accompanied by actuarial disincentives to earlier
retirement. We report here the most prudent Tow' scenarios. The out-
come (columns 5 and 6) in most countries is a positive growth in labour
supply. However, considering a longer time span, 2000-50, the sign
would remain negative for most countries in the Greying world, partic-
ularly in Japan and Italy. The participation rates of the standard work-
ing-age population (Table 8.4) would decline slightly in 2000-25 in most
countries due to the ageing process and the associated retirement
process, and rise slightly in the following period. A more substantial rise
is expected if further reforms are legislated that favour female and older
worker participation.
In synthesis, there is a remarkable variety of regional situations. In the
settlement and newcomer countries the evolution of the population and
of the participation rates is such as to assure a growth in labour supply.
In countries like Italy, characterized by current low employment rates, a
given constant level of labour demand may be met in the short run by
employing the large pool of domestic idle labour and by resorting to
immigration, but a relative shortage of labour might appear, ceteris
paribus, in the long run (Aprile et al. 2002). Japan may encounter a sim-
ilar situation (Feldman 2004). Independently of the regional patterns,
however, the composition of the labour force is presumably changing
everywhere in favour of the female and older sections, augmenting,

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
192

Table 8.4 Evolution of participation rates in OECD countries: scenarios by the


OECD

Baseline scenario, Further re forms,b


changes over changes over
Levels in
2000a

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
2000-25 2025-50 2000-25 2025-50

Settlement countries
Australia 73.8 -0.3 -0.6 2.9 2.3
Canada 76.3 1.2 0.2 3.8 4
New Zealand 75.2 -2.3 -0.9 1.4 0.4
United States
average 77.2 -1.7 0.8 1.4 2.2
Newcomers
Mexico 62.3 6.5 0.2 9.1 9.1
Turkey 51.8 -10.2 -2.4 -6.3 -8.9
Gently greying Europe
Iceland 86.6 2.3 -0.9 2.5 1.5
Norway 80.7 1.7 1.1 4.1 5.2
Sweden 78.9 -5 0 -4.3 -0.2
Netherlands 74.6 2.2 1.6 9 10.3
Luxembourg 64.2 2 1.8 10.3 11.9
Ireland 67.4 9.7 1.1 16.3 17.4
United Kingdom 76.6 -1.2 0.8 1.7 2.5
Portugal 71.1 1.4 0.4 7.8 8.2
Greying world
Denmark 80 -2.3 0.3 -0.7 -0.4
Finland 74.3 0.3 1.4 3 4.2
Austria 70.6 -4.6 0.4 -0.4 -0.1
Belgium 65.2 1.7 0.4 8.7 9.1
France 68 -2.6 0.3 6.5 6.8
Germany 72.2 2.2 0.3 9.6 9.9
Switzerland 80.5 1.7 -0.6 5.4 4.6
Italy 60.3 2.9 0.2 8.1 8.3
Spain 66.7 3.2 1.7 11.8 13.5
Greece 63 8.4 0.6 11.1 11.7
Czech Republic 71.6 -1.4 -5.3 3.8 -1.4
Hungary 60.2 -1.9 -5.7 0.3 -4.8
Poland 65.8 -0.7 -6.2 4.2 -1.9
Slovakia 69.9 -1.3 -5.8 2.4 -3.3
Korea 64.3 -3.2 0.9 3.9 4.5
Japan 72.5 1.4 -0.3 10.7 10.5
OECD unweighted
average 70.7 0.3 -0.5 4.9 4.5

Notes:a Percentages of population aged 15-64. b Conservative scenario.


Source: Burnieaux etal. (2003).

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Sergio Cesaratto 193

ceteris paribus, the pressure on the male, prime age section of the labour
supply.
In addition in all countries the old-age dependency ratio, defined as
the ratio between the inactive over-65s on the labour force (those aged
over 15 that participate in the labour market), actually rises whatever

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
the scenario. Worth noticing are the trends in the total dependency ratio,
defined as the ratio between the inactive population and the labour
force.5 In the OECD baseline scenario this ratio rises less than the old-
age ratio given the compensating effect exerted by the diminished num-
ber of both dependent young and adults. In a number of cases it even
becomes negative in the 'further reforms' scenario, with the effect being
felt more strongly in the more greying countries. We shall return to this.

The economics of the impact of the demographic


changes on the labour market
Classical economists showed a primary interest in the relations between
population, wages and accumulation. The centrality in their theories of
the notion of wage rate as the amount of commodities that workers and
their families must receive in order to survive (Garegnani 1984) natu-
rally suggests an interest in the circumstances that regulate the standard
of living and the reproduction of the labour force. In addition, the deter-
mination of the wage rate on the basis of the bargaining power of labour
propounds a connection between the speed of the accumulation
process - which influences labour demand and wages - and population
developments - which influences labour supply. Having said this, with
the exception of Malthus, the Classical economists were far from hold-
ing the simplistic view of these interrelations that has often been attrib-
uted to them. In particular: (1) labour demand, and not just the wage
level, was seen as inducing population changes; moreover (2) the level
of wages could influence population in either direction; and finally,
(3) unfavourable population developments and accumulation rates had
no mechanical or immediately negative influence on the wage rate,
which depended on enduring social habits and conventions (Stirati
1994: chapter 4 and 175-6). While Marx regarded the existence of an
excess of labour population over the job opportunities as a necessary
check on wages, he considered population movements too slow to
account for the preservation of a labour reserve army over the cycle
and during the process of accumulation - a role that was not, nonethe-
less, excluded in the long run. The most direct role in preserving
the industrial reserve army was rather assumed by the modalities of
the accumulation process itself, in particular: by (a) the weakening of

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
194 Pensions and Distribution in an Ageing Society

the accumulation process, for a given 'organic composition of capital'


and, were its pace putting too much pressure on a given labour supply,
by (b) the progressive substitution of variable capital (direct labour) with
constant capital in the production process that diminished labour
demand in the course of the accumulation process (Marx [1867] 1974,

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Vol. I: chapter XXV: sections 1-3, for example 580-1 and 597-9).6
In the Marginal theory, population, although behind the determina-
tion of the labour supply schedule, played a very indirect, de facto sec-
ondary role in the determination of the wage rate which results from the
relative factors' scarcity and the technical conditions of production. The
interest in the complex relations between the wage rate and labour pop-
ulation developments investigated by the Classical economist was lost, as
much as the existence of a surplus of labour population, involuntary
unemployed, as a check on wages, whose interpretation was also the fruit
of the Classical conflict view of income distribution. The excess of popu-
lation in working age over the employed population is seen by Marginal
theory as consisting of voluntarily or temporarily unemployed.
According to Neoclassical theory, a fall in the rate of growth of labour
supply does negatively affect the patterns of employment and output. It
is, however, expected that the per capita capital endowment rises. As a
result of the mutation in the relative scarcity of factors, the wage rate is
expected to rise and the profit rate to fall.7 In principle, therefore, a fall
in the population in working age is accommodated by the Neoclassical
view like any other mutation of factors' endowments. However, we
are authorized to suspect that, in practice, a situation of greater labour
scarcity is seen by these economists with disquiet, given its impact on
the bargaining power of labour and distribution (see for example Cotis
2003).8 The impact of ageing on the costs of PAYG in relation to output,
to which we shall return, is viewed with even more open apprehension.
The Classical-Keynesian approach looks at the demographic develop-
ments in the context in which employment is determined by the pat-
terns of effective demand. According to this approach, employment is
labour-demand and not labour-supply led as in the Neoclassical
approach. Historical experience suggests that, in the past, capitalism has
never suffered from labour scarcity, mainly because it has resorted to
migration flows. Peasants, women, children and immigrants - that is the
'pre-capitalist' world that overlaps the 'reserve army' in which Marx
included the elderly - in different historical phases have constituted
pools of labour supply from which capitalists have drawn workers in
their millions, when necessary (Patnaik 2003). This is not to underesti-
mate the novelty of the demographic developments that occur which,

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Sergio Cesaratto 195

as we have seen, not even robust migration can arrest. What must nec-
essarily be reiterated in analysing these events is that whatever the
events on the labour supply side, labour demand does not mechanically
depend on labour supply through the flexibility of the wage rate, as
argued by Neoclassical economists, but on effective demand.

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
In dealing with the relative fall in the working-age population - possibly
bounded by migration, but only within the limits suggested by population
density, political and cultural considerations - the non-conventional
approach would point in the directions suggested by Marx and recon-
sidered by Kalecki (1943), that it is necessary for capitalism to be
endowed, on average, with a labour reserve army in order to discipline the
workforce. The first reaction of capitalism to a reduction in the labour
supply - for a given labour demand - would be to recruit those who are
still outside the labour market, women, old workers, but also the young
and even invalids (as stated in a official document of the OECD 2004).9
A persistent situation of relative labour scarcity for a given level of
labour demand may also determine a reaction on the side of the capital-
ist class applied to reconstituting the industrial reserve army through
variations of output levels - achieved by deliberate economic policies or
by a fall in investment decisions. It is mainly in this sense that, in a
Classical-Kaleckian context, circumstances on the labour supply side
can influence labour demand and output. So, although in principle a sit-
uation of relative labour scarcity may favourably affect the direct and
social wage rates and raise the propensity to consume, thus sustaining
effective demand, it may also induce a reaction on the side of the dom-
inant classes aimed at restraining labour demand and preserving an
industrial reserve army. Marx's theory of the labour reserve army would
also suggest a wave of labour-substituting innovations.
The real threat is that, especially in Europe, the ageing process associ-
ated to the alarm over a labour shortage and the consequent fear of the
necessity of increasing the share of income going, directly or indirectly
to labour, might lead to neo-Malthusian, economic stagnation policies
in order to preserve an industrial reserve army. The present policies of
weakening the labour-market institutions that reinforce the labour bar-
gaining power may also be explained in this perspective. In this context,
policies devoted to diminishing the expected level of pensions may
change the life-cycle conceptions that were consolidated in the second
half of last century, inducing the social acceptance of a higher retire-
ment age up to the point of again making acceptable the idea of work-
ing in old age - which, given the higher longevity, would mean over
65 years or so - to avoid poverty.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
196 Pensions and Distribution in an Ageing Society

Ageing and the cost of PAYG

Scenarios on the cost of PAYG on GDP


The second question associated with ageing is the increasing weight of
PAYG on output. The conventional wisdom is that this will determine an

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
increase in the tax burden, which might act as a disincentive on labour
and saving supply, or increase public deficits and debt which, by absorbing
saving and determining a rise in the interest rates, would undermine capi-
tal accumulation. Let us consider some OECD estimates of future PAYG
costs on GDP in order to estimate the order of magnitude involved (Dang
et al. 2001). Behind the projections there is a supply-side model in which
labour supply and productivity growth determine the rate of growth of
GDP. This is unsatisfactory from a Keynesian point of view, according to
which the GDP trend is based on the pattern of effective demand - which
depends on policies (including pension policies) rather than on 'natural'
market forces. Nonetheless, this conventional kind of exercise may pro-
vide an order of magnitude for the phenomenon under observation.
The OECD bases its estimates on the following equation:

PENS POP+55 pop20~64 Av. Benefit Recipients


GDP POP ' Employment Av. Productivity POP+5S
20 64

PENS/GDP is PAYG's costs on gross output. POP^^/POP2064 is the ratio


between the population over 55 over the population in working-age
(standard definition). This ratio measures the ageing process that influ-
ences the relative old-age pension costs since many workers retire before
65. POP20'64/Employment is the inverse of the employment rate. This
ratio measures the aforementioned expected rise in the share of work-
ing-age population in employment given, on the one side, the fall in the
potential labour-supply and, on the other, the increasing female labour-
market participation and the measures to reduce early retirement.
Observe that a fall of this ratio, say due to a fall of POP20'64 for a given
employment level, indicates that a decreasing number of inactive adults
is dependent on the active adults, and this may compensate the ageing
burden. The term Av. Benefit/Av. Productivity is the ratio of the average
pension benefit on per capita productivity, measuring the effects of pen-
sion reforms - such as the abolition of the real indexation of pensions to
wages or, in Italy and Sweden, the effects of the NDC reforms that link
the pension benefit to the average (not the final) wage and reduce them
whenever the expected years of retirement increase. Finally, the terms
Recipients/POP+55 is the eligibility rate, which also reflects the introduc-
tion of more restrictive rules of access to retirement and pensions.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Sergio Cesaratto 197

The OECD estimates the rate of change over 2000-50 of PENS/GDP as


the summation of the rate of change of the components on the right
hand side of equation (8.1). Table 8.5 shows the results.
Initial spending ratios, shown in the first column, depend not just on
the stage reached in the ageing process, but also on the employment

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Table 8.5 Changes in old-age pension spending 2000-50, OECD estimates
Contributions of
Total old- Total old-
,age pension age pension Old-age
spending; spending; depend- Empoly- Eligi-
level in level in Variation ency ment Benefit bility
2000 2050 2000-50 ratio ratio ratio ratio

Settlement countries
Australia 3 4.6 1.6 2.5 -0.1 -0.5 -0.2
Canada 5.1 10.9 5.8 5.1 0 -0.6 1.3
New Zealand 4.8 10.5 5.7 4.7 -0.1 1 0
United States 4.4 6.2 1.8 2.4 -0.1 -0.2 -0.3
Average 4.3 8.1 3.7
Gently greying Europe
Norway 4.9 12.9 8 3 0.1 3.9 1.2
Sweden 9.2 10.8 1.6 3.9 -0.5 -2.1 0.4
Netherlands 5.2 10 4.8 3.8 -0.5 0.2 1.4
United Kingdoim 4.3 3.6 -0.7 1.7 0.1 -2.5 0.1
Average 5.9 9.3 3.4
Greying world
Denmark 6.1 8.8 2.7 2.7 -0.3 -1.5 1.7
Austria 9.5 11.7 2.2 7.6 -1.9 -1.1 -2.4
Belgium 8.8 12.1 3.3 4.7 -0.7 -1.6 1
France 12.1 15.9 3.8 7.6 -0.5 -3.4 0.4
Germany 11.8 16.8 5 6.4 -0.7 -2.7 2.1
Italy 14.2 13.9 -0.3 10.1 -3.2 -5.5 -1.5
Spain 9.4 17.4 8 8.6 -2.6 0 2
Czech Republic: 7.8 14.6 6.8 8.2 -0.8 -0.1 -0.1
Hungary 6 7.2 1.2 2.9 -1 -0.3 -0.4
Poland 10.8 8.3 -2.5 7.3 -1.3 -5.9 -2.1
Korea 2.1 10.1 8 4.8 -1 0.2 5
Japan 7.9 8.5 0.6 5.1 -1.2 -3.9 0.9
Average 8.9 12.1 3.2
OECD
unweighted
average 8.1 11.6 3.5 5.6 -1.0 -1.5 0.6
Coeff. of
variation* 0.43 0.35

Notes: * standard deviation/average.


f
Source: Dang et al. (2001).

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
198 Pensions and Distribution in an Ageing Society

rate - which, in countries like Italy where it is particularly low, deter-


mines a heavier pension burden on the employed. It also depends on
the national characteristic of the national pension system - which is
much less generous in countries in which the benefit is not broadly
linked to pre-retirement earnings but is mainly 'flat-rate' (the traditional

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
example is the UK). The coefficient of variation (the standard deviation
over the average) measures the scatter of the national data around the
average. It falls from 0.43 in 2000 to 0.35 in 2050 (0.32 if the expendi-
ture in the UK is considered as 6.5 per cent, which now sounds more
reasonable),10 suggesting that in 2050 there will be a greater relative
homogeneity among the OECD countries. Although the ratio of over 55
on the population in working age is increasing at a different pace, the
pension reforms endeavoured by some countries (and further reforms
will take place in the next future) affecting the level of pensions in rela-
tion to productivity (and wages) and restricting the eligibility rights, and
the rise in the rate of employment of the working age population in
countries where this was relatively low, contribute to this greater homo-
geneity. The case of Italy is very representative, since in spite of present-
ing the most extensive ageing process (with Japan), this country
succeeds in stabilising (albeit in the long run) the expected spending
ratio as a result of the NDC reform and of a rising employment ratio. The
dramatic effect on the level of benefits with reference to per-capita
output is quite evident from the table. The social problem that this per-
spective opens does not regard only Italy.11 The OECD is clearly more
worried about the impact of the increasing old-age spending (which
should include also health spending) on taxation and the public debt
than on the social effects.

Compensation factors: productivity growth and the


(relative) constancy of the total dependency ratio
The role of productivity growth in alleviating the rise of the ratio of
pension spending to output needs also to be carefully examined.
Productivity growth is not a panacea, as is often envisaged. In a nutshell,
if productivity growth is used to offset ageing (the rise in the number of
retirees over employment), than it cannot be used to increase pensions
in line with increased productivity. It might therefore be necessary to
redistribute income from wages and profits towards the retirees if we
want to keep pensions in line with the growth of incomes in the active
population. Nonetheless, productivity gains, which have been spectacu-
lar over the last two centuries, will lead to a substantial increase in the
level of net real income of active workers in spite of a possible higher

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Sergio Cesaratto 199

contribution rate (Palley 2002). The burden of PAYG can also be


transferred to non-wage earners either by increasing real wages in
proportion with contributions, or by using general taxation. As Wray
(1999: 1 et passim) has pointed out, most economists fail to distinguish
between the financial imbalances of PAYG, that may well increase into

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
the future, given the current parameters that govern the schemes, and
the 'real problems involved in producing a sufficient quantity of
resources to care for future retirees', that cannot realistically be consid-
ered as an insurmountable burden. The viability of PAYG is, to a large
extent, a question of distribution of the social product between genera-
tions and social classes, and this has no mechanical negative impact on
economic growth and welfare. One big question concerns the difficulty
of raising taxes, especially over financial capital in this era of financial
liberalization and tax competition. In this regard, it must be noted that
financial liberalization has been a deliberate political choice pursued,
inter alia, in order to create problems for the financing of the Welfare
State.
It may also be asked if the cost of an ageing society is not compensated
by the lower costs of youth and dependent adults. As aptly observed by
Concialdi, ultimately 'it is the ratio of the population out of work to the
population employed that gives the most reliable idea of the likely evo-
lution of future social expenditures' (Concialdi 1999: section 4). Indeed,
as also seen above, the total dependency ratio (the inactive population
on the labour force) rises less than the ratio of the old-age dependency
ratio (inactive elders to the labour force). Conventional economists have
actually led us to regard the ageing process as a progressive contraction of
the active population in favour of the inactive, forgetting that this is a
process partially offset by the falling number of dependent adults. They
are probably trapped into this way of thinking by the Neoclassical theory
that our economies are, on average over the business cycles, in full
employment.
Concialdi defines as the 'overall economic dependency ratio' the pro-
portion of people out of work (whether inactive or unemployed) to
people actually employed (Concialdi 1999: section 4). The inactive sec-
tion of the population has two components: the inactive young and the
non-working adults. As far as the former group is concerned, the debate
on the relative costs of their upbringing and education versus the sup-
port costs of the old has not led to conclusive results (see, for instance,
Denton and Spencer 1999). The prevalent opinion seems that once the
reduced spending for the shrinking young section of society is duly
taken into account, this will not be a decisive counter force to the

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
200 Pensions and Distribution in an Ageing Society

increasing relative costs of ageing. With regard to the second group, as


we have seen, the ageing process should accelerate the participation in
the labour market of women and of mature workers, reducing the share
of inactive adults (in practice many capitalist economies will approach a
genuine full employment). Concialdi estimates the overall economic

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
dependency ratio for the European Union over the next two or three
decades on the basis of various hypothesis on the trends of total employ-
ment. The principal conclusion he draws is that there is, in principle,
some chance that future population changes will not increase the eco-
nomic burden on workers (Concialdi 1999: section 4). In a nutshell, this
author assumes that in the next two decades or so in Europe employ-
ment will not fall significantly (and will possibly increase slightly). Since
also total population is not expected to fall - merely to grow older - the
overall economic dependency ratio will not rise greatly. So, if productivity
rises, the dependent and the active population can both share in the
productivity gains without changing the distribution of the social out-
put between the two sections of the population:
when we look at the whole population of the European Union, vari-
ations in the numbers of the dependent population will be relatively
small over the next 25 years. There was an estimated dependent pop-
ulation of 222 million people in 1995. The figure for the year 2020
will be between 218 and 230 million. Consequently, the average cost
of the dependent population could approximately follow the rate of
real economic growth without increasing the burden on workers.
(Concialdi 1999: section 5)
If the overall economic dependency ratio rises, the dependent population
can only share part of productivity growth, but its welfare will still
rise as long as the latter grows at a higher rate than the former. Let us
define etot = DepPop/Nw as the overall economic dependency ratio and
d = (DepPopbd)/(Nw7r) as the weight of the dependent population on
output; Nw is the number of workers, hd is the average transfer to the
dependants and ir is output per worker (time subscript omitted). The
last expression d = etot (bd/ir) can be written. If we want d to remain con-
stant, then:
bd=w- etot (8.2)
where the hat indicates the rates of growth. According to Concialdi
(1999: figure 5), in spite of the rise in etoV an expected TT higher than etot
will allow a substantial rise in bd in many European countries at least up
to 2020 (for example around 40-50 per cent in France, Germany and

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Sergio Cesaratto 201

Italy). Note that it is the approximation to full employment conditions


that does the trick (idle labour resources tend to disappear and this com-
pensates for the rise in the old population). In this regard our earlier
conclusions must be recalled: it is the contraction in the labour reserve
army that may represent the real obstacle to an effective coping with an

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
ageing society, not the lack of labour resources.
Concialdi's method is interesting, and it deserves to be developed in
future research. This should also better assess the impact of the chang-
ing composition of the dependent population - less young and prime
age inactives, more old - on private and social support costs, and on the
financing sources (say, inactive spouses would transit from family sup-
port to old-age pensions), since this has serious income distribution
implications.

Final remarks

This chapter has tentatively explored the possible economic impact of


ageing on the labour market and the viability of PAYG.
The conventional wisdom is that the fall in fertility may negatively
influence the level and growth of employment, at least in the developed
countries. Not all countries would be on the same footing in this regard:
some possess reserves of labour that are currently (shamefully) kept idle;
traditional settlement countries may welcome a larger number of immi-
grants than congested non-settlement regions. In addition, in most
countries the emerging situation may produce a spontaneous increase in
the demand for the labour services of older workers, many of whom are
presently still being dismissed against their will. Finally, per capita pro-
ductivity growth, including some reversal of working hours, may free
labour resources from the current activities. The conventional view
strongly reflects the standard theory whereby the level of employment
is governed by the labour supply through the flexibility of wages. We
reject this view. This does not imply, however, discharging the impact
of the falling supply of labour population on to the labour market.
Assuming as a working hypothesis the persistence of employment at the
present levels (with output growth resting on productivity growth),
the falling supply of domestic labour population is not necessarily a
constraint on output growth, insofar it is in most countries compen-
sated by higher participation rates (including later retirement), technical
progress, immigration and, perhaps, some recovery in fertility. The chal-
lenge of a shrinking labour supply will, however, seriously bite in some
countries like Japan and Italy. The novel dimension added in this

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
202 Pensions and Distribution in an Ageing Society

chapter is whether the additional sources of labour supply are such as to


preserve an industrial reserve army and if capitalism can live without a
significant labour reserve army that keeps the whip of competition on
the workers. It is in this sense that, according to non-conventional eco-
nomic analysis, a shrinking labour supply may affect employment and

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
growth by inducing deflationary choices, particularly in Europe. The
more so if ageing brings about a rising cost of the dependent population.
Most projections forecast a rise in PAYG's cost over income. Whereas
differences within developed countries will persist, the tendency - as
presently predicted - is towards a greater homogeneity. We discussed the
role of productivity growth in attenuating the expected rise in PAYG's bur-
den on social output. In synthesis, this factor can help to keep the effects
of a rising dependency ratio at bay, but cannot at the same time be used to
raise pension benefits in line with per capita output growth. If we wish to
avoid a relative impoverishment of the retirees, we may have to accept the
idea that wages and/or profits have to share part of the productivity gains
with the retirees. To look at the old-age dependency ratio alone might,
however, be too limited a prospect. Some authors suggest looking at a total
dependency ratio that includes also the dependent youth and the depend-
ent adults, both shrinking components. The total dependency ratio would
therefore suggest a less gloomy future. The composition of the dependent
population would, however, change in favour of the old component, more
dependent on public (tax-financed) public support compared to the other
components (young and spouses) sustained by the family. The old compo-
nent is generally considered also to be more costly. Although the total flow
to the dependents changes less than expected, the increase and changing
composition of the flow may open up questions of political acceptability.
We may finally note that according to the Classical-Keynesian
approach there is no mechanical relation between an increasing amount
of resources going to the old, or more generally to the dependents, and
economic growth. On the one hand social transfers have positive effects
on aggregate demand and hence on growth. On the other hand they
may negatively affect the political climate and the incentive to invest.
The incentive to invest may be negatively affected also by tax competi-
tion in open economies. This is not, however, a mechanical result and
may be avoided if a robust and persistent social consensus sustains those
social transfers, so as to induce capitalists to accept a lower post-tax rate
of profit. International social dumping might be avoided by a degree of
tax coordination and, perhaps, by the reintroduction of forms of capital
control. From an analytical point of view what must be strongly noted
is the absence, in a heterodox framework, of an automatic negative

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Sergio Cesaratto 203

effect of increasing social transfers on accumulation: these effects may


take place, but their origin resides in the refusal of a changing income
distribution, not in some mechanical relation.

Notes

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
1 This is an echo of the procedure famously advanced by Marx in the Grundrisse:
It seems to be correct to begin with the real and the concrete, with the real
precondition, thus to begin, in economics, with for example the popula-
tion, which is the foundation and the subject of the entire social act of pro-
duction. However, on a closer examination this proves false. The
population is an abstraction if I leave out, for example, the classes of which
it is composed. These classes are in turn an empty phrase if I am not famil-
iar with the elements on which they rest. For example, wage labour, capi-
tal, etc. ... Thus, if I were to begin with population, this would be a chaotic
conception of the whole, and I would then, by means of further determi-
nation, move analytically towards ever more simple concepts, from the
imagined concrete towards ever thinner abstractions until I had arrived at
the population again, but this time not as a chaotic conception of a whole,
but as a rich total of many determinations and relations. (Marx 1857-58
[1973]: 100)
2 The reader is referred to table 8.2 in Cesaratto (2005) that provides a broad idea
of the most likely comparative evolution in different regions based on UN pro-
jections concerning the period 2000-50 (UN-Population Division 2002a).
3 Of course, for mainstream economists the potential and the effective labour
supply do not coincide, given the existence of people that remain voluntarily
outside the labour market. Only with the contraction of the number of people
in labour age, do potential and effective labour supply tend to correspond.
4 The OECD experts assume that migration flows remain at their recent past
levels, which have been very high for the USA but not necessarily so for
European countries, so these estimates should be looked upon with the pro-
viso of a conservative hypothesis on the migration flows for many countries.
5 The reader is referred to table 8.6 in Cesaratto (2005) also based on Burnieaux
etal. (2003).
6 Marx could therefore conclude that the
law of capitalist accumulation, metamorphosed by economists into a pre-
tended law of Nature, in reality merely states that the very nature of accu-
mulation excludes every diminution in the degree of exploitation of
labour, and every rise in the price of labour, which could seriously imperil
the continual reproduction, on an ever enlarging scale, of the capitalistic
relation. (Marx [1867] 1974, Vol. I: 582)
In other words, the existence of a labour reserve army does not depend, accord-
ing to Marx, on fertility variations, too slow to be effective for this purpose, but
on the patterns of accumulation. It should be noted that in Marx labour sup-
ply does not directly affect labour demand, as the existence of persistent unem-
ployment shows (this is true also for the other Classical economists, cf. Stirati
1994: chapter 6), but may affect the rate of accumulation, and therefore indi-
rectly labour demand, by influencing distribution.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
204 Pensions and Distribution in an Ageing Society

7 The rise in the wage rate may induce some voluntary unemployed, whose
reserve wage is below the new level of the wage rate, to enter the labour mar-
ket, thus raising the activity and employment rates.
8 An example of Neoclassical analysis is provided by Boersch-Supan (2003).
9 '[R]eforms need to identify disabilities correctly, distinguishing between
minor and major disabilities in order that the term "disabled" is no longer

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
equated with being unable to work' (Boersch-Supan 2003: 10-11).
10 The projections for the UK are over-optimistic. According to the Economist
(Sept. 2004) the disarray of the British private-pillar 'casts doubts on the
sustainability of Britain's cheap public pensions'. This opinion follows the
results of a government commission on pensions that presented very
pessimistic forecasts about expected pension incomes in Britain.
11 The OECD experts specify that the expected decline in average benefits
relative to average productivity over the period 2000-50 will be - 1 6 % in
Belgium, - 1 1 % in Denmark, - 2 1 % in France, -20% in Germany, - 3 0 % in
Italy, - 3 8 % in Japan, - 5 1 % in Poland, - 2 2 % in Sweden, and - 4 7 % in the
United Kingdom.

References
Aprile, R., De Persio, P. and Lucarelli, A. (2002) 'Una previsione di medio lungo
periodo dei tassi di attivita secondo un approccio generazionale', Economia &
Lavoro, 36: 61-87.
Bloom, D.E. and Canning, D. (2004) Global Demographic Change: Dimension
and Economic Significance, Federal Reserve Bank of Kansas City Symposium on
Global Demographic Change: Economic Impacts and Policy Challenges,
Jackson Hole, Wyoming, http://www.kansascityfed.org/PUBLICAT/SYMPOS/
2004/pdf/BloomandCanning.Paper.0923.pdf.
Boersch-Supan, A. (2003) 'Labour Market Effects of Population Aging', Labour, 17:
5-44.
Burnieaux, J.M., Duval, R. and Jaumotte, E (2003) Coping with Ageing: A Dynamic
Approach to Quantify the Impact of Alternative Policy Options on Future Labour sup-
ply in OECD Countries, OECD Economic Department Working Paper no. 371
(Paris: OECD).
Cesaratto, S. (2002) 'The Economics of Pensions: A Non-Conventional
Approach', Review of Political Economy, 14: 149-77.
Cesaratto, S. (2005) Pension Reform and Economic Theory: A Non-Orthodox Analysis
(Cheltenham, UK and Northampton, MA, USA: Edward Elgar).
Cesaratto, S. (2006a) 'Notes on the Pros and Cons of the Notional Defined
Contribution Reforms', Journal of Income Distribution forthcoming.
Cesaratto, S. (2006b) 'The Transition to Fully Funded Pension Schemes: A Non-
Orthodox Criticism', Cambridge Journal of Economics forthcoming.
Concialdi, P. (1999) 'Demography, Employment and the Future of Social
Protection Financing, in Ministry of Social Affairs and Health', paper presented
at the conference 'Financing Social Protection in Europe', Helsinki (www.vn.fi).
Cotis, J.-P (2003) 'Reform of the European Labour Markets: A View from the
OECD', mimeo.
Dang, T.T., Antolin, P. and Oxley, H. (2001) Fiscal Implications ofAgeing: Projections
of Age-Related Spending, OECD, Economic Department Working Paper no. 31
(Paris: OECD).

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Sergio Cesaratto 205

Denton, FT. and Spencer, B.G. (1999) Social and Economic Dimensions of Aging
Population, Social and Economic Dimensions of an Aging Population, Research
Papers (1).
Feldman, R.A. (2004) 'Japan Economics - Immigrants: Torrent or Trickle?',
Morgan Stanley, Global Economic Forum (8 January 2004), http://www.
morganstanley.com/GEFdata/digests/20040108-thu.html

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Garegnani, P. (1984) 'Value and Distribution in the Classical Economists and
Marx', Oxford Economic Papers, 56: 291-325.
Kalecki, M. (1943) 'Political Aspect of Full employment', in M. Kalecki (1971)
Selected Essays on the Dynamics of the Capitalist Economy (Cambridge: Cambridge
University Press): 138-45.
Marx, K. (1857-58) Grundrisse, Foundations of the Critique of Political Economy
(Rough Draft) reprinted 1973 (Harmondsworth: Penguin Books).
Marx, K. (1867) Capital. A Critical Analysis of the Capitalist Production (I—III),
reprinted 1974 (London: Lawrence & Wishart).
Palley, T.I. (2002) 'Social Security: Prefunding is Not the Answer!', Challenge, 45:
97-118.
Patnaik, P. (2004) 'What Should be the Scope of "Development Economics"?',
http://www.networkideas.org/themes/macroeconomic/oct2003/mp06_
Development_Economics.htm.
Stirati, A. (1994) The Theory of Wages in Classical Economics (Aldershot, UK and
Brookfield, USA: Edward Elgar).
UN Population Division (2000) Replacement Migration: Is It a Solution to Declining
and Ageing Population? (New York: UN).
UN Population Division (2002a) World Population Ageing: 1950-2050 (New York:
UN).
UN Population Division (2002b) Completing the Fertility Transition (New York:
UN).
UN Population Division (2003) World Population in 2300 (New York: UN).
Wray, L.R. (1999) The Emperor Has No Clothes: President Clinton's Proposed Social
Security Reform, Jerome Levy Economic Institute o Bard College, Public Policy
Note (2). http://www.levy.org/pubs/pn/pn99_2.pdf.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
9
Do Profits Affect Investment
and Employment? An

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Empirical Test Based on the
Bhaduri-Marglin Model*
Ozlem Onaran and Engelbert Stockhammer

Introduction
This chapter aims at clarifying the macroeconomic effect of changes in
functional income distribution empirically for a range of developed and
developing countries. In doing so, our goal is to discuss the crucial pol-
icy issues related with neoliberal policies in the developing, as well as
developed countries in the post-1980 era; by focusing o n the effects of
distributional policies we seek to contribute to the explanation of the
reasons for the stagnant accumulation and employment growth rates.
Both the structural adjustment agenda in the developing countries, and
the debate about the European u n e m p l o y m e n t have been cases where
mainstream economics has pushed for policy changes favouring a pro-
capital redistribution of income, and a deregulation of the labour

* This chapter puts together the empirical conclusions of two previous papers
by the authors, and improves on the theoretical and political implications of the
findings. Several findings, indicated in the text are reprinted from Structural Change
and Economic Dynamics, vol. 15, Stockhammer, E. and Onaran, O. 'Accumulation,
distribution and employment: a structural VAR approach to a Kaleckian macro
model', 421-47, copyright© 2004, with permission from Elsevier, and from
Emerging Markets Finance and Trade vol. 41, Onaran, O. and Stockhammer, E., 'Two
Different Export-Oriented Growth Strategies: Accumulation and Distribution in
Turkey and South Korea,' no. 1, 65-89, copyright© 2005, with permission from
M.E. Sharpe, Inc. At several stages, we have benefited from comments by Eckhard
Hein, Amit Bhaduri, Philip Arestis, and the fruitful discussions in the Eighth
Workshop on 'Wages, Distribution and Growth' of the Research Network
'Alternative Conceptions of Macroeconomic Policies under the Conditions of
Unemployment, Globalization and High Public Debt', Berlin, 29-30, October 2004.

206

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Ozlem Onaran and Engelbert Stockhammer 207

market. For neoclassical economics unemployment is, in the last


instance, a labour market phenomenon. It is due to 'too high' real
wages, which in turn are a result of so-called labour market 'distortions',
like labour market regulations and trade unions. In contrast, Post-
Keynesians argue that unemployment is the result of demand deficien-

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
cies on the goods markets, and that the latter result particularly from a
slowdown in investments.
The resolution of this controversy requires the test of the dynamic
interaction between distribution, accumulation, growth and employ-
ment. For this purpose, in this study a Kaleckian-Post-Keynesian macro-
economic model, which is an extended version of the Bhaduri and
Marglin (1990) model, serves as the starting point. The merit of a
Kaleckian model for our purposes is that it highlights the dual function
of wages as a component of aggregate demand as well as a cost item, as
opposed to mainstream economics which perceives wages merely as a
cost item. Depending on the relative magnitude of these two effects,
Kaleckian models distinguish between profit-led and wage-led regimes,
where the latter is defined as a low rate of accumulation being caused by
a high profit share. Allowing for capacity utilization to vary in these
models gives rise to the possibility of a wage-led regime, that is a higher
rate of accumulation as a result of an increase in the wage share, if the
demand effect on investment is stronger than the profit effect.
Which regime prevails in a certain economy is an empirical question.
Are actual economies wage-led or profit-led? Current orthodoxy implic-
itly assumes that they are profit-led, and thus supports the neoliberal pol-
icy agenda. The purpose of the chapter is to carry this discussion into the
empirical terrain, and to test whether accumulation and employment are
profit-led in two groups of countries. We do so by means of a structural
vector autoregression (SVAR) model. The model is estimated for USA, UK
and France to represent the major developed countries, and for Turkey
and Korea to represent developing countries. The latter are chosen since
they represent two different export-oriented growth experiences. The
results of the adjustment experiences of both countries are in striking
contrast to orthodox theory, however, they also present counter-exam-
ples to each other in terms of their ways of integrating into the world
economy. Thereby, they provide examples for comparing different eco-
nomic policies among the developing countries as well.
The rest of the chapter is organized as follows. The next section intro-
duces the theoretical Kaleckian-Post-Keynesian model. This is followed
by a brief discussion of the estimation method and presents the hypothe-
ses to be tested by the empirical analysis. We then summarize the

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
208 Do Profits Affect Investment and Employment?

estimation results for developed and developing countries respectively,


followed by our conclusions, as well as open questions and challenges for
future research, with a discussion of policy implications of the results.

The model

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
For analysing the dynamic effects of distribution on growth, accumula-
tion and employment in these two groups of countries, we utilize a post-
Keynesian open economy model, which is an extension to the model of
Bhaduri and Marglin (1990). We augment the goods market block of this
model by a demand-driven labour market, a reserve army effect in the
Marxian sense, and technological change. Table 9.1 is a summary of this
linear open economy model.
The model developed by Marglin and Bhaduri (1990) is a more general
formulation of earlier neo-Kaleckian models by Rowthorn (1981), Dutt
(1984), Taylor (1985) and Blecker (1989), and allows for profit-led as well
as for wage-led growth regimes.1 This generality borrows itself to the
decomposition of the profit rate (r) into the profit share (TT), capacity uti-
lization (z) and (technical) capital productivity (k):

r= =
i fll = 7rz/c (91)
Then, for the sake of simplicity, assuming that technical capital produc-
tivity is constant, the rate of accumulation (g{), which is the ratio of new
investment to the stock of capital (£), can be formulated as a function of
the past values of the profit share (IT), and capacity utilization (z), which
constitute the current expected rate of profit. Equation (9.2) in Table 9.1
presents an extended linear version of this accumulation function, where
the effect of productivity growth on investment is also incorporated.
The goods market part consists of behavioural functions for accumula-
tion, savings and net exports. This part is then complemented by a distri-
bution function, a labour productivity function and an unemployment
function.
Equation (9.3) in Table 9.1 is a simple Cambridge savings function,
where the ratio of domestic savings to capital stock is a function of
capacity utilization and income distribution, that is the profit share.
Assuming that workers have a lower marginal propensity to save than
capitalists, b 2 is positive and measures the differences in savings propen-
sity between profit incomes and wage incomes.
Equation (9.4) in Table 9.1 incorporates international trade focusing
on the effect of distribution and growth on net exports, leaving the
other crucial variable of an open economy outside the model. Profit

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Ozlem Onaran and Engelbert Stockhammer 209

Table 9.1 Summary of the model

i ^
(9.2) Accumulation = a + UlZt l + ai7Tt 1 + a x
gt = Y ° ~ - &^
(9.3) Cambridge Savings equation ^domestic = ^ + ^ ^

(9.4) Net exports nxt = -h1zt + h2irt

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
(9.5) Income distribution 7rt = d0 + dxzt + d2ut + d?gxt
(9.6) Employment ut = n- exgt -e2Azt - e37rt + e 4 u t _i
+ esgxt
(9.7) Productivity growth gXt = T0 + T& + T2Zt

£ accumulation (growth of capital stock)


™Sdomestic
domestic savings / capital stock
Z capacity utilization (capital productivity)
77 profit share
nx net exports (normalized by capital stock)
u unemployment rate
g* productivity growth
All coefficients are positive numbers

Notes: See Stockhammer and Onaran (2004) for a detailed discussion of the theoretical
background.

share is inversely related to unit labour costs. Accordingly net exports


(again normalized by capital stock) are a positive function of the profit
share and a negative function of capacity utilization (since imports are a
positive function of the domestic demand).
The fifth equation in Table 9.1 models the distribution of income as a
positive function of capacity utilization via pro-cyclical mark-up, a pos-
itive function of the unemployment rate (u), reflecting labour's bargain-
ing position via the Marxian reserve-army effect, and finally the growth
of labour productivity. The latter will positively affect the profit share if
wages are imperfectly indexed to productivity growth.
Equation (9.6) in Table 9.1 models the labour market. Unemployment
is a negative function of output, thus the change of capacity utilization
and the growth of capital stock. Next, if the cost of labour is important
for labour demand, as the neoclassical theory would suggest, the profit
share (being inversely related with the real wage, after controlling for
productivity) is expected to have a negative effect on unemployment.
Unemployment will also depend on past unemployment via a hysteresis
effect. Finally, if technological change is not accompanied by a growth in
demand, the growth of labour productivity could lead to an increase in
unemployment. The constant, e0f captures labour supply shocks.
Finally, equation (9.7) in Table 9.1 models the growth of labour pro-
ductivity (x) as determined by accumulation and capacity utilization.
Exogenous technical progress is captured by the constant term, r0.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
210 Do Profits Affect Investment and Employment?

The goods market equilibrium is determined by investment being


equal to total domestic and foreign savings, that is:
pi _ ^Stotal _ ^Sdomestic _ ^.y /g o\

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Capacity utilization implied by the goods market equilibrium can be
written as:

[gt + (h2 - b2)7Tt] (9.9)


b1 + h1

The effect of an increase in the profit share on capacity utilization is inde-


terminate, and will depend in the medium run on the relative responsive-
ness of consumption and investment to profits. Contemporaneously as
well, this effect, thus the sign of %, will be indeterminate. If exports react
strongly to the profit share, whereas domestic consumption decreases only
mildly, (i.e. domestic savings increase mildly), then Jf > 0. Such a growth
regime is called exhilarationist. Whereas if savings differentials are large
compared to the net export effect of the profit share, then % < 0, and the
regime is called 'stagnationist' (Bhaduri and Marglin 1990). However,
when the lagged effects through investment also kick-in in the longer
term, the overall effect of profit share will depend on the relative magni-
tude of its positive direct effect on investment, the positive international
demand effect, and the negative effect on domestic consumption.
Finally substituting (9.7) and (9.9) into (9.2), we get accumulation as
a function of distribution:

_, (ax + a5r2 \r
gt = a0 + a5r0 + \ h + h + <*sTiJgt-i

+ \a2 + ax{\ + asr2) h z h y ^ (9.10)

Again here the effect of the profit share on accumulation, idgl/dir^), can
be decomposed to the direct positive effect of the profit share on accu-
mulation (the partial dgl/di:^^ a2), the positive international demand
effect ((a_-//dzt_!) x (dz^Jdnx^) x (dnx^j/dn^) = (a 3 (l + a5T2)fr2)/
(fri+^i)) anc * the negative domestic consumption effect ((dgl/dzt-i) x
{dzt_1ldi:t_l) = (-«i(l+a5T2)^2)/(^i+^i)). Depending on the relative mag-
nitudes of these effects, an increase in the profit share leads either to an
increase in accumulation, in which the regime of accumulation is profit-
led, or to a decrease, i.e. to a wage-led regime of accumulation.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Ozlem Onaran and Engelbert Stockhammer 211

Empirical m e t h o d o l o g y a n d hypotheses

The main methodological motivation behind this study is to model the


dynamic relationship between distribution, accumulation, capacity
utilization and employment considering b o t h lagged and contempora-

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
neous interactions within a systems approach that goes beyond the
limited framework of comparative statics. The existing empirical contri-
butions in the literature about the effect of distribution o n accumula-
tion, growth and employment do not address the issue of simultaneity

Table 9.2 Hypotheses

HI Demand-led - ^ < 0; - ^ < 0 Goods market determines labour


dz
labour market * market

H2 The effect of -^ = ? Is accumulation wage-led or


distribution on d7r profit-led?
growth
dir .
>0
H3 Reserve-army effect du Unemployment lowers wages

dir
H4 Imperfect wage >0 Productivity increases do not lead to
indexation dgx equivalent wage increases

H5 Technological j du
^- > 0 Productivity increases cause
unemployment %x unemployment

H6 Neoclassical du <0 Wage decrease increases


labour market dir employment

dgX Increase in wages leads to


H7 Substitution <0
dir substitution of labour with capital,
which in turn increases productivity

H8 Export-led >0 An increase in net exports leads to an


accumulation dnx increase in accumulation

H9 Profit-led exports ^ >0 Lower unit labour costs, i.e. a higher


profit share, increase international
competitiveness
dU
H10 Export-led <0 Exports increase the labour demand
employment dnx

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
212 Do Profits Affect Investment and Employment?

(for developed countries see Bowles and Boyer 1995; Gordon 1995a,
1995b; Hein and Kramer 1997; Bhaskar and Glyn 1995; Stockhammer
2004a, 2004b; Hein and Ochsen 2003; for developing countries see
Yenturk 1998; Onaran and Yenturk 2001; Seguino 1999; Sarkar 1992). To
overcome this shortcoming we employ a structural vector autoregres-

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
sion (SVAR) analysis which incorporates the contemporaneous interac-
tion as well as the lagged relationship. SVAR helps to capture the
complex simultaneous interaction between distribution, accumulation,
growth and employment, and the system aspect that is crucial to the
theoretical model.
The VAR model allows past values of all variables to influence present
values of any variable. Thus, results that are not in accordance with the
structural model outlined above are possible due to lagged effects. The
structural model provides the motivation and shapes the interaction of
the contemporaneous effects only. Thus it will be useful to summarize
the hypotheses to be explored empirically. The first five hypotheses
summarized in Table 9.2 follow directly from the model presented
above. Hypotheses six and seven are standard theses of the neoclassical
theory about the labour market. Hypotheses eight-ten are related to the
typical policy recommendations of the neoliberal structural adjustment
programmes, particularly in developing countries.
All the effects discussed above (except H2) are partial effects. The VAR
framework used does not distinguish between partial and total effects,
but gives the effects at different points in time. Only the estimated con-
temporaneous effects are clearly partial effects. We will interpret the
effects in the first two or three periods as partial effects in the short run,
but also discuss the longer term effects, wherever significant.

Summary of the results for developed countries

The advantages of VAR models, unfortunately, come with a disadvan-


tage: Given the lagged structure that incorporates the dynamic effects to
the estimation, it requires a long-enough time series. For a given length
of time-series data, the implication is a limit on the number of variables
that can be included into the system. This limit leads to differences in
the specifications for developing and developed countries.
In the case of the developed countries, the focus is on the interaction
between the labour market and the goods market. However, the foreign
trade is not modelled explicitly. Nevertheless, the estimated coefficients
and impulse responses of the profit share on capacity utilization (and of
course other variables) will include indirect effects via export demand.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Ozlem Onaran and Engelbert Stockhammer 213

Moreover, shocks to capacity utilization do include shocks coming from


fiscal policy, monetary policy and the foreign sector; in fact they include
all shocks to effective demand other than investment.
The SVAR system estimated consists of accumulation (of the business
sector), capacity utilization (output gap), the profit share (of the busi-

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
ness sector), the growth rate of labour productivity and the unemploy-
ment rate.2 The model is estimated for the periods 1970:1-1997:2,
1966:1-1997:2, and 1972:1-1997:1 for the UK, the USA and France
respectively, based on semi-annual data. The different periods are due to
data availability. The VAR is estimated with four lags.
The results of the response by various variables as a result to shocks to
the relevant variable in the hypothesis to be tested (impulse response
analysis) are summarized in Table 9.3. A more detailed technical discus-
sion can be found in Stockhammer and Onaran (2004).
The Keynesian-Kaleckian model performed fairly well, and in line with
the theoretical model; strong support is found for the demand-led labour
market hypothesis. The goods market variables play a strong role in deter-
mining unemployment, and shocks to accumulation as well as capacity
utilization have statistically significant negative effects on the rate of
unemployment. How long these effects last differs across countries.
Distribution seems to play little role in determining goods market
outcomes. None of the effects in the impulse responses were statistically
significant. The result may be due to offsetting effects of profitability
and demand, which would be consistent with the theoretical frame-
work. However, this also might be suggesting a theoretical challenge
about the role of profit share in investment models. We will discuss
more on this issue in the next section.
We found no evidence for the reserve-army effect. A shock to unem-
ployment has little or no effect on the profit share. Only in the UK was
there a positive effect, but not statistically significant. This finding is not
consistent with the literature, and may be due to the generous lags of
the dependent variable.
A shock to productivity growth has a statistically significant positive
effect on the profit share in the USA and France. So wages are not per-
fectly indexed. An innovation to labour productivity growth also has a
significant and upward impact on unemployment in all countries, and,
in fact, rather persistently so. Thus technological development does
not automatically generate demand, and can lead to technological
unemployment.
Weak or no evidence was found for both of the neoclassical labour
market hypotheses. In France and the UK, a shock to the profit share

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
214 Do Profits Affect Investment and Employment?

Table 9.3 Summary of impulse responses for the UK, the USA and France

UK USA France

HI Demand-led market Yes. Yes. Yes.


H < 0 and H < 0 g and z, Both sig. Both g and z sig. Both g and z sig.

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
or close to sig.
H2 Distribution-led No effect, Insig. Insig.
regimes Insig. (g profit-led) (g profit-led)
(z exhilarationist) (z stagnationist)
H3 Reserve-army Yes. No. No.
effect g > 0 insig. No effect
H4 Imperfect wage No. Yes. Yes.
indexation | g > 0 sig. for three Contemporary
periods effect sig.
contemporary
effect sig.
H5 Technological Yes. Yes. Yes.
unemployment Long Sig to 6 lags Sig to 4 lags
— >0 contemporary contemporary
effect sig. effect sig.
H6 Neoclassical labour No. Yes. No.
market % < 0 No effect But sig. only Insig.
after 7 periods
H7 Substitution No. No. No.
— <0 Insig. / no effect Insig. Insig.

Note: Sig = statistically significant.


Source: Stockhammer and Onaran (2004).

had basically no significant effect on unemployment. Only in the USA,


and only after seven periods, a shock to the profit share led to a decline
in the unemployment rate. Again, the profit share had no significant
negative effect on productivity. Thus the substitution hypothesis was
also rejected.

Summary of the results for developing countries

In the case of the developing countries, that is Korea and Turkey, the
effect of international trade on growth and employment on the one
hand, and the source of international competitiveness on the other
hand are important points of focus for policy analysis. Therefore, we not

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Ozlem Onaran and Engelbert Stockhammer 215

only explicitly model international demand, but also decompose the


effect of exports and imports, in order to be able to test widely accepted
mainstream policy assumptions regarding export orientation. It is par-
ticularly important to highlight the opposite effects of different export-
oriented growth strategies on distribution and employment: decreasing

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
wage share, low growth, low investment, low employment in Turkey
and increasing wage share, high growth, high investment, high employ-
ment in South Korea.
The inclusion of the international trade block came at the cost of
excluding the explicit modelling of productivity change. The effects of
changes in productivity can only be interpreted as exogenous shocks.
Moreover, due to data limitations, several proxies had to be used
instead of the variables in the theoretical model. Second, since the agri-
cultural sector follows a completely different pattern, particularly in
terms of labour demand, where unpaid family work and self-employment
is important, and can lead to significant rates of disguised unemploy-
ment, we estimate the model as a whole for the non-agricultural econ-
omy. However, then it is very hard to measure the non-agricultural
unemployment or employment rate due to problems in anticipating the
sector specific labour supply. As a result we simply use the employment
level in logarithms to model the labour market block. Accordingly, the
SVAR system estimated consists of investment/GDP instead of accumu-
lation, growth instead of capacity utilization, the profit share, and the
logarithm of employment (all variables for the non-agricultural sector).
The model is estimated for the periods 1972-2000 for Korea, and
1965-97 for Turkey, based on available annual data. The VAR is esti-
mated with two lags.
The results of the response by various variables as a result of shocks to
the relevant variable in the hypothesis to be tested (impulse response
analysis) are summarized in Table 9.4. A more detailed technical discus-
sion can be found in Onaran and Stockhammer (2005).
The results also show that demand is the main driving force behind
employment, and accumulation is an important component to enhance
the job-creation capacity of the economy. Moreover, the employment
regime is not profit-led, and quite on the contrary to the arguments of
neoclassical economics, it decreases with a decrease in the wage share. In
South Korea the wage-led employment pattern is more evident, whereas
in Turkey the cumulative negative effect dies away after five periods.
This can be explained by the effect of the profit share on demand. In
Turkey an innovation to the profit share creates a negative response of
the investment rate in the next period, and the shock continues for

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
216 Do Profits Affect Investment and Employment?

Table 9.4 Summary of impulse responses for South Korea and Turkey

South Korea Turkey

HI Demand-led labour Yes. Yes.


market g < 0 and g and z both sig. g and z both sig.

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
— <0
H2 Distribution-led g wage-led insig.
regimes z significantly (g slightly wage-led in
stagnationist the second period)
(z in the first two periods
significantly stagnationist)
H3 Reserve-army Yes. Yes.
effect % > 0 Sig. even in long run Sig for six periods
H6 Neoclassical labour No. No.
market % < 0 Contrarily, positive Contrarily, positive effect
effect for 4 periods
H8 Export-led Yes. No.
accumulation Sig. strong persistent Slightly significant only
after 6 periods
H9 Profit-led exports No. Yes.
Positive but insig. Sig. Persistent and
strong positive effect
H10 Export-led Yes. No.
employment Sig. persistent and Sig. persistent and
strong positive effect strong negative effect

Note: sig. = statistically significant.

another period, and then dies out without leading to any significant
improvement in investment. In South Korea an increase in the profit
share creates a strong and persistent negative effect o n accumulation.
Regarding the effects on capacity utilization, an increase in the profit
share is immediately transformed into a decline in growth, indicating a
stagnationist regime in b o t h countries in the short run. The recovery of
the growth rate is due to the improvements in exports. However, in
Turkey it takes rather long - three periods - for the positive effect of
increased exports to lead to a recovery, and in South Korea the recovery
does not take place at all.
Reflecting the crucial differences in the design of export-oriented
growth strategy in the two countries, in South Korea the response of the
investment rate to international competition is very strong and persist-
ent, whereas in Turkey the response hardly shows u p with a lag of three
years and is never too strong. Also in Turkey exports increase w h e n unit

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Ozlem Onaran and Engelbert Stockhammer 217

labour costs decline (that is, a positive response of exports to profit


share), and thus when domestic demand contracts. However in South
Korea, a shock to the profit share has no significant effect on exports.
Turkey's export growth based on low wages and increased use of existing
capacity rather than new investments proves to be unable to stimulate

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
investments, whereas in South Korea export competitiveness is the pri-
mary stimulus behind investment decisions of firms. In Turkey, invest-
ments are stimulated by domestic demand, whereas in South Korea
exports are even more important than domestic demand. In South
Korea, exports are a systematic target of industrial policy, and competi-
tiveness is based on improvements in productivity. The consequence of
this striking difference in the export-oriented growth strategies
shows up also in the labour demand. The response of employment to an
increase in exports is persistently negative in Turkey, whereas it is
strongly and persistently positive in South Korea. This result points at a
very important policy implication indicating that the increase in com-
petitiveness, which is maintained by low wages, does not transform into
higher employment. Another important implication of the results for
Turkey is that they provide counter-evidence to the expectations about
an increase in labour intensity of production following an increase in
export orientation.
Finally, although distribution does not immediately adjust to changes
in labour-market conditions in the model, the lagged effects are signifi-
cant and in the expected direction according to the reserve army theory.

Conclusions and challenges for future research

The results indicate that the Kaleckian model overall performs well;
estimations are mostly according to the predictions of the model,
although within large confidence intervals. The Keynesian and Kaleckian
hypotheses about the labour market are confirmed: Accumulation and
capacity utilization/growth have a strong impact on employment.
Goods market variables have a strong impact on unemployment and the
economy is driven by investment expenditures; accumulation also
impacts strongly upon capacity utilization. The neoclassical hypotheses
of the labour market are not validated. There is little evidence of
employment reacting to wages (profit share), and no evidence for sub-
stitution. The ineffectiveness of labour costs on employment does not
differ much between developed vs. developing countries; the only dif-
ference is that in the latter there is even a negative response of employ-
ment to a decline in the wage share. The findings also suggest that

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
218 Do Profits Affect Investment and Employment?

productivity growth does play an important role. It is not distributionally


neutral and causes unemployment.
However, the results also point at some challenges for the model. No
statistically significant effect of the 'profit share' was found on invest-
ment and growth in developed countries, as well as in one of the devel-

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
oping country cases, Turkey (apart from the slightly significant but very
small effect in the second period). There is basically no result in terms of
the Kaleckian distinction of wage-led vs. profit-led regimes for a wide
range of different countries, and it also is not easy to generalize that
developing countries tend to be more wage-led just based on the Korean
case. Are the international demand and profitability effects, on the one
hand, and the domestic demand effect, on the other hand, exactly off-
setting each other in all the other countries that we studied? Although
that is theoretically possible, it is not likely in the VAR setting, particu-
larly because there would be some inter-period differences in the way
the lagged effects operate through different channels, and it is unlikely
that there is no period where there is a significant effect. It is also inter-
esting that this is the case for many different countries.
The next question is whether the 'profit share' is an appropriate
measure for income distribution. At the conceptual level it is useful,
because it serves the dual task in the model as a proxy for wages in the
distribution and savings functions, and a proxy for profits in the invest-
ment function. However, there may be measurement problems. First
there is the issue of taxes. The savings differential through which the
profit share is expected to affect consumption, works through net
income, i.e. post-tax income, whereas the profit share measures pre-tax
income distribution. The same is true for the profit share effect on labour
demand. If there are significant changes in the tax wedge between post-
tax wages and gross compensation, the profit share might be a bad proxy.
However, since tax structures change slowly, it would be surprising, if this
problem dominated the VAR estimations. Second, profit is value added
minus labour compensation. Thus it includes the income of self-
employed, whereas wage payments to management are counted as
wages. Although these are important concerns, clearly the profit share is
statistically negatively (and significantly) correlated with real wage. Thus
the profit share variable is certainly not dominated by noise due to meas-
urement errors.
Finally, the results bring up the question whether the model is looking
at the right variables. Are other factors affecting investment, such as
expectations, financial structure, state policies and institutions, more
important? Although the wage-led accumulation regime scenario and

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Ozlem Onaran and Engelbert Stockhammer 219

the effect of demand on accumulation explain part of this story, there


certainly are more to that in explaining the striking difference in invest-
ment rates between these countries. Within the institutional and class
structures of these economies, there are many factors that determine
accumulation other than demand and distribution.

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
The limit of the VAR framework is that with generous lag structures
the number of variables that can be incorporated is restricted due to a
lack of degrees of freedom. However, the limits are also related to the dif-
ficulties in quantifying institutional structures. For example, within a
business environment created by active state policies there was a virtu-
ous cycle of an increasing wage share, high investment, high productiv-
ity and high growth in South Korea, as opposed to the Turkish case with
the vicious circle of a decreasing wage share, low growth, low invest-
ment and low productivity. Unfortunately, it is difficult to model
and test the complicated role of the state's economic and specifically
industrial policy by adding simple and measurable variables. State
expenditures would be too coarse a measure, because the policy aspect
lies in the details of these budgetary expenditures that even go beyond
the composition between current vs. investment expenditures, such as
subsidy and incentive structures. Similarly the multi-foreign exchange
system for different industries and even firms in Korea cannot simply be
captured by the rate of depreciation of the official exchange rate. Such
complexities make it hard to carry the institutional information into the
time series framework.
The incorporation of the financial would also improve the model.
Unfortunately, not only the limitations of SVAR, but also limitations
regarding the data to measure these effects related to financial variables
and expectations, leave these crucial aspects unexplored. Real interest
rates were completely insignificant in the estimations; obviously they
were unable to capture the full complexity of the structural change in
the financial system and the role of the institutions for the cases we
studied.
We also experimented with the specifications including inflation and
the change in inflation to reflect the macroeconomic environment.
Again, no major changes in the impulse responses occurred, though,
unsurprisingly, confidence intervals increased.
The complicated link between the wage share and investments could
to some extent be uncovered with a model that decomposes the wage
share into real wages and productivity. Such an analysis could provide
an insight in how wage bargaining, investments and technological
change interfere.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
220 Do Profits Affect Investment and Employment?

Policy implications

Important policy implications follow from the results. If we turn around


the result about the ineffectiveness of distribution on accumulation
and employment, we can conclude that actual economies are 'not

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
profit-led'. Thus, a pro-capital incomes policy is neither a necessary nor
a sufficient condition to achieve higher accumulation and growth. On
the contrary, the decline in domestic demand can have detrimental
effects on the long term growth potential of the economy.
Secondly, demand is the driving force behind employment. The
increase in competitiveness, which is maintained via wage suppression,
does not necessarily transform into higher employment. The limits in
creating employment via low wages and a growth regime based on the
use of existing capacity rather than new investments point out the
significance of active policies to stimulate accumulation. This alterna-
tive line of economic policy necessitates a different perspective on inter-
national competitiveness, which is based on enhancing productivity.
Moreover, if distribution is neutral with respect to investment, then
there is room for egalitarian redistribution policies, without harming
the growth potential of the economy.
In terms of the development agenda, the responses of accumulation,
growth and employment to distribution are suggestive in explaining
some crucial aspects of the mechanism behind the inability of orthodox,
market-based, export-oriented growth strategies relying on decreasing
wage shares to stimulate accumulation and employment. The centrality
of demand and the inability of low wages as a policy tool to stimulate
investment point at important policy lessons for the design of an alter-
native export-oriented growth strategy. Clearly, institutional settings
and state policies matter more than distribution in achieving a high,
investment and productivity-led export performance. Obviously,
the neoliberal reader could think that state policies were at the heart
of the structural problems of the Korean economy, that resulted in over-
investment in sectors with falling profit rates. However, these state poli-
cies led to unforeseen growth rates, and the need of the Asian model was
to revise its state industrial policies, and not to abolish them. Speculating
more on the design of the relevant industrial policy tools is beyond the
scope of this chapter, however, there are more policy issues that should
be addressed.
Are such policies available simultaneously to all developing countries
trying to compete for a limited global market? Obviously that brings in
the questions about the design of a new international system targeting

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Ozlem Onaran and Engelbert Stockhammer 221

at coordinated and expansionary macroeconomic policies, which would


benefit not only the developing but also developed countries. Although
the existing balance of power relations between the multinationals
dominating the world markets and the working masses of the world, an
alternative international macroeconomic policy seems quite unachiev-

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
able even in the context of EU, which is claiming to be not only an eco-
nomic but also a political union. The neo liberal policies representing
the interests of the firms, preventing any coordinated policy, which
could target demand management is hiding behind the discourse of
market efficiency and anti-inflation targets. Although the global chorus
of neoliberalism ranging from academics to central bank experts is
repeating the need for tight fiscal and monetary policy, we still conclude
by repeating the need for a coordinated international expansionary
macroeconomic policy.

Notes
1 See Blecker (1999) for a discussion of the extension of the model to the open
economy and Blecker (2002) for a review of other possible extensions to neo-
Kaleckian models.
2 A series of tests was performed to ensure the robustness of the results. First, it
was checked whether the results were sensitive to variable specification. The
profit share of the total economy was used instead of the profit share of the
business sector. The employment share (employment divided by working-age
population) was used instead of the unemployment rate. Instead of the output
gap, detrended capital productivity and GDP growth were used. In neither
case were there major changes in the results.

References
Bhaduri, A. and Marglin, S. (1990) 'Unemployment and the Real Wage: The
Economic Basis for Contesting Political Ideologies', Cambridge Journal of
Economics, 14: 375-93.
Bhaskar, V. and Glyn, A. (1995) 'Investment and Profitability: The Evidence from
the Advanced Countries', in G. Epstein and H. Gintis (eds), Macroeconomic
Policy After the Conservative Era. Studies in Investment, Saving and Finance
(Cambridge: Cambridge University Press): 175-96.
Blecker, R. (1989) 'International Competition, Income Distribution and
Economic Growth', Cambridge Journal of Economics, 13: 395-412.
Blecker, R. (1999) 'Kaleckian Macromodels for Open Economies', in J. Deprez and
J.T. Harvey (eds), Foundations of International Economics: Post Keynesian
Perspectives (London, New York: Routledge): 116-49.
Blecker, R. (2002) 'Distribution, Demand and Growth in neo-Kaleckian Macro-
Models', in M. Setterfield (ed.), The Economics of Demand-led Growth
(Cheltenham, UK: Edward Elgar): 129-52.
Bowles, S. and Boyer, R. (1995) 'Wages, Aggregate Demand, and Employment in
an Open Economy: An Empirical Investigation', in G. Epstein and H. Gintis

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
222 Do Profits Affect Investment and Employment?

(eds), Macroeconomic Policy After the Conservative Era. Studies in Investment,


Saving and Finance (Cambridge: Cambridge University Press): 143-71.
Dutt, A. (1984) 'Stagnation, Income Distribution and Monopoly Power',
Cambridge Journal of Economics, 8: 25-40.
Gordon, D. (1995a) 'Growth Distribution, and the Rules of the Game: Social
Structuralist Macro Foundations for a Democratic Economic Policy', in

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
G. Epstein and H. Gintis (eds), Macroeconomic Policy After the Conservative Era.
Studies in Investment, Saving and Finance (Cambridge: Cambridge University
Press): 335-83.
Gordon, D. (1995b) 'Putting the Horse (back) Before the Cart: Disentangling the
Macro Relationship between Investment and Saving', in G. Epstein and H. Gintis
(eds), Macroeconomic Policy After the Conservative Era. Studies in Investment, Saving
and Finance (Cambridge: Cambridge University Press) 57-108.
Hein, E. and Kramer, H. (1997) 'Income Shares and Capital Formation: Patterns of
Recent Developments', Journal of Income Distribution, 7(1): 5-28.
Hein, E. and Ochsen, C. (2003) 'Regimes of Interest Rates, Income Shares, Savings
and Investment: A Kaleckian Model and Empirical Estimations for Some
Advanced OECD-Economies', Metroeconomica, 54(4): 404-33.
Marglin, S. and Bhaduri, A. (1990) 'Profit Squeeze and Keynesian Theory', in
S. Marglin and J. Schor (eds), The Golden Age of Capitalism. Reinterpreting the
Postwar Experience (Oxford: Clarendon Press): 153-86.
Onaran, 6. and Stockhammer, E. (2005) 'Two Different Export-oriented Growth
Strategies: Accumulation and Distribution in Turkey and South Korea', Emerging
Markets Finance and Trade, 41(1): 65-89.
Onaran, O. and Yenturk, N. (2001) 'Do Low Wages Stimulate Investments? An
Analysis of the Relationship between Distribution and Investments in Turkish
Manufacturing Industry', International Review of Applied Economics, 15(4):
359-74.
Rowthorn, R. (1981) 'Demand, Real Wages and Economic Growth', Thames Papers
in Political Economy. Autumn: 1-39, reprinted in Studi Economici, 1982 (18):
3-54.
Sarkar, P. (1992) 'Industrial Growth and Income Inequality: An Examination of
"Stagnationism" with Special Reference to India', Journal of Quantitative
Economics, 8(1): 125-38.
Seguino, S. (1999) 'The Investment Function Revisited: Disciplining Capital in
South Korea', Journal of Post Keynesian Economics, 22(2): 313-38.
Stockhammer, E. (2004a) The Rise of Unemployment in Europe. A Keynesian
Approach (Northampton: Edward Elgar).
Stockhammer, E. (2004b) 'Explaining European Unemployment: Testing the
NAIRU Theory and a Keynesian Approach', International Review of Applied
Economics, 18(1): 3-24.
Stockhammer, E. and Onaran, O. (2004) 'Accumulation, Distribution and
Employment: A Structural VAR Approach to a Kaleckian Macro Model',
Structural Change and Economic Dynamics, 15: 421-47.
Taylor, L. (1996) 'Stimulating Global Employment Growth', in J. Eatwell (ed.)
Global Unemployment. Loss ofJobs in the 90s (Armonk, NY: M. E. Sharpe), 147-72.
Yenturk, N. (1998) 'Ajustement et accumulation: la Turquie', Canadian Journal of
Development Studies, xix (1): 55-77.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
10
Class Conflict and the Cambridge
Theory of Income Distribution*

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Thomas I. Palley

Why the interest in income distribution?


Understanding the determination of income distribution is a central
concern of economics. Ricardo regarded it as the central issue of eco-
nomics, writing that '(determining) the laws which regulate this distri-
bution is the principal problem in political economy' (Ricardo 1821: 5).
Income distribution is clearly an important social and political concern,
having ramifications for fairness and the social and political stability of
society. How we explain income distribution also has major implica-
tions for how we understand the economy, thereby impacting attitudes
and approaches to policy.
The dominant approach to income distribution is marginal produc-
tivity theory, which emphasizes demand and supply for factors of pro-
duction. The determination of income distribution is therefore part of
the working of the price system. Distribution is determined by competi-
tive market forces that ensure factors are paid their contribution to pro-
duction, and the process of determining factor prices in turn links the
determination of factor employment. The theoretical implication is that
there is no ethical basis for intervention as competitive markets bar
exploitation and ensure factors are paid what they are worth. Indeed,
intervention stands to distort the market process, creating unemploy-
ment and reducing welfare.
The marginal product approach to income distribution sanitizes the
economy of conflict and the need for intervention. Other theories of

* This chapter originally appeared as 'Class Conflict and the Cambridge Theory of
Distribution,' in B. Gibson (ed.), The Economics of Joan Robinson: A Centennial
Celebration, Cheltenham: Edward Elgar, 2005. My thanks to Edward Elgar for
permission to reprint the material.

223

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
224 Class Conflict and the Cambridge Theory of Income Distribution

income distribution have dramatically different implications, yet they


are neglected and even suppressed within modern economics. Why that
is so is an interesting question in itself. The current chapter expands the
Cambridge theory of income distribution, which is a Keynesian theory
in which income distribution is determined macroeconomically by the

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
forces of aggregate demand (AD) rather than the price system. The
chapter expands the Cambridge approach to incorporate labour market
conflict.
The Cambridge theory was developed by Kaldor (1956), and its key
insight concerns the role of aggregate demand (AD) in determining
income distribution. The core idea is that AD needs to adjust to the level
of full employment output, and this is accomplished by adjustment in
the pattern of income distribution. Pasinetti (1962) subsequently intro-
duced class into the analysis, distinguishing between capitalists' and
workers' income shares and the AD impact of their differential propen-
sities to save. However, though adding class to the determination of
income distribution, Pasinetti's model is strangely devoid of class con-
flict in the traditional Marxian sense - that is, class conflict centred on
the labour market and bargaining strength. In Pasinetti's framework
class enters through behavioural propensities, with the propensity to
save differing across classes. This chapter joins this behavioural propen-
sity channel with traditional labour market class conflict.
The balance of the chapter is as follows. The next section describes the
sociological structure of the economy and its relation to income distri-
bution. I then recapitulate the Cambridge Post-Keynesian (CPK) theory
of income distribution, and describe the Kaleckian extension of the CPK
approach that includes less than full employment outcomes. The fifth
section incorporates labour-market class conflict into the extended
Kaleckian CPK model, followed by a comparative statics and stability
analysis. The issue of ownership and its relation to income distribution
is introduced, and a final section concludes the chapter.

Structure of the model

The key analytic contribution of the chapter is to distinguish the


income distribution effects of labour market conflict from those of
product market competition. Kaleckians have always recognized the sig-
nificance of both labour market conflict and product market competi-
tion, but these two forces have been lumped together under the 'degree
of monopoly'. The logic by which the chapter disentangles labour mar-
ket and product competition effects is illustrated in Figure 10.1, which

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Thomas I. Palley 225

National
Income

ir

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
i f v
Wages Profits

i r V

1
1 r V v \r

Workers' Manager-capitalists'
wages Distributed Retained
wages

V v
Dividends paid Dividends paid
to workers to manager-capitalist

Figure 10.1 The national income tree

shows the national income tree. National income consists of wages and
profits. Wages are in turn paid to workers and managers. The latter are
also identified as capitalists. Profits are partly retained by firms, and
partly distributed as dividends to shareholders. Dividends are in turn
shared between workers, who have part-ownership, and manager-
capitalists who own the rest of the firm.1 The chapter treats the division
of income between wages and profits as being primarily influenced by
the extent of product market competition, while the division of the
wage bill is determined by labour market bargaining power.
The model makes several important theoretical innovations. First, it
introduces managerial pay, an area that has taken on great significance
with the Chief Executive Officer (CEO) pay and share-option explosion
of the last 20 years. Second, the concern with distribution of the wage
bill introduces a second margin for income distribution effects, supple-
menting the traditional Cambridge focus on the profit share. Third, the
presence of a wage distribution channel means that the economy can
simultaneously exhibit 'wage-led' and 'profit-led' tendencies.2 An econ-
omy is defined as wage-led if improved income distribution (a lower

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
226 Class Conflict and the Cambridge Theory of Income Distribution

profit share) raises AD: it is profit-led if improved income distribution


lowers AD. In standard Kaleckian models of growth and income
distribution the economy is either wage-led or profit-led. In the current
paper, an economy can simultaneously have both properties.
This can be seen as follows. Shifts in the wage distribution from man-

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
agers to workers, holding the profit share constant, can increase AD so
that the economy is wage-led. At the same time, increases in the profit
share, holding the wage distribution constant, can increase investment
so that the economy simultaneously displays profit-led characteristics.
This combination may best describe the US economy.3
At the policy level, the model identifies several margins on the
income tree where policy can intervene. A major policy recommenda-
tion is that progressive policy focus on the distribution of the wage bill
rather than the profit share. Redistributing the wage bill is always expan-
sionary, whereas redistributing the profit share can be contractionary if
the economy is profit-led. Second, the model offers insight into the
effects of changes in the business sectors' profit retention ratio.

The CPK model revisited

The CPK approach to growth and distribution was pioneered by


Kaldor (1956). The standard short-run Kaleckian macroeconomic
model (derived from Kalecki 1942) is characterized by three features:
(1) income distribution is exogenously given, (2) income distribution
influences AD, and (3) the level of output then adjusts to equal the level
of AD.4 Putting the pieces together, the pattern of income distribution
therefore influences the short-run level of equilibrium income.
Kaldor (1956) reversed this reasoning. Instead of assuming income
distribution to be exogenous, Kaldor took output as exogenously given
and equal to its full employment level. Given that AD must still equal
output, Kaldor argued that in the long run income distribution would
adjust. Rather than having output adjust to income distribution, as in
the short-run Kaleckian model, income distribution adjusts to ensure a
level of AD consistent with full employment income.
Assuming a positive propensity to save out of profit income and no
saving out of wage income, this gives rise to the famous Cambridge
equations for the profit share and profit rate given by:

P/Y = I/sPY (10.1a)


P/K = I/sPK (10.1b)

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Thomas I. Palley 227

P/Y=l/s„Y

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Investment share

Figure 10.2 The Kaldor (1956) model

where P = profits, Y = output, I = investment spending, K = capital


stock, and sP = propensity to save out of profits. These equations consti-
tute investment-saving balance (IS) equations in which income distri-
bution has adjusted to ensure AD equals full-employment output. The
Kaldor model is illustrated in Figure 10.2 which shows the profit share as
a function of the investment share. The important feature of Kaldor's
analysis is that it examines the special case where the investment share
is consistent with full employment output. This investment share is
denoted I/Y* in Figure 10.2.
Pasinetti (1962) extended Kaldor's model by introducing two social
classes - capitalists and workers. Like Kaldor, he too focused on the case
of full-employment steady-state growth. The key analytic contribution
was to give a class structure to income distribution and savings behav-
iour. The assumptions of the model are that capitalists receive just profit
income, workers receive b o t h profit and wage income, and capitalists
have a higher propensity to save t h a n do workers. 5 Given these condi-
tions, Pasinetti shows that the functional distribution of income and the
profit rate depended exclusively on capitalists' propensity to save and
the level of full-employment investment spending. 6 The equilibrium
conditions t h e n get restated as:

P/Y = I/sKY (10.2a)

P/K = I/sKK (10.2b)

where sK = capitalists' propensity.


Pasinetti's result has been the subject of significant attention. The
introduction of government saving leaves the result unchanged (Dalziel

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
228 Class Conflict and the Cambridge Theory of Income Distribution

1991); so too does the introduction of life-cycle saving (Baranzini 1982).


However, the introduction of financial factors changes the result, and
workers' propensity to save matters for steady-state income distribution.
Palley (1996a, 2002a) shows that in a world with bank-created inside
debt (that is an endogenous money world) workers' saving propensity

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
matters for income distribution. This is because they pay interest on
bank loans, which are costless to produce. This interest increases capi-
talists' incomes, necessitating a reduction in the profit share to maintain
a full-employment investment-saving balance. Interestingly, the result
does not hold in a loanable funds world in which capitalists make loans
in the form of real resources that are transferred to workers. Palley
(1997) also shows that in a model with money and an inflation tax,
workers' saving also matters because they are taxed disproportionately
on their money holdings.

The Kaleckian extension of the CPK model

The Kaldor-Pasinetti approach analyses the determination of income


distribution under the assumption of full employment. This is a
strangely un-Keynesian assumption, since Keynes (1936) took pains to
explain in The General Theory that he thought full employment was a
special case of classical economics.
Several authors (Rowthorn 1982; Dutt 1984, 1990; Lavoie 1995) have
contributed to development of a more general Kaleckian model of
growth and income distribution that extends the CPK model. The
important contribution of these authors is to introduce less than full
employment conditions. These extended models involve adding an
investment function equation, and a mark-up or real-wage equation.
The mark-up and real-wage equations perform identical functions -
namely determining the profit share. This last feature reveals how
Kaleckian models have difficulty distinguishing the income distribution
impacts of labour market conflict from those of product market compe-
tition. Labour market conflict and product market competition are con-
flated and work through the mark-up, which impacts the price level, the
real wage, and the profit share.
The logic of these models is easily illustrated. Let price be a mark-up
over average wage costs and given by:

p = [l + m]w/a (10.3)

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Thomas I. Palley 229

where p = price, m = mark-up, w = nominal wage and a = constant


average product of labour. In this case, the profit share can be shown
to be:

P/Y =m/[l + m] (10.4)

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Multiplying by the output-capital ratio yields:

P/K = mk/[l + m] (10.5)

where k = output-capital ratio, which ratio is a positive function of the


rate of capacity utilization. In addition, if the mark-up is assumed to be
a positive function of capacity utilization and the exogenously given
degree of product market competition, the mark-up schedule can be
written as:

P/K = m(u, c)k(u)/[l + m(u, c)] mu>0, rac<0, ku>0 (10.6)

where c = degree of product market competition.


To this mark-up schedule is added a Kaleckian investment equation
given by:

I/K = a0 + axu + a2P/K + a3P/Y av alt a3, a4>0 (10.7)

where u = capacity utilization rate. Investment spending is therefore


assumed to be a positive function of capacity utilization (a x > 0), the
profit rate (a2 > 0), and the profit share (a 3 >0). There has been much
discussion of what constitutes appropriate specification of the invest-
ment function (see Lavoie 1995). There are many drivers influencing
investment spending. These include capacity expansion, cost reduction,
and technology adoption. The Kaleckian equation incorporates vari-
ables that legitimately influence all of these drivers. Capacity utilization
is directly relevant to the need for capacity expansion; the profit rate
affects firms' willingness to adopt new technologies; and the profit share
can be thought of proxying for cash-flow effects that have been found to
be empirically important in microeconomic firm-level based studies
(Fazzari, Hubbard and Petersen 1988).
Substituting equation (10.4) into equation (10.7) then yields:

I/K = a0 + a\u + a2P/K + a3m(u, c)/[l + m(u, c)] (10.8)

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
230 Class Conflict and the Cambridge Theory of Income Distribution

Now substituting (10.8) into (10.2b) yields:

P/K = {a0 + axu + a3m(u, c)/[l + m(u, c)]}/[sk - a2] (10.9)

where sK = capitalists' propensity to save. Equation (10.9) is a reformu-

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
lated IS curve in which investment is endogenous and depends on
capacity utilization, the profit rate and the profit share. The full Post-
Keynesian-Kaleckian growth model consists of equations (10.6) and
(10.9). Equation (10.6) is a microeconomic profit-rate equation that is
derived from the pricing behaviour and cost structure of firms. Equation
(10.9) is the IS schedule. Together equations (10.6) and (10.9) jointly
determine capacity utilization, u, and the profit rate, P/K.
The model is illustrated in Figure 10.3 which depicts the model in
[u, P/K\ space. Equation (10.6), the profit-rate equation, is denoted by
MM. Equation (10.9), the investment-saving balance equation, is
denoted by IS. The slope of the IS equation in [u, P/K\ space is in princi-
pal ambiguous and Figure 10.3 is drawn under the assumption that it is
positively sloped. A positive slope requires sK > a2, which is the more
likely case given that the link between investment and capacity utiliza-
tion is empirically weak.7 The mark-up equation, depicted by the MM
schedule, is drawn as flatter than the IS schedule reflecting the fact that
empirical evidence suggests the mark-up is fairly stable over the business
cycle.8 The intersection of the IS and MM schedules corresponds to
a [u, P/K\ combination for which the goods market clears (that is,
investment-saving balance holds), and for which the profit share and
profit rate are consistent with the microeconomic pricing decisions of

IS
MM

| P*IK

u
~ Capacity utilization
Figure 10.3 The Kaldor-Pasinetti-Kalecki model

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Thomas I. Palley 231

firms. These schedules jointly determine P/K and u. This in turn allows
determination of I/K, K/Y and m. Determination of m determines P/Y,
which then allows determination of I/Y.
Figure 10.3 can then be manipulated to generate some standard
Kaleckian comparative static results.9 An increase in capitalists' propen-

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
sity to save shifts the IS schedule left, lowering the equilibrium profit
rate and rate of capacity utilization. An exogenous decrease in the level
of competition increases the mark-up and shifts the MM schedule up,
lowering the equilibrium profit rate and rate of capacity utilization.
In principal, the financial factors alluded to earlier, concerning worker
borrowing of inside bank money and the inflation tax, can also be
included. These factors affect the IS schedule by impacting overall sav-
ing, and they allow financial factors to impact the determination of the
equilibrium profit rate and rate of capacity utilization. An increase in
worker bank borrowing shifts the steady state IS schedule down, and
lowers the equilibrium profit rate and rate of capacity utilization. The
reasoning is that workers pay interest on their debts that is distributed to
capitalists who own the banks. This raises aggregate saving because of
capitalists' higher propensity to save, necessitating a reduction in the
profit rate which lowers investment and capacity utilization.

Bringing class back to Cambridge

Though having a class structure embedded in aggregate demand (the


Pasinetti contribution), class conflict in the Kaleckian model is opaque.
This is because it is made to operate through the mark-up, which in turn
depends on the rate of capacity utilization. However, traditionally, class
conflict over income distribution has been thought of as operating
through the labour market.
One way of introducing labour market conflict is through an Okun's
law relationship, whereby there is a monotonic negative relationship
between capacity utilization and unemployment. In this case, the rate
of capacity utilization can be thought of as proxying for the unemploy-
ment rate, so that labour market class conflict operates indirectly
through the rate of capacity utilization. This is the approach adopted
by Dutt (1992) in a model in which workers' target real wage is impacted
by the rate of unemployment.
However, this approach effectively conflates capacity utilization and
unemployment rate impacts. In effect, worker-firm conflict over wages
in the labour market is treated as identical to firm-firm competition
over the mark-up in product markets. This is a problem that has always

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
232 Class Conflict and the Cambridge Theory of Income Distribution

been present in the Kaleckian model. Product market competition and


labour conflict are distinct economic forces that have differential
impacts and work through different channels.
The distinction between the profit-wage functional distribution of
income and the distribution of wage income, identified in Figure 10.1,

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
provides an avenue for distinguishing between these two effects. The
model that is developed below argues that inter-firm competition affects
the mark-up and the profit share, while labour market competition
affects the distribution of the wage bill across workers and managers.
Modelling this requires respecifying the IS relation so that it includes
managerial pay. The mark-up side of the model, as represented by the
MM schedule, remains unchanged. Analytically, the effect is to intro-
duce labour market conflict into the model via the IS schedule. The logic
is that labour market conflict affects the wage distribution, and the wage
distribution in turn impacts AD.
Finally, in addition to decomposing the wage bill into wages paid to
workers and manager capitalists, the model also introduces profit reten-
tions as a way of financing investment. Such retentions have firms sav-
ing on their own behalf to finance investment, and it can have
important macroeconomic implications - yet, it has traditionally been
ignored in Cambridge distribution theory analysis.
Aggregate income, wages, profit and ownership satisfy the following
adding-up constraints:

Y=W +P (10.10a)
Ww + WK = W (10.10b)
PW + PK + R = P (10.10c)
zw + zK = 1 (lO.lOd)

where W = wage bill, Ww = wage bill paid to workers, WK = wage bill


paid to manager-capitalists, Pw = profits attributable to workers,
PK = profits attributable to manager-capitalists, R = corporate retained
profits, zw = workers' ownership share, and zK = manager-capitalists'
ownership share. Profits distributed to workers and manager-capitalists
are given by:

Pw = zw[P-R] (10.11a)
PK = zK[P-R] (10.11b)

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Thomas I. Palley 233

Note that ownership of the capital stock has a critical impact on


distribution by affecting the distribution of profit. This is a feature that
has also been ignored in Cambridge models of income distribution, and
it is an issue that is discussed further below.
To these accounting relations is now added behavioural content. First,

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
the ratio of workers' wage bill to that of manager capitalists is given by:

WW/WK = y (10.12)

y is treated as parametric for purposes of comparative static analysis. In


practice, this ratio depends on the state of technology which determines
the ratio of non-supervisory to supervisory labour.10 It also depends on
bargaining power, union density, workers' militancy, labour market
policies concerning employee rights at work, minimum wage laws,
unemployment insurance compensation, and the scope of the social
safety net. The effect of this distributive parameter is to create a channel
for labour market distributional impacts that is separate and distinct
from the impact of product market competition on the mark-up.
The second behavioural relationship concerns firms' profit retentions.
This is assumed to be governed by:

R = $(t,a)P 0 < p < l , (3f>0 (10.13)

where p = retained profit ratio, t = dividend tax rate, and a = exogenous


shift factor. The level of retentions is a positive function of profits. In
addition, the retained profit ratio is positively related to the dividend
tax rate, with a higher tax encouraging firms to hold on to profits.
The IS schedule for the expanded model is then given by:

sw[Ww + Pw] + sK[WK + PK]+R = I (10.14)

where sw = workers' saving propensity, and R = level of profit retentions.


Using the relations given by (10.10a)-(10.10d), (lO.lla)-(lO.llb),
(10.12) and (10.13), this IS schedule can be restated as:

I/Y= a 0 + nxP/Y (10.15a)


I/K = n0k(u) + VL^/K (10.15b)

where fl0 = [swy+sK]/[l+y] andfli = {sw[l-zK] + sKzK + [1 - s ^ l - z ^ ] -


SKZKW*, a) - [s^r+sJ/[H-y]}. The term [ 1 - s ^ [ 1 - z ^ - s ^ z J attaching

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
234 Class Conflict and the Cambridge Theory of Income Distribution

to /3(£, a) is the net increase in aggregate saving coming from an increase


in retained profit. Retained profits increase corporate saving, but they
diminish household sector saving by reducing distributed profit
income. Substituting equation (10.8), determining the I/K ratio, into
(10.15b) then yields an IS schedule in [u, P/K\ space given by:

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
P/K = {a0 + axu + a3m(u, c)]/[l + m(u, c)] - Ci0k(u)}/[Ci0 - a2]
(10.16)

The critical feature of this IS curve is that it embeds the labour market con-
flict parameter y, which affects AD. This is consistent with the logic of class
conflict affecting AD, and is distinct from product market competition
effects on the mark-up and profit share. Note, however, that these product
market effects still affect AD through the term a3m{u, c)/[l+m(u/ c)]. This is
because investment spending, per equation (10.7), is assumed to be posi-
tively related to the profit share. The slope of the IS schedule is ambiguous,
and more likely to be negatively sloped if investment is very sensitive to
the profit rate (that is, a2 is large).
The full model now consists of equation (10.16), describing the IS
schedule, and equation (10.6) describing the MM schedule. The general
reduced forms for these equations are given by:

P/K = M(u,c) (10.17a)


P/K = I(a0, alf a2, a3, u, c, sw, sK, y , zK, ~t, a) (10.17b)

Signs above functional arguments are signs of partial derivatives. The


graphical analogue of the model, under the assumption of a negatively
sloped IS schedule, is similar to Figure 10.3.

Stability analysis, comparative statics and policy

The stability of the model is analysed in the Appendix for the case where
the IS is positively sloped in [u, P/K] space. The model can be either stable
or unstable. Stability is impacted by whether the economy is wage-led or
profit-led (see Bhaduri and Marglin 1990). In the profit-led case, capacity
utilization increases when the profit rate is above that needed for goods
market equilibrium. In the wage-led case, capacity utilization decreases
when the profit rate is above that needed for goods market equilibrium.
As shown in the Appendix, stability also depends on the relative slopes of
the IS (goods market) and MM (mark-up) equilibrium schedules.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Thomas I. Palley 235

Comparative statics yields the following conclusions. An exogenous


increase in investment, represented by an increase in the coefficient a0,
shifts the IS schedule up. Both the profit rate and capacity utilization
rate increase. This is consistent with the standard Keynesian construc-
tion of the macroeconomy. Increases in the coefficients alf a2, a3, all of

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
which increase the sensitivity of investment, also shift the IS up and
result in a higher profit rate and higher rate of capacity utilization.
Increases in the propensity to save of capitalists or workers, sw and sK,
shift the IS schedule down. This lowers the profit rate and rate of capac-
ity utilization. Increased saving is therefore contractionary, the standard
Keynesian result.
Figure 10.4 illustrates the case of an exogenous increase in the level of
product market monopoly power (that is, a decrease in c) that raises the
mark-up, perhaps brought about by a merger wave. This shifts up both
the MM and IS schedules, so that the effect on the profit rate and capac-
ity utilization is ambiguous. Note, the IS shifts up because investment is
a positive function of the profit share. If this profit share effect on
investment is weak (that is, a3 is small), the upward shift of the IS sched-
ule will tend to be small, and it is more likely that the profit rate and
capacity utilization fall. This corresponds to a wage-led construction of
the economy, in which worsening of the functional distribution of
income lowers AD and economic activity. Alternatively, if the profit
share effect on investment is strong (that is, a3 is large), then the IS shift
will be large and it is more likely that the profit rate and capacity uti-
lization will rise. This corresponds to a profit-led construction of the

MM'
B MM

2 P*/K
CL

Li" Capacity utilization

Figure 10.4 Ambiguous effect of an exogenous increase in the degree of


monopoly power in the Kaldor-Pasinetti-Kalecki model

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
236 Class Conflict and the Cambridge Theory of Income Distribution

Q: J S ......IS'
CD
/ / M M

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
03

O P*/K

-•
Capacity utilization

Figure 10.5 Expansionary effect of a redistribution of the wage bill to workers in


the Kaldor-Pasinetti-Kalecki model

economy, in which worsening of the functional distribution of income


raises AD and economic activity by stimulating investment.
Figure 10.5 illustrates the effect in worker bargaining power which
raises y and shifts the wage distribution towards workers. This shifts
right the IS schedule, leading to an unambiguous increase in the profit
rate and capacity utilization.11 Distinguishing the wage share from the
distribution of wages is a critical policy distinction. Improving the dis-
tribution of the wage bill is always expansionary. This is because it posi-
tively impacts consumption, but has no impact on investment since the
profit share and profit rate are left unchanged. As such, improving the
wage distribution should be the principal focus of progressive macro-
economic policy. In contrast, increasing the wage share can be contrac-
tionary if the economy is profit-led in character. Focusing on the wage
share therefore constitutes more complicated policy.
Finally, from a theoretical perspective, distinguishing between the
wage share and the distribution of the wage bill allows the economy
simultaneously to exhibit wage-led and profit-led characteristics. This
contrasts with existing constructions of the Cambridge growth and dis-
tribution model which impose an either or condition. The labour con-
flict channel, operating through the wage distribution, is always
wage-led - so that shifts in the wage bill towards workers are expansion-
ary. However, investment may be profit-led, exhibiting a strong depend-
ence on the profit share - so that shifts in the functional distribution
from wages to profits raise investment and economic activity. This dual
construction helps make sense of developments in the US economy over

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Thomas I. Palley 237

the last 25 years. Changes in the distribution of the wage bill, exempli-
fied by the explosion of CEO pay, have been contractionary.12 Side-by-
side, shifts in the functional distribution of income towards profits may
have been expansionary since there is some evidence that investment
spending in the USA is exhilarationist - that is positively influenced by

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
the profit share (Gordon 1995).
Increasing capitalists' ownership share, zK, shifts the IS schedule down
so that the profit rate and capacity utilization fall unambiguously. This
suggests that measures to change the distribution of wealth in a pro-
gressive direction, through wealth or inheritance taxes, may be expan-
sionary. If saving falls in response to such taxes, this would make them
even more expansionary. However, all bets are off if investment also falls
in response to wealth and inheritance taxes. Then, they could be
counter-productive and lower capacity utilization and growth. Lastly,
consideration of ownership shares also suggests why worker pension
plans can exert a long-run favourable impact in that they shift owner-
ship and profit income over to workers, thereby having a long-run
favourable impact on AD and the economy.
A final experiment concerns dividend taxes, t, and exogenous changes
in firms' decisions about retained profit, a. This experiment has
implications for the debate over reducing double taxation of dividends.
Increases in the dividend pay-out, resulting from lower taxes on divi-
dends or a change in firms' decisions, shift the IS schedule up. They are
therefore expansionary, raising the profit rate and capacity utilization.
The economic logic of this effect is easily understood in terms of equa-
tion (10.14). Increased dividend payouts reduce firms' saving by a full
dollar, but households only save a part of the increase in dividend
saving. Consequently, aggregate saving decreases and AD increases.
The above argument suggests that recent US tax changes reducing
double taxation of dividends may be expansionary, to the extent they
induce higher dividend payouts.13 However, there is an important
caveat to this. The justification for including P/Y in the investment
function is that it proxies for some form of cash-flow variable. In this
case, the aggregate investment function is better stated as:

I/K = a0 + axu + a2P/K + a3R/Y av a2f a3>0 (10.7')

Investment therefore depends on retained profits as a share of GDP,


rather than total profits. Now, if firms increase dividend payouts they
will reduce investment spending. If a 3 is large, the net effect could be to
shift the IS down and lower the profit and capacity utilization rates.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
238 Class Conflict and the Cambridge Theory of Income Distribution

The second caveat concerns balance-sheet effects that are not


modelled here. Changing dividend tax rates may just induce a shift
between debt and equity financing, leaving net payments unchanged.
In this case there would be no change in net corporate retentions, and
only the government budget would be impacted. This would result in

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
larger budget deficits, which are expansionary. However, these issues
push beyond the scope of the current chapter which has not addressed
the government sector and its relation to the household and corporate
sectors.

Ownership

A last issue concerns that of ownership, which is relevant for income dis-
tribution because it affects the distribution of dividend income. This is
an issue that is important for Cambridge theory but has not been
addressed. The above analysis was conducted on the basis of constant
ownership shares (unchanged zK and zw), the traditional assumption of
Cambridge theory. However, ownership is endogenous, and may change
as part of the adjustment process.
The reason why ownership matters is simple. Cambridge theory
emphasizes how income distribution adjusts to bring AD into alignment
with output. There are two ways to do this. One is to change the profit
share, which redistributes income between wages and profit. The other
is to change the pattern of ownership, thereby changing the distribution
of profit income between workers and capitalists.14 Cambridge theory
has always operated under the assumption that income distribution
alone does the adjustment via a changed mark-up - that is by adjust-
ment of the profit share. However, when there is investment-saving
imbalance ownership shares will also be changing. If capitalists are sav-
ing too much and there is excess saving, then their ownership share will
be rising. The reverse holds when workers are saving too much.
The process of changing ownership shares operates through back-
ground financial variables. Thus, if capitalists have excessive saving,
these savings can be thought of as being directed to equity purchases.
This drives up the price of equities and reallocates equity ownership to
capitalists. Consideration of these financial effects is beyond the scope
of the current chapter. Instead, the intention is to point out that saving
patterns impact ownership shares, and ownership shares impact the dis-
tribution of income and aggregate demand.
The addition of ownership concerns introduces an additional steady
state equilibrium condition. Now, in steady-state, capitalists must be

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Thomas I. Palley 239

saving just enough to finance their share of investment, thereby main-


taining their ownership share. This imposes the following steady-state
ownership condition:

SK {W/[l + y] + zK[P ~ R]} = zK[I ~ R] (10.17)

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
If capitalists receive no wage income the condition reduces to:

sK{zK[P - R]} = zK[l - R] (10.17')

This quickly generates amended Pasinetti-style conditions for income


distribution in an economy with corporate saving given by:

P/Y = I/sKY + R[l - l/sK]/Y (10.18a)


P/K = I/sKK + R[l - l/sK]/K (10.18b)

Corporate retentions, R, therefore reduce the profit share and profit rate.
The logic is that corporations are saving on behalf of capitalists, thereby
reducing the need for profit income to finance investment. This simple
derivation also illustrates how the Pasinetti conditions are in fact a form
of steady-state ownership condition.
Appropriate substitution into equation (10.17) combined with simple
algebraic manipulation yields:

a0 + axu + a2P/K + a3m(u, c)/


ZK={sKk(u)/[l + y]}/
[1 + m(u,c)] - [1 +sK][l -p\P/K
+ - - +n +/?
z(s.y,P,P/K,
V 7 , p , i / i x ,
u)
u, ( 1 0 1 9 )

Signs above functional arguments represent signs of partial derivatives.


From a partial equilibrium standpoint, increases in capitalists' propen-
sity to save increase capitalists' ownership share. Increases in workers
share of the wage bill decrease capitalists' share, and increased firm
profit retention ratios also decrease manager-capitalists' share.15
However, on top of this there are general equilibrium effects, because
changes in ownership shares impact AD, the profit rate and capacity uti-
lization that in turn feedback to influence ownership patterns. If an
increase in capitalists' propensity to save drives down the profit rate and
the utilization rate, this may induce negative manager-capitalist income
effects that outweigh the effect of an increased propensity to save, so
that the capitalist ownership share may fall. In other words, capitalists

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
240 Class Conflict and the Cambridge Theory of Income Distribution

can conceivably save themselves out of ownership. This is the asset


stock equivalent of the Kaleckian dictum that 'workers spend what they
earn, while capitalists earn what they spend'.

Conclusion: further issues and future research

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
This chapter has expanded the CPK-Kaleckian model of distribution to
include a labour market conflict channel that is distinct from the prod-
uct market competition channel. This labour channel works through
conflict over distribution of the wage bill, whereas product market com-
petition impacts the profit share. Kaleckians have long emphasized the
significance of both product market competition and labour market
conflict for income distribution; however, these two forces have been
conflated under the degree of monopoly, and the Kaleckian paradigm
has not been able to disentangle them.
The addition of the new channel enriches the structure of the model,
allowing it to simultaneously exhibit both wage-led and profit-led ten-
dencies. The model speaks to real-world concerns in that there have been
significant changes in the distribution of the wage bill, as well as changes
in the functional distribution of income. Both types of change matter for
macroeconomic outcomes, and the model captures both types.
The distinction between wage-share and wage-bill distribution has
important theoretical and policy implications. At the theoretical level, it
explains why economies can exhibit both wage-led and profit-led char-
acteristics. Redistribution of the wage bill to workers always raises AD
and economic activity by raising consumption. However, if the econ-
omy is profit-led, lowering the profit share can retard activity by lower-
ing investment spending. At the policy level, this suggests that
progressive policy should focus on altering the distribution of the wage
bill, rather than the profit share as has been the traditional focus.
Redistribution from managers to workers is always expansionary.
Redistribution from profits to wages is expansionary if the economy is
wage-led, and contractionary if it is profit-led. In the latter case, this
generates a growth versus equity trade-off. Unions may do a bit of both
types of redistribution - that is from managers to workers, and from
profits to the wage bill. This is strongly expansionary if the economy is
wage-led, but the effect is ambiguous if the economy is profit-led.
This dual wage-led-profit-led characteristic also helps make sense of
developments in the US economy over the last three decades. The dete-
rioration of the wage distribution has reduced AD (though this effect has
also been masked by increased household borrowing), but this has been

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Thomas I. Palley 241

offset by the positive impact on investment from a rising profit rate and
profit share. This helps explain why some pessimistic macroeconomic
prognostications regarding the effects of worsening income distribution
have not been realized.16
Finally, the model also addresses sociological criticism of Pasinetti's

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
model regarding its lack of a managerial capitalist class that draws
income from both profits and wages. The fact that both classes now
have two different sources of income also allows for reconciliation
between the Kaldor-Kalecki approach to saving behaviour, and that of
Pasinetti. Kaldor and Kalecki assumed different propensities to save out
of wage and profit income, a pattern of behaviour that can be justified
on behavioural rule of thumb grounds. People tend to consume most of
their wages, while leaving their savings accounts to compound. Pasinetti
emphasized different propensities to save across classes, but classes
saved at a common rate regardless of source of income. Now, it is possi-
ble to have behavioural rule of thumb saving within classes, and these
rules can vary across classes. One possible configuration is 0 <sww <sKW
<sWP<sKP<l, where sww = worker propensity to save out of wage
income, sKW = capitalist propensity to save out of wage income,
sWP = worker propensity to save out of profit income, and s^ = capitalist
propensity to save out of profit income.
To conclude, the Cambridge approach to growth and income distri-
bution remains vitally relevant. Though the main stream of economists
may be in denial about the major features of capitalism, the Cambridge
model is not. Looking to the future, there is need for an empirical and
analytic simulation agenda that builds on the theoretical framework
provided by the Cambridge approach to growth and distribution. Such
work will amplify its real world policy relevance.

Appendix

Stability analysis for the IS-MM model


The stability analysis for the two-equation goods-market mark-up model is as fol-
lows. It is assumed that capacity utilization increases in response to excess
demand in the goods market, and falls in response to excess supply. The profit
rate adjusts via changes in the mark-up, and the mark-up falls through product
market competition when above its equilibrium level. Conversely, it rises via
product competition when below its equilibrium level.
These dynamics can be represented by the following adjustment equations:

u = yEDG(u,p}K) (10A.1)
P/K =tyEM(u,P/K) (10A.2)

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
242 Class Conflict and the Cambridge Theory of Income Distribution

These equations can be linearized around the local equilibrium as:

u = <pEDGu(u - u*) + <pEDGP/K(P/K - P*/K) (10A.3)


P/K = \\fEMu(u - u*) + ^EMP/K(P/K - P*/K) (10A.4)

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
The profit-led case corresponds to EDGP/K > 0. Graphical analysis of stability for
this case is provided in Figures AlO.l and A10.2. In Figure AlO.l the MM curve is
flatter than the IS curve, and the model is cyclically stable. There is some casual
evidence that this configuration applies in the USA, since investment spending
has some profit-led tendencies, and firms' mark-up appears fairly constant over
the business cycle.
The wage-led case corresponds to EDGP/K < 0. Graphical analysis of stability for
this case is provided in Figures A10.3 and A10.4. In Figure A10.3 the MM curve is

IS

r n
CD
co MM

*t

Capacity utilization

Figure AlO.l Profit-led dynamics with IS steeper than MM

MM

CD
r IS
CO

*y
-r

Capacity utilization

Figure A10.2 Profit-led dynamics with IS flatter than MM

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Thomas I. Palley 243

L /
IS

^/±t MM

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
7^ •
Capacity utilization

Figure A10.3 Wage-led dynamics with MM flatter than IS

IS
o

Capacity utilization

Figure A10.4 Wage-led dynamics with MM steeper than IS

flatter than the IS curve, and the model is saddle-path stable. In Figure A10.4 the
MM is steeper than the IS, and the model may be cyclically stable or explosive.

Notes
1 The introduction of managerial wage income muddies the functional defini-
tion of 'workers' to be sure. The notion of what it means to be a worker in this
context, however, is beyond the scope of the chapter.
2 This distinction is attributable to Bhaduri and Marglin (1990). They
term wage-led economies as 'stagnationist', and profit-led economies as
'exhilarationist'.
3 Gordon (1995) reports that the US economy appears to have profit-led
tendencies in that investment responds positively to the profit share, while
consumption is impacted by income distribution variables.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
244 Class Conflict and the Cambridge Theory of Income Distribution

4 In the short-run Kaleckian model the distribution of income is determined


by the exogenously given mark-up.
5 This condition is needed to prevent workers out-saving capitalists, and
thereby driving them out of existence.
6 The simple logic of Pasinetti's result is that in equilibrium workers' and
capitalists' ownership shares of the capital stock are constant. This means

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
that the profits must adjust so that, given capitalists' propensity to save, cap-
italist saving exactly equals the share of investment they must finance to
maintain their ownership share.
7 The IS equation is given by P/K = {a0 + axu + a3m(u, c)/[l + m(u, c)]}/
ISK ~ a2]- Differentiating totally with respect to P/K and u yields h(P/K) =
{«i + «3[[1 + ™\mu - mrnu]/[l + m]2}8u/[sK - a2] so that 8(P/K)/(u ={ax +
a3[[l + m]mu - mmJ/fl + m]2}/[sK - a2]. The numerator is positive. The
entire expression, the slope of the IS, is positive if sK > a2.
8 Domowitz et al. (1986) and Chirinko and Fazzarri (1994) find acyclical or
mildly pro-cyclical mark-ups. Bils (1987) reports counter-cyclical mark-ups.
When a real-wage labour market closure (Dutt 1992) is used instead of a
product market closure, the mark-up is implicitly assumed to be counter-
cyclical since the real wage rises with capacity utilization. In effect, the MM
schedule is negatively sloped rather than positively sloped.
9 Because of the inherent ambiguity of the slope of the IS schedule, these
results are only illustrative.
10 Technology is usually viewed as exogenous. Neo-classical Marxists, such as
Bowles and Gintis (1990) and Skillman (1991), emphasize that technology is
endogenously selected by capital, which controls the production process.
This choice influences the ratio of non-supervisory to supervisory workers, a
feature emphasized by Gordon (1996).
11 The necessary condition is that sw < sK, which is the normal assumption in
Pasinetti growth models, being needed to stop worker-saving from driving
capitalists into extinction.
12 Their adverse impact on AD has been offset by rising household borrowing.
However, such borrowing is an unsustainable process, and the stagnationist
impulse must eventually come out in full (Palley 2002b).
13 This argument is in addition to the fiscal stimulus argument, whereby lower
dividend taxes raise the government budget deficit.
14 This claim is easily understood by examining the expression for AD in the stan-
dard Kaleckian model, given by: yd =cwwN + c^jnwN + ckzkmwN + I + G,
where cw = workers' propensity to consume, cK = capitalists' propensity to con-
sume, w = wage level, and G = level of government spending. AD consists of
worker spending out of wages, worker spending out of worker income, capital-
ist spending out of profit income, plus investment and government spending.
In the Kaleckian macro model, ownership shares and the mark-up are constant,
and output adjusts to AD. In the Kaldor-Pasinetti model, output is fixed at
potential, and AD adjusts to ensure balance. This can be done either by adjust-
ing the mark-up (m) or by adjusting ownership shares (zk, zw).
15 Increased firm retentions decrease managers' ownership share because firms
now do saving on behalf of workers. If profits were paid out, workers would
tend to consume more of the payout than managers, thereby reducing their
ownership share.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Thomas I. Palley 245

16 The effects of worsening income distribution may also have been masked by
a series of non-repeatable adjustment mechanisms including consumer bor-
rowing, a rising stockmarket and disinflation that has reduced household
mortgage burdens. These different channels of alleviation are examined in
Palley (2002a).

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
References
Baranzini, M. (1982) 'Can the Life-Cycle Help in Explaining Income Distribution
and Capital Accumulation?', in M. Branzini (ed.), Advances in Economic Theory
(New York: St Martin's Press).
Bhaduri, A. and Marglin, S. (1990) 'Unemployment, the Real Wage: The
Economic Basis of Competing Political Ideologies', Cambridge Journal of
Economics, 14: 375-95.
Bils, M. (1997) 'The Cyclical Behavior of Price and Marginal Cost', American
Economic Review, 77: 838-57.
Bowles, S. and Gintis, H. (1990) 'Contested Exchange: New Microfoundations for
the Political Economy of Capitalism', Politics and Society, 18: 165-222.
Chirinko, R. and Fazzari, S.M. (1994) 'Economic Fluctuations, Market Power, and
Returns to Scale: Evidence from Firm-level Data', Journal ofApplied Economics, 9:
47-69.
Dalziel, PC. (1991) 'A Generalisation and Simplification of the Cambridge
Theorem with Budget Deficits', Cambridge Journal of Economics, 15:
287-300.
Domowitz, I., Hubbard, G. and Petersen, B. (1986) 'Business Cycles and the
Relationship between Concentration and Price-Cost Margins', Rand Journal of
Economics, 17: 1-17.
Dutt, A.K. (1984) 'Stagnation, Income Distribution, and Monopoly Power',
Cambridge Journal of Economics, 8: 25-40.
Dutt, A.K. (1990) Growth, Distribution and Uneven Development (Cambridge:
Cambridge University Press).
Dutt, A.K. (1992) 'Conflict Inflation, Distribution, Cyclical Accumulation and
Crises', European Journal of Political Economy, 8: 579-97.
Fazzari, S., Hubbard, R.G. and Petersen, B. (1988) 'Financing Constraints and
Corporate Investment', Brookings Papers on Economic Activity, 1: 141-95.
Gordon, D. (1995) 'Putting the Horse (Back) before the Cart: Disentangling
the Macro Relationship between Investment and Saving', in H. Gintis and
G. Epstein (eds), Macroeconomic Policy after the Conservative Era: Studies in
Investment, Saving and Finance (New York: Cambridge University Press): 57-108.
Gordon, D. (1996) Fat and Mean: The Corporate Squeeze of Working Americans and
the Myth of Managerial Downsizing (New York: Free Press).
Kaldor, N. (1956) 'Alternative Theories of Distribution', Review of Economic
Studies, 23: 83-100.
Kalecki, M. (1942) 'A Theory of Profits', Economic Journal, 52: 258-67.
Keynes, J.M. (1936) The General Theory ofEmployment, Interest and Money (London:
Macmillan, now Palgrave Macmillan).
Lavoie, M. (1995) The Kaleckian Model of Growth and Distribution and its
Neo-Ricardian and Neo-Marxian Critiques', Cambridge Journal of Economics, 19:
789-818.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
246 Class Conflict and the Cambridge Theory of Income Distribution

Palley, T.I. (1996a) 'Inside Debt, Aggregate Demand, and the Cambridge Theory
of Distribution', Cambridge Journal of Economics, 20 (July): 465-74.
Palley, T.I. (1997) 'Money, Fiscal Policy, and the Cambridge Theorem', Cambridge
Journal of Economics, 26 (September): 633-9.
Palley, T.I. (2002a) 'Financial Institutions and the Cambridge Theory of
Distribution', Cambridge Journal of Economics, 26: 275-7.

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Palley, T.I. (2002b) 'Economic Contradictions Coming Home to Roost? Does the
US Economy Face a Long-Term Aggregate Demand Generation Problem?'
Journal of Post Keynesian Economics, 25 (Fall): 9-32.
Pasinetti, L. (1962) 'Rate of Profit and Income Distribution in Relation to the Rate
of Economic Growth', Review of Economic Studies, 29: 267-79.
Ricardo, D. (1965) Principles of Political Economy and Taxation, [1821, 3rd edn]
(London: Dent Dutton).
Rowthorn, R.E. (1982) 'Demand, Real Wages, and Economic Growth', Studi
Economici, 18: 3-53.
Skillman, G. Jr. (1991) 'Efficiency vs Control: A Strategic Bargaining Analysis of
Capitalist Production', Review of Radical Political Economics, 23: 12-21.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
11
The Dynamics of Profit- and
Wage-led Expansion: A Note

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Amit Bhaduri

Any redistribution of income between profits and wages would have


contradictory effects in terms of aggregate demand, so long as the
propensities to consume are different for the two classes. For instance,
the lowering of the real wage rate would tend to depress total consump-
tion expenditure by redistributing income against the wage-earners with
a higher propensity to consume. At the same time, it might also encour-
age investment by increasing the margin of profit per unit of sale.
Depending on which effect dominates quantitatively, two alternative
regimes or paths both led by demand emerge: the consumption- or wage-
led path which has also been called 'stagnationist', and the investment-
or profit-led path called 'exhilarationist' (Bhaduri and Marglin 1990;
Marglin and Bhaduri 1990).
The essential formalism is simple. With full capacity (potential) out-
put normalized at unity, the saving of the economy is written as:

S = shz, l > z , h>0 (11.1)

where 5 is the saving propensity out of profit, and all wage is assumed to
be consumed for expositional simplicity; h = P/Y is the share of profit in
output; and z = Y/Y* is the degree of capacity utilization, with Y* = 1,
that is the normalized level of full capacity output.
We assume investment depends positively on both capacity utilization
(z) and the profit margin (m) or the profit share (h), which are by defini-
tion positively related as: h = m/(l + m). Assuming for expositional
simplicity, static expectation, the investment function is written as:

I = I(z,h), I2>0, 4>0 (11.2)

247

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
248 The Dynamics of Profit- and Wage-led Expansion

From (11.1) and (11.2), the slope of the locus of saving-investment


equality, the IS-locus, is derived through total differentiation as:

dh sh - Iz

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
The positive slope of (11.3) indicates that a higher profit share (h) results
in higher capacity utilization (z), placing the economy in the exhilara-
tionist regime. In this case the stimulating effect of the higher profit share
on investment expenditure outweighs its depressing effect on con-
sumption expenditure. When the slope of (11.3) is negative, the economy
is said to be in the stagnationist regime for the opposite reason. However,
the two regimes have normally been distinguished by assuming as valid the
stability condition of the one-variable income adjustment process through
the Keynesian multiplier mechanism. This requires saving to be more
responsive than investment to changes in income, making the denominator
of the right-hand side of expression (11.3) positive:

sh-Iz>0 (11.4)

Consequently,

lh-sh>0 (11.5)

implying that investment is more responsive than saving to the profit


share, sets the economy on a profit-led exhilarationist path. If, on the
other hand,

lh-sh<0 (11.6)

the economy is on a wage-led stagnationist path for the opposite reason.


However, this analysis remains valid only so long as the distribution
of income (h) is treated as an exogenous variable. In effect, this makes
the underlying dynamical system correspond to the usual single variable
income adjustment process of Keynesian theory with income distribu-
tion given, but capacity utilization (z) adjusting to excess demand or
supply in the product market. With capacity utilization (z) as the only
endogenous variable in the system, the dynamic adjustment equation,
in view of (11.1) and (11.2) becomes:

^ = a[l(h,z)-shz] (11.7)

where a > 0 is some arbitrary positive speed of adjustment.

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Amit Bhaduri 249

However, the above argument could be reversed by making income


distribution (h) the endogenous variable adjusting to excess demand or
supply in the product market, while the degree of capacity utilization is
assumed exogenously fixed, say at the full employment or full capacity
level. In this case an excess of demand in the product market would raise

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
the price level; and, if the money wage rate failed to keep pace in per-
centage terms, the real wage rate would fall, the margin and share of
profit would rise. This would tend to close the gap between investment
and saving at the exogenously given full employment or full capacity
level of output. This theory of distribution, sketched originally by
Keynes (1930, see also 1936: vii), was developed later as the Keynesian
theory of distribution (Kaldor 1956; Pasinetti 1962; Robinson 1962;
Marglin 1984). The theory presumes not only full employment, but also
'forced saving' by the workers, even in a situation of full employment.
On the other hand, if this latter assumption is abandoned, the possibil-
ity of a 'profit squeeze', rather than 'forced saving' by the workers, has to
be included in the analysis. In this case the money wage rate would rise
faster than the price level, and the share of profit would fall rather than
rise in response to excess demand in the product market. Thus, a more
general version of the dynamic adjustment equation for the 'Keynesian'
theory of distribution might be written as:

^ = p[l(h,z)-shz] (11.8)

where the speed of adjustment is /3 > 0 in the case of 'forced saving' by


the workers, but /3<0 in the case of a 'profit squeeze'.
Note the central difference between adjustment equations (11.7) and
(11.8). In the former, the profit margin as well as the profit share are
given which makes the real wage inflexible, so that the entire burden of
adjustment falls on capacity utilization (Kalecki 1971). In the latter case,
the margin is flexible but capacity utilization is fixed by assumption,
making adjustment work exclusively through the real wage rate, that is
income distribution.
Consequently in a more general case in which neither the profit mar-
gin and share (h) nor the degree of capacity utilization (z) is treated as
exogenous, and both are endogenous variables reacting simultaneously
to an excess demand or supply in the product market, we have a coupled
dynamical system consisting of (11.7) and (11.8). The product market
clearing, equilibrium path of the system depicted by the IS-curve with
its slope given in (11.3) still remains the same. However, the stability

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
250 The Dynamics of Profit- and Wage-led Expansion

property of this two-variable dynamical system need no longer


correspond to the one-variable stability condition (11.4) of the usual
Keynesian system.
Moreover, since both z and h are assumed to be endogenous, for a and
P positive, a positive relation between z and h leading the trajectory of

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
the economy to the profit-led exhilarationist regime. This means that
'forced saving' by the workers implies that the economy is necessarily
on an exhilarationist path. Obversely, with a > 0 , but /3<0 in the 'profit
squeeze' case, the economy is necessarily placed on a wage-led stagna-
tionist path. This can be seen easily from dividing (11.7) by (11.8) to
obtain the integral curve with the slope:

I=I <»*>
where a>0, but /3 can be positive or negative.
The stability of the dynamical system (11.7) and (11.8) may be exam-
ined by considering the function:

V(t) = | [I(h, z) - shz]2 (11.10)

The function V is a Liapunov function with all the desirable properties.


It enjoys stability in the large, that is positive definiteness and unbound-
ness as (I-S) tends to infinity, provided also dV7dr<0 (see for example
LaSalle and Lefschetz 1961; Minorsky 1962; Gandolfo 1995). This last
condition can be checked by differentiating (11.1) and (11.2) with
respect to time, and substituting from (11.7) and (11.8) to obtain:

^ = [/ - S]2[p(Ih - sz) - a(sh - Iz)]< 0 (11.11)

Therefore, global stability requires:

P(Ih - sz) < a(sh - Iz) (11.12)

Using (11.3) and (11.12), a complete classification of the various cases


of profit-led or exhilarationist as well as wage-led or stagnationist
sub-regimes according to their stability property becomes possible
(Table 11.1). Note, however, that this classification is done in two successive

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Amit Bhaduri 251

Table 11.1 Classification of regimes (a > 0, p > 0 or p < 0)

Stability property from (11.11)


Sign of
dz For/3>0 For (3<0
(Ah Nature of ('forced savings'of ('profit squeeze')

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Cases from (11.3) regime workers)

Al Positive Profit-led Stablity ambiguous, Unambiguously stable,


sh-Iz>0 and more likely the but not possible from
and larger is — (H-9), since z and h are
Ih-sz > 0 endogenous

A2 Negative Wage-led Unambiguously stable, Stability ambiguous,


5/2-4 > 0 but not possible from more likely, the larger
and (11-9), since z and h —
lh-sz<0 are endogenous '™

Bl Negative Wage-led Unambiguously Stability ambiguous


sh-Iz < 0 unstable, but not less likely, the larger
and possible from (11.9), a
Ih-sz < 0 since z and h are
endogenous

B2 Positive Profit-led Stability ambiguous, unambiguously


sh-L < 0 and less likely the unstable, but not
a
and larger is — possible from (11.9),
Ih-sz < 0 since z and h are
endogenous

* See Appendix for explanation of Al, A2, Bl and B2.

steps to highlight the difference between a single endogenous variable


stability analysis given by conditions (11.3) to (11.6), and the stability
analysis in the case of the dynamical system (11.7) to (11.9) when both
the variables, h and z are endogenous.
In the case of the general dynamical system with both variables
endogenous, it emerges as a general result that neither the profit-led nor
the wage-led path of expansion is unambiguously stable. The stability
depends critically on the relative magnitudes of the speeds of adjust-
ment, that is the absolute value of the ratio (a/p). Thus, when p >0, that
is the case of 'forced saving' by the workers, only the profit-led path is
relevant, and the usual one-variable Keynesian stability condition (11.4)
may or may not matter in determining the stability of this profit-led
regime depending on the ratio of the relative speeds of adjustment. The

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
252 The Dynamics of Profit- and Wage-led Expansion

higher (lower) is the ratio, implying faster (slower) speed of adjustment


of capacity utilization relative to that of income distribution, the more
(less) likely is the stability of the profit-led regime, depending on
whether the sensitivity of saving is more (less) than that of investment
to changes in income (that is, condition 11.4 satisfied or not). In a sim-

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
ilar manner, in the case of 'profit squeeze', that is p < 0, leading to wage-
led expansion, again the Keynesian stability condition may or may not
be relevant depending on the (absolute) value of the relative speeds of
adjustment. It follows, the Keynesian stability condition is neither nec-
essary nor sufficient without considering simultaneously the relative
speeds of adjustment of capacity utilization and of income distribution.
The critical role played by the relative speeds of adjustment points
towards an interesting possibility of extending this analysis. It is often
argued that Keynesian analysis neglects the 'supply side'. One way of
taking into account the supply side would be to incorporate into the
analysis the fact that the speed of adjustment of capacity utilization
tends to decrease as the degree of utilization increases, and various bot-
tlenecks begin to appear on the supply side. This means a can be treated
as a decreasing function of z to partly incorporate considerations on the
supply side. This would lead to non-linearities which we have avoided
dealing with in this chapter, and which must remain a matter of future
research.

Appendix

The 'profit-squeeze' case


z = a[l(h, z) - shz]; h =- /x [/(ft, z) - shz]
2
V(t)=±(I-S)

^ = (I - S)\lhh + Izz - szh - shz]


at
= (I-S)[(Ih-sz)h+{Iz-sh)z]
= (I - S){(Ih - sz)[ - /*(/ - S)] + (Jz - sh)a(I - S)}
= (I - S)2[ - /x(4 - SZ) + a(Iz - Sh)]<0
For ^ < 0 => - fi(Ih - sz) + a(Iz - sh) < 0
dr
- fi(Ih - SZ)<- a(Iz - Sh)

or fi{sz - Ih) < a(sh - Iz)

(11.12a) instead of (11.12): a(sh - Iz) > fi(sz - Ih); \p\ = /i > 0

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Amit Bhaduri 253

Cases
Al: LHS > 0; RHS < 0 .-.unambiguously stable.
A2: LHS. > 0; RHS > 0 /.stability ambiguous; larger f ^ j more likely stability.

Bl: LHS < 0; RHS < 0 .-.stability ambiguous, larger 1^1 stability less likely.

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
B2: LHS < 0; RHS > 0 .-.unambiguously unstable.

LHS=left-hand side; RHS=right-hand side.

References
Bhaduri, A. and Marglin, S. (1990) 'Unemployment and the Real Wage: The
Economic Basis for Contesting Political Ideologies', Cambridge Journal of
Economics, 14: 375-93.
Gandolfo, G. (1995) Economic Dynamics, 3rd edn (Berlin: Springer).
Kaldor, N. (1956) 'Alternative Theories of Distribution', Review of Economic
Studies, 23: 83-100.
Kalecki, M. (1971) Selected Essays in the Dynamics of the Capitalist Economy
(Cambridge: Cambridge University Press).
Keynes, J.M. (1930) A Treatise on Money, 2 vol. (London: Macmillan, now Palgrave
Macmillan).
Keynes, J.M. (1936) The General Theory ofEmployment, Interest and Money (London:
Macmillan, now Palgrave Macmillan).
LaSalle, J.P. and Lefschetz, S. (1961) Stability of Liapunov's Direct Method with
Applications (New York: Academic Press).
Marglin, S.A. (1984) Growth, Distribution and Prices (Cambridge, Mass.: Harvard
University Press).
Marglin, S. and Bhaduri, A. (1990) 'Profit Squeeze and Keynesian Theory', in
S. Marglin andJ.B. Schor (eds), The Golden Age of Capitalism (Oxford: Clarendon
Press).
Minorsky, N. (1962) Nonlinear Oscillations (New York: Van Nostrand).
Pasinetti, L. L. (1962) 'The Rate of Profit and Income Distribution in Relation to
the Rate of Economic Growth', Review of Economic Studies, 29: 267-79.
Robinson, J. (1962) Essays in the Theory of Economic Growth (London: Macmillan,
now Palgrave Macmillan).

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
This page intentionally left blank

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Index

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
Notes: f = figure; n = note; t = ta bold = extended discussion or
heading/word emphasized in the main

African Americans 120-1 balanced budget 11


ageing bankruptcy 5, 68, 125, 141-5
pensions and income distribution banks 68, 71, 73, 231
6, 181-205 bargaining 94, 99, 107-9, H l ( n 7 ) ,
ageing and cost of 'pay-as-you-go' 224, 233
social security organisation collective bargaining 2 2 , 5 1 ,
196-201, 2 0 4 ( n l 0 - l l ) 77-84, 88(nl8-23)
ageing and labour market 187, see also wage-bargaining
189-95, 203-4(n3-9) 'beggar-my-neighbour' policies 9
demographic changes (impact on Belgium 59, 60t-61t, 176t, 190t,
labour market) 193-5, 192t, 197t, 2 0 4 ( n l l )
203-4(n6-9) unemployment: causes 23,
labour-supply and dependency-ratio 24t-26t, 37t, 37, 41t, 46(n2)
scenarios 187, 189-93, benefit duration 25t, 27, 28t, 30t,
203(n3-5) 32, 33, 35, 36, 43, 45, 52
aggregate demand (AD) 6, 7, 50, 53, Bhaduri-Marglin model (1990) 234,
57, 58, 86(n5), 117, 127, 132, 202, 243(n2), 247
207, 224, 226, 231-2, 234-40, effect of profits on investment and
244(nl2, n l 4 ) , 247 employment 6, 2 0 6 - 2 2
aggregate governance indicator 157 borrowing 16, 124, 124t, 125n,
aggregate investment function 237 133, 137
agriculture 95, 102, 153, 157, 215 Britain see United Kingdom
'alliances for jobs' 79, 84, 88(n23) budget deficits 11, 12, 15, 2 1 ,
assets 72, 124, 128, 129, 133, 113(nl8), 196, 238, 244(nl3)
134t, 141 budget surplus 114(n26)
Atkinson's inequality measure 135, business cycle 2 1 , 109, 124, 124t,
136t, 137, 138t, 139t, 140t, 144t, 125t, 127, 134, 199, 230, 242
146(n9) business sector 95, 97-9, lOOt, 108,
Australia 175t, 190t, 192t, 197t 213, 221(n2)
unemployment: causes (sclerosis
versus macroeconomic policy) Cambridge Post-Keynesian (CPK)
23-7, 37t, 41t, 45 theory of income distribution
Austria 23, 24t-26t, 37t, 4It, 46, 59, 6-7, 223-46
60t-61t, 103, 175t, 190t, 192t, Kaleckian extension 224, 2 2 8 - 3 1 ,
197t 244(n7-9)
key insight/core idea 224
baby-boom generation 182, 183, 186 revisited 226-8, 244(n4-6)
balance of payments 12, 72, 106 see also class conflict

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
256 Index

Cambridge savings function 208, 209t capital-labour ratio 54


Canada 51, 56, 176t, 186, 190t, capitalism 3, 6, 18, 69, 194, 195, 241
192t, 197t conflictive/cooperative 10
unemployment: causes 23-9, 35, capitalists 195, 202, 208, 224, 227,
37t, 41t, 46, 47(n5, n7, 9) 231, 238-9, 244(n5, n i l )
capacity utilization 6, 7, 56, 58, 59, cash flow 229, 237

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
63, 207-13, 215-17, 230, 230f, catering lll(n3)
234-7, 239, 241, 242f, 243f, central bank experts 221
244(n8), 247, 248, 249, 252 central bankers 38-40
capacity utilization rate 229, 231 central banks 12, 23, 70-2, 87(n8)
capital 5, 63, 87(nl3), 112(nl2), charge-off rates 125, 141-5
128, 138, 173, 202, 203(nl), children 122, 123, 194
244(nl0) City of London 9-10
balance of power (vis-a-vis labour) civil society 152
177(nl6) class 123, 195, 199, 203(nl), 219,
constant 194 224, 227, 231-4, 241, 244(nl0)
demand for 127 middle class 113(n20), 129, 130
entrepreneurial 87(nl2) ruling classes 182
external 87(nl2) class conflict 6-7, 40, 224, 231, 234
financial 199 class conflict and Cambridge Theory
fixed 134n of Income Distribution 6-7,
intangible 55 223-46
and labour (degree of caveats 237-8
substitutability) 53-4 CPK model revisited 226-8,
capital accumulation 54, 57, 193-4, 244(n4-6)
196, 203, 203(n6), 206-13, Kaldor model 227f, 227
215-17,220 Kaleckian dictum 240
effect of demand 219 Kaleckian extension of CPK model
export-led 21 It, 216t 228-31, 244(n7-9)
per labour hour 55 national income tree 225f
rate 208 ownership 2 3 8 ^ 0 , 244(nl4-15)
capital expenditures 125, 125t significance of income distribution
capital flows 9, 13 223-4
capital income 124, 135, 139, 141 stability analysis 234-8,
capital liberalization 106,173 244(nll-13)
capital markets 8 7 (n 12) theoretical innovations 225-6
deregulation 11, 12 class structure 227
imperfect 2, 87(nl2) Cobb-Douglas production function
liberalization 12 53,54
capital mobility 129 Cold War 10
capital productivity 221 (n2) collateral 135, 136, 137
capital shortage 53, 54, 56-7 collective agreements 97, 111 (n3)
capital stock 2, 4, 208, 209t, 233, commodities 70-3, 193
244(n6) companies/firms 18, 20, 58, 59, 73,
business sector 59, 60t, 63 83, 84, 101, 103, 105, 108,
physical and intangible, 55 lll(n2), 112(nl0, 15), 125, 130,
capital stock: relation to 173, 219, 221, 244(nl5)
unemployment 4, 49-66 cost structure 230
capital-labour power relations domestic 74
87(nl3) manufacturing 106

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Index 257

companies/firms - continued corporate governance 128


multinational corporations (MNCs) corporations 12, 125, 239
172-4,221 internal resources 125t, 125n
price-setting 75 productive uses 125t
profit-retention 233 US 127
state-owned 173 use of non-financial resources 125t

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
comparative advantage 151-2,153, corruption control 157
167, 171, 172 cost-of-living index 103, 111 (n3)
'compensation' see wages counter-cyclicality 21, 22
competition 3, 10, 17, 18, 202, credit 68, 72, 73, 104
231-2 endogenous 137
financial 133 non-bank 5, 133, 145
imperfect 54 credit cards 5, 133, 141, 144t
inter-firm 231-2 credit supply 132, 133, 135, 136, 142
international 102, 105, 216 credit-card debt (CCR) 124n, 133,
labour market 232 137-41, 143-5, 145(n6)
perfect 15 see also debt
product-market 6-7, 224, 225 creditors 72, 73
'competitive corporatism' 4, 79 culture 182
competitiveness 4, 106, 113(n25), currencies 106
217, 220 attacks on 174
construction 102, lll(n3) devaluations 5, 9, 101, 106, 108
consumer credit 124t, 124n, 141-5, domestic 72
145(n6) national 113(n25)
consumer debt/consumer borrowing currency depreciation 106, 107,
5, 117, 125, 132-45, 145-6(n4-9), 113-14(n25)
245(nl6) currency market concerns 21
see also credit-card debt currency speculation 21
consumer price index (CPI) 67, 82n, Cyprus 176t
86(nl) Czech Republic 190t, 192t, 197t
consumer prices 45, 68, 156
consumer spending 127, 134t, 134n, debt 5, 68, 72, 133, 134, 141,
145(n4) 146(n8), 196, 198
consumers 71, 72, 85, 86(n5), 141 see also household borrowing/debt
consumption 7, 12, 72, 73, lll(n3), debt composition 142, 143
118, 124, 132-4, 218, 236, 240, debt deflation 69, 72, 73
243(n3) debt and equity financing 238
domestic 210 debt service 142, 143, 144t
mass 10 debt-to-income ratio 16
negative domestic effect 210 debt-service costs 133
private 74f, 75f debt-servicing 16, 141-3, 231
consumption expenditure/spending debtors 68, 72, 73
134n, 134, 247, 248 decentralization 17
consumption growth 126t, 127, 132, 'decentralization of production' 101,
134 lll(n2)
consumption-led regime 12 default 141, 143
'COORD' (coordination index of wage deficit-financing 11
bargaining amongst union and deflation 2, 6, 71, 72, lll(n3)
employers) 27, 28t, 30t, 32-6, actual 86
43-5, 47(n6-7) demand-led, sustained 68

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
258 Index

deflation risks: role of wages and deregulation 129, 151


wage-bargaining 4, 67-92 developed countries 22, 109, 182-7,
deflationary policies/choices 106, 201-2, 206-7, 212-14, 217-18, 221
113(n22), 202 see also OECD countries
deflationary processes/tendencies 4, developing countries 22, 152, 156,
68, 69, 71 167, 177(nlO), 183, 185, 206, 207,

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
demand 16, 59, 132, 213, 219 212, 218, 220, 221
composition 17 effects of economic liberalization on
domestic 12, 85, 209, 217, 218, 220 income distribution 5
and employment 215, 220 development 101, 151
effect on unemployment and development level 16, 153, 167, 170,
wages 62 177(nl3)
effective 189 disinflation 3, 4, 22, 27, 32, 43,
excess 241, 248-9 70-2, 80, 84, 245(nl6)
foreign 72 dividend taxes 233, 237, 238,
insufficiency 14, 17 244(nl3)
international 210, 215, 218 dividends 118, 125, 125t, 128, 225f,
over-contraction 12 225, 238
short-term, long-term 14 division of labour 18
see also aggregate demand downsizing 16, 128
demand deficiencies 206 downturns 73, 80, 95
demand growth 118, 209
demand shocks 4, 50, 55 Earned Income Tax Credit 130
temporary 68 earnings 102, lll(n3), 129
demand-led expansion actual (per employee hour) 8It, 82t
dynamics of different regimes 7, actual rates of change 97
247-53 aggregate lll(n3)
'profit-squeeze' case 252 gross 88(nl8), 95, 96, lll(n4)
demand-management 3, 9, 10, 11, real net lll(n4)
12-13, 14, 221 see also employee remuneration
democracies 9, 10, 15 economic activity 4, 7, 49, 63, 68,
demographic changes lll(n6), 235, 236, 240
impact on labour market 193-5, cyclical fluctuations 10
203-4(n6-9) real 50
demographic transition 183, 185 economic growth 1-8, 22, 73, 78,
demographic trends 182-7, 188-9t, 80,95, 118, 130, 131, 151, 165,
194, 203(n2) 173, 181, 199, 202, 208, 212, 215,
global scenarios 182-5 218-19, 226, 237, 241
replacement migration 186-7, country rates 46
188-9t cross-country spillover effects 29
Denmark 176t, 190t, 192t, 197t, 'endogenous' 18
204(nll) export-oriented 5, 153, 167, 207,
unemployment: causes 23, 215, 216, 217, 220
24t-26t, 35, 37t, 4It, 37, 46 full-employment steady-state 227
dependency-ratio scenarios 187, growth rate 16, 18
189-93, 203(n3-5) labour-market flexibility 2-3,
see also old-age dependency ratio 9-19
dependent population 202 real 200
changing composition 201 slow/sluggish 43, 84, 172
rising cost 202 sustainable 152

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Index 259

economic liberalization employee contributions 96,


effects on income distribution 5, lll(n4-5)
151-80 employee protections 20, 21
economic policy 68, 102 employee remuneration 80, 8It, 82t
domestically oriented 9 see also wages
flexible and counter-cyclical Employee Retirement Income Security

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
fluctuation 21 Act (ERISA) (USA, 1974) 128
theories and counter-theories employees 10, 55, 69, 70, 83, 84,
13-16 87(nl5), 88(nl8-19), 95, 96, 97,
economic recovery 85 98t, 101, 108, l l l ( n 5 , n7)
economics 2, 203 ( n l ) public sector 113(n20, n23)
classical 4, 6, 93, 94, 181, 228 real hourly average earnings 59
classical-Kaleckian 195 rights at work 233
classical-Keynesian 181, 194, 202 see also labour
mainstream 207 employers 10, 70, l l l ( n 2 - 3 , n 5 , n7),
monetarist 70 129
new classical 1, 7(nl), 70, 177(nl4) paying low wages 130
New Keynesian 1, 7(nl), 177(nl4) social security contributions 79n,
non-conventional 202 88(nl8)
see also Keynesianism; neoclassical employment 1^8, 22, 43, 50, 53, 57,
economics 63, 64, 68, 70-3, 84, 86,
economists 14, 56, 68, 121, 199, 86-7(n5-12), lOOt, 101, 108,
203(n6), 241 112(nl5), 114(n26), 118, 151-4,
classical 193, 194, 203(n6) 156, 157, 167, 172-4, 193, 194,
conventional 199 196, 200-2
mainstream 203(n3) effect of profits (Bhaduri-Marglin
neo-classical Marxist 244(nl0) model) 6 , 2 0 6 - 2 2
economy/economies 8 7 (n 11) effects of wage moderation,
capitalist 200 113(nl8)
closed 9, 71, 77 equilibrium level 54
credit money 70-3 export-led 21 It, 216t, 217
domestic efficiency/efficient high 75
organization 151-4, 165, 167, income distribution effects 168-71
171, 172, 174 industrial 97, 103-7, l l l ( n 8 )
international 20 irregular 102
monetary 69 large firms 102
open and globalized 3 limited duration 102, 113 (n 17)
education 157, 171, 177(nl2), 199 manufacturing 127
educational attainment 120, 121, 'natural' level 189
129, 131, 167, 191 'peculiar paradox' 13
educational inequality 121 policies 16-18
effective demand 2, 14, 72, 86, 94, short-term barrier 72
181, 194-6, 213 temporary 102
elasticity of substitution total 105f, 107
capital and labour 53, 54, employment conditions 228
112(nl2) employment creation/growth 13,
elections 9, 15 101, 206
Emergency Home Finance Act (USA, employment legislation 1, 50, 51
1970) 133 employment protection 1, 22, 40,
emerging markets 152 50,51

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
260 Index

employment rates 124, 191, 204(n7) nominal 103


employment share 221 (n2) official 219
employment trends 80, 96 rates of change 107t
'England' see United Kingdom real 58, 103, 105, 106, 113(n25)
entrepreneurs/entrepreneurship 63,104 exchange rate agreements 108
'equilibrium' 15 exchange rate markets 174

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
euro 13, 21 exhilarationist regimes 210, 237,
Euro-sclerosis 22 243(n2), 248, 250
euro-zone area 20 export competitiveness 217
Europe 12, 21-2, 27-8, 34-7, 44f, export demand 212
44, 46, 50, 52, 55, 57, 79, 94, export growth 22, 172, 217
108, 188t, 189t, 195, 202, export markets 13
203(n4), 206 export performance, productivity-led
economic interdependence 33 220
Southern 186, 189 export prices 103, 106
Western and Northern 186 export sector 152
European Central Bank (ECB) 13, 2 1 , export surplus 11-12, 13, 16, 86
38, 67, 85, 86, 8 8 ( n l 9 , n26) export-led expansion 1 6 , 1 7 4
monetary policy ('anti-growth exports 35, 46, 104, 113(nl9), 151,
bias') 84 171,211,215
European Commission 113(n25) capital-intensive (or not) 127-8
European Economic C o m m u n i t y 52 income distribution effects (by
European Monetary System (EMS) sector) 167-8, 177(nl3)
104, 106, 110 'increase labour d e m a n d ' 21 I t
European Monetary Union (EMU) merchandise 156
country sample (Austria, Belgium, net 208, 209t, 210
Germany, Spain) 49, 50, 63 profit-led 21 It, 216t
deflation risks: role of wages and
wage-bargaining 4, 67-92 factor demand schedules 93
inflation rate 78f factors of production 223
labour income share 75-6 fairness/equity 152, 223
larger countries 80, 85 families 121, 130, 175t, 193, 202
monetary and fiscal policies 70 high-income 122
remuneration of employees 79f low-income 118, 123
restrictive policy mix 85, 88(n24) unpaid work 215
smaller countries 80 see also households
unit-labour-cost growth 78f family income 118, 121, 122, 124
unit-labour-cost growth and family subsidies 113 (n23)
inflation rate 75f Federal Home Loan Mortgage
wage policy dilemma 84-6, Corporation ('Freddie Mac', USA)
88(n24-7) 132
wage trends and extent to which Federal National Mortgage
scope for distribution is Corporation (FNMA, 'Fannie
exploited 8 It Mae', USA) 132
European Union 9, 20, 29, 87(nl4), financial crises 152
187, 188t, 189t, 200, 221 financial distress 132-45,
exchange rate 145-6(n4-9)
depreciation 219 financial institutions 133
new 103 financial intermediaries 68, 73

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Index 261

financial liberalization 199 economic institutes 64(n2)


financial markets 12, 17, 38, 129 'exit' from poverty 129
financial services 128 inflation rate (consumer prices)
Finland 23, 24t-26t, 35, 37t, 37, 41t, (1991-2003) 78f
176t, 190t, 192t labour income share (1960-2003)
fiscal convergence criteria 21 75-6

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
fiscal discipline 5 , 1 5 3 , 1 5 6 , 1 7 1 'largest economy in EMU' 84
fiscal policy 10, 2 1 , 22, 70, 73, 104, post-reunification boom 80
110, 172, 213 productivity, labour costs, exchange
expansionary 12 rates (1979-2000) 107t
pro-cyclical 85 remuneration of employees
restrictive/tight 2, 13, 114(n26), (1991-2003) 79f
221 reunification 78
fiscal retrenchment 172 unemployment: causes 20, 23,
fiscal stimulus 244(n 13) 24t-26t, 35, 36, 37t, 41t
forced saving 249-52 unit-labour-cost growth
foreign direct investment (FDI) 5, (1991-2003) 78f
153-56, 164, 171, 174 unit-labour-cost growth and
foreign exchange reserve 12 inflation rate (consumer prices)
France 56, 57, 80, 104, 106, 176t, (1961-2003) 74f
188t, 190t, 192t, 197t, 200, wage policy dilemma 84-6,
2 0 4 ( n l l ) , 207, 213, 214t 88(n24-7)
unemployment: causes 23, 24t- wage trends and extent to which
26t, 35, 36, 37t, 41t the scope for distribution is
freedom of association 22 exploited (1996-2003) 82t
Friedman, M. 11 wages policy 69
full employment 3, 9, 10, 4 3 - 5 , western 83
47(n9), 53, 55, 103, 113(nl8), globalization 3, 11, 12, 15, 129, 152
181, 189, 199-201, 224, 226, goods market 6, 87(nl3), 206-7,
227, 249 208, 210-13, 217, 230, 234, 241
fund managers 128 'determines labour market' 21 It
interaction with labour market 212
GDP 35, 46(n3), 47(n4), 104, 126t, governance 152, 153, 162, 163t, 163,
156, 157, 163, 164, 215 165-8, 171, 174, 177(nl3)
GDP deflator 58, 59, 62t, 67, 86(nl) government insurance and pension
GDP at factor costs 76f, 87(nl5) reserves 134n
GDP growth 23, 24t, 196, 221(n2) Government National Mortgage
real 45, 126t Association (GNMA, 'Ginnie
gender inequality 120 Mae', USA) 132
geography 16, 165 government saving 227
Germany 9, 13, 56, 57, 59, 60t-61t, government size
62t, 103, 104, 106, 188t, 190t, income distribution effects 168-71
192t, 197t, 200, 2 0 4 ( n l l ) governments 38, 59, 130, 172,
collective bargaining system 83, 244(nl4)
88(n21) economic credibility 16-17
consumer price index 82n economic policies 12
deflation risks: role of wages and effectiveness 157
wage-bargaining 4, 67-92 Great Depression (1929-) 69
eastern 83 Greying countries 189-93

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
262 Index

health 129, 177(nl2), 198 income distribution 1-8, 10, 44,


high employment (debate over causes) 58, 81t, 82t, 94, 110, 117, 122,
20-3 206, 208-13, 215, 217-20, 248,
household borrowing/debt 118,134, 249, 252
137, 240, 244(nl2) Cambridge Theory 6-7, 223-46
regression results (1980-2003) cross-country 40, 47(n8)

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
136t effect on growth 21 It, 214t, 216t
total 140, 141 effects of economic liberalization 5
see also indebtedness effects of labour-market conflict
household economic distress 141-5, 224
145-6(n7-9) functional 75, 87(nl4), 227, 235,
household income 5, 42, 117, 123, 236, 237, 240
132-3, 137, 141-5, 155-6, 175t pensions 6, 181-205
households 73, 175t, 237, 238 profit-wage functional 232
bankruptcy 125 significance 2 2 3 - 4
economic distress 5 , 1 1 7 - 5 0 steady state 228
top versus bottom/middle (income unequal 2
strata) 122 income growth 134, 143
total number (HH) 142, 145(n7) income inequality 44, 47(n8), 123,
see also families 126t, 137, 154-6, 166
housing 12, 145(n7) labour-market institutions and
Housing Price Index (HPI) 135, 136t, 4 0 - 3 , 47(n8)
138t, 139t, 140t, 144t income tax 96, l l l ( n 5 )
Housing and Urban Development Act indebtedness 73, 87(nl2)
(USA, 1968) 132 see also mortgage debt
h u m a n capital 2, 120, 152-4, 157, individuals 14, 15, 16
165, 171, 172 industrial pay inequality 155
h u m a n development 185 industrial policy 113(nl9), 219, 220
hyperinflation 164, 171, 1 7 7 ( n l l ) industrial relations 79, 101-4
hysteresis effects 54, 209 industrial restructuring 110
industrial sector 71
immigration 6, 123, 129, 186, 191, industrialization 154
194, 201 'new Keynesian'view 172,
import costs 58 177(nl4)
import prices 74 industries 75, 130, 219
import surplus 11-12, 13, 16, 86 domestic 10, 173
imports 151, 156, 209, 215 'Industry' 95, 96-7, 99, 100, 102,
labour-intensive (or not) 127-8 l l l ( n 4 ) , 112(nl5), 157
impulse response analysis 213, 214t, composition l l l ( n 3 )
215, 216t 'leading role played by wage-setting'
income 2, 86 97
aggregate 232 real gross and net wages
disposable 124t, 125, 126t, (1972-2000) 96f
132-41 inequality 152, 164
national 132 across- and within-group 120, 121
net real 198 capital versus labour 118, 127,
personal 129, 134n, 134 128, 135, 136, 140-5
post-tax 218 consumer debt and financial
pre-tax 218 distress 132-45, 145-6(n4-9)

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Index 263

inequality - continued invalids 195, 204(n9)


explanation for rise 127-30 inventory valuation adjustment 125n
household troubles (USA) 5 , 1 1 7 - 5 0 investment 2, 4-7, 12, 50, 52-4, 56-7,
within labour 118, 128, 132, 59, 64, 73, 118, 130, 134n, 151,
136-7, 140-5 152, 154, 157, 174, 195, 202, 231,
and macro-economy 124-7, 232, 235, 236, 239, 241, 243(n3),

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
130-2, 145(n3) 244(n6, n l 4 ) , 247, 249, 252
trends 118-24, 145(nl-2) demand-effect 207
inflation 5, 10, 24t, 40, 49, 53, 56, domestic 173
58, 67, 80, 81t, 82t, 87(n8), effect of profits (Bhaduri-Marglin
108-9, 113(n20), 153, 156-7, 162, model) 6, 2 0 6 - 2 2
168, 171, 174, 219 full-employment 227
annual rates 45 Kaldor model 227f, 227
government-targeted 105 private 72, 103, 172
inflation targets 85-6, 87(n5), 221 profit-effect 207
inflationary expectations 11 public 85, 103, 172
inflationary pressure 3, 59 slowdown 127, 207
inflationary processes 71 investment rates 53, 215-16, 219
inflationary tendencies 68 investment-saving (IS) balance 227,
inheritance tax 237 230-8, 242, 242f, 243f, 244n(9),
innovation 68, 130-1 249-9
labour-substituting 195 Ireland 23-9, 3 2 - 3 , 37t, 4It, 43, 45,
link with rising inequality 5, 145 46, 46(n2), 47(n7), 5 1 , 52, 80,
institution-building 152 176t, 190t, 192t
institutional changes 102-8, IS-MM model 241-3
113-14(nl8-26) Italy 4, 57, 80, 112(nl0), 176t, 188t,
institutional investors 127, 128, 174 190t, 191, 192t, 196-8, 201,
institutions 218, 219, 220 204(nll)
multilateral 152, 176(nl) rates (1979-2000) 107t
insurance companies 128 unemployment: causes 23,
insurance funds 18 24t-26t, 36, 37t, 41t
interest 118, 133, 134
interest costs 87(n7) Japan 13, 67, 176t, 186, 188t, 190t,
'action parameter of monetary 191, 192t, 197t, 198, 201,
policy' 70 204(nll)
equilibrium real 87(n5) unemployment: causes 23,
nominal 87(n5) 24t-26t, 37t, 4 I t
real 86(n5) 'job-less'recovery 118,127
tool of monetary policy 70
interest rates 3, 12, 2 1 , 22, 24t, 27, 71, Kaldor model 227f, 227
72, 104, 114(n26), 134, 173, 196 Kaldor-Pasinetti model 244 (n 14)
CPI 45 Kaldor-Pasinetti-Kalecki model
long-term 57 235f, 236f
nominal 85, 106 Kaleckian models 6f, 207, 217, 226,
real 16, 24t, 38-40, 43, 47(n4), 56, 228-9, 231-2, 244(n4, 14)
85, 105, 219 Kaleckians 224, 240
short-term 23, 45 Keynesian multiplier 29, 46, 248
International Monetary Fund (IMF) Keynesian-Kaleckian model 213
50, 67, 68 hypotheses confirmed 217

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
264 Index

Keynesianism 6, 16, 17, 46, 57, 68, impact of demographic changes


172, 196, 248-52 193-5, 203-4(n6-9)
Korea, Republic of 176t, 188t, 189t, imperfect 70
190t, 192t, 197t, 207, 214-17, interaction with goods market 212
218 internal 130
multi-foreign exchange system Italian 102

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
219 malfunctioning 14
structural problems 220 neoclassical hypotheses 6, 21 It,
213-14, 214t, 216t
labour 5, 53-4, 112(nl2), 127, 138, people voluntarily outside
152, 173, 177(nl6), 181, 194 203(n3)
bargaining position and power persons outside 195
129, 193, 195, 209 rigidities/sclerosis 20, 23, 50, 52
constant average product 229 structural characteristics/factors 2,
exploitation 203 (n6) 70
share of national income 137 USA 130
'unreasonable' demands 130 labour market, demand-led
unskilled 127, 152, 153, 167, 171, (hypothesis) 21 I t
172, 177(nl3) 'strong support' 213, 214t, 216t
see also workers labour reserve army 11, 193, 194,
labour costs 71, 84, 217 195, 201, 203(n6)
actual 102 'industrial reserve army' 6, 182,
non-wage 88(nl8) 195, 202
rates of change (1979-2000) 107t labour scarcity 182, 194, 195
trends 80 labour supply 6, 54, 94, 112(nl2),
labour demand 15, 93, 127, 189, 124, 182, 193-6, 201, 203(n6)
193, 195, 203(n6), 215, 217, 218 abundant 189
diminished 194 additional sources 202
labour force/workforce 16, 55, 157, effective 203(n3)
172, 195 potential 187, 196, 203(n3)
down-sizing 12 scarce 189
domestic 6 scenarios (2000-50) 190t
family members (number) 123 shrinking 202
inadequately skilled 50 labour supply shocks 209
participation 189, 200, 201 labour surplus 194
reproduction 193 labour unions see trade unions
total 102 labour-market bargaining power 225
labour income 118, 135-8, 145(n3), labour-market conditions 4, 93, 94,
172 96, 98, 108-10, l l l ( n 2 , n6),
labour income share 75, 76f, 85, 113-14(n25), 217
87(nl5), 118, 119t, 132, 135, broader view 100-2,
140, 141 112-13(nl5-17)
labour market/s 1, 17, 29, 54, 64, labour-market conflict 224, 228,
101, 204(n7), 215, 217, 224, 231 232, 234, 236, 240
ageing (economic impact) 201 labour-market deregulation 206
demand-driven 208 'not sufficient condition for
deregulation 50 improvement in employment'
female 196 52
impact of ageing 6 labour-market 'distortions' 206

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Index 265

labour-market flexibility 1, 20-2, disequilibria 153, 163, 164


50-3, 6 4 ( n l - 2 ) efficiency 156
current focus 'overstated' 64 environment 219
labour-market flexibility and factors 3, 22, 36, 52
economic expansion 2 - 3 , 9-19 forces 37
policies for high employment imbalance 164

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
16-18 mismanagement 165
theories and counter-theories for outcomes 36, 240
economic policies 13-16 perspective 13
labour-market institutions 3, 22, 23, policy instruments 174
27, 25t-26t, 32-7, 44, 45, 49, 50, problems 78
52, 69, 195 prognostications 241
microeconomic 23 theory 14, 16
and income inequality 4 0 - 3 , variables 23, 28, 29, 38, 43
47(n8) macroeconomic background,
labour-market policies 46(n3), 233 unequal fortunes, unstable
labour-market reforms 51, 52, 54 households (USA) 124-7,
current focus 'overstated' 64 145(n3)
'failed to reduce u n e m p l o y m e n t ' macroeconomic policy 1-3, 21-3,
51 29, 38, 44f, 47(n4), 152, 236
Germany 64(n2) dysfunctional 22
labour-market regulations 206 expansionary 43, 44
labour-market spending 26t external constraints and
labour-standards, international 22 institutional changes 102-8,
labour-supply scenarios 187, 113-14(nl8-26)
189-93, 203(n3-5) need for coordinated international
laissez-faire 27, 44f, 45, 113(nl8) expansionary 221
laws/legislation 1, 49, 50, 51, 102 restrictive mix 84, 88(n24)
'Leontief's Paradox' 127-8 macroeconomic stability 153, 154,
Liapunov function 250 162, 165-6, 167, 171, 174
liberalization 151, 152 proxies 156-7
life-cycle 195, 228 macroeconomics 15-16, 17,
life-expectancy 186 224, 232
lira 103 macro-economy
devaluation (1992) 5, 101, 106, 108 inequality and 130-2
nominal exchange rate (1971-6) Keynesian construction 235
113(n25) manager-capitalists 225f, 225, 232,
loans 73, 125, 133, 141, 142, 239, 241, 244(nl5)
176(nl) managerial pay 232
non-bank 143 managers 15, 127, 131, 226, 240
Luxembourg Income Study 155, 175t professional 128
rising power 128
Maastricht Treaty 13, 108, 114(n26) manufacturing 56, 57, 107t, l l l ( n 3 ) ,
macroeconomic 127-8, 130, 153, 156, 172
balance 163 marginal productivity theory 223
conditions 27 marginal propensity to consume 132
consequences 5 marginal propensity to save 208
data 24t, 132 marginal theory 194
discipline 151, 152, 153 marginalist theory 112(n 12)

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
266 Index

mark-up 71, 73, 74, 87(nl3), 209, 'monetary constitution' 104


228-35, 241, 242, 244(n4, n l 4 ) monetary policy 12, 2 1 , 22, 49, 68,
counter-cyclical/pro-cyclical 244(n8) 70, 84, 86(n2-3), 87(n7), 88(n26),
mark-up pricing 71 85, 110, 174,213
market efficiency 221 deflationary 104
market failure 54 expansive 72

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
market forces 151, 152, 171 growth-friendly 86
competitive 223 national 173
'natural' 196 political economy 3 8 - 4 0
market ideology 15 restrictive 2, 13, 104
market mechanism 1 5 , 1 5 1 symmetrically effective 73
market share, international 77 tight 221
market system 69 m o n e y 72, 77, 100
'market worth' 15 endogeneity in a credit m o n e y
markets 10, 151, 165 economy 71
domestic/internal 11, 12, 18, 172 m o n e y aggregates 68
foreign/external 11, 18 m o n e y demand 72
flexible 2 money market rates 45
limited global 220 m o n e y supply 70, 71, 72
proxies representing 156 m o n e y wage rate 249
unregulated 176(nl) monopoly power 235, 235f
world 221 mortality 183
marriage 129 mortgage debt 124t, 124n,
Marshall Plan 10 125, 132, 133, 137, 138, 140,
Marshall-Lerner condition 72 245(nl6)
Marxism 11, 208, 224 real 135, 136t, 138t
mergers and acquisitions 173, regression results 138t
177(nl5), 235 see also consumer debt
metalworkers' trade unions 104 multipliers
methodological individualism 3, 14, cross-country Keynesian 46
15, 16 trade-based cross-country 43
microeconomic mutual funds 128
data 132, 145
factors 3, 52 NAIRU see Non-Accelerating
institutional settings 36 Inflation Rate of Unemployment
labour-market institutions 22, 32 national income 5, 53, 136t, 138t,
models 120 139t, 140, 140t, 144t, 145, 225f
policy 44 labour share 135
pricing decisions 230 natural resources 153, 167, 171, 172,
variables 28, 43 177(nl3)
migration 182, 185-7, 194, 195, neo-Kaleckian models 208, 221(nl)
203(n4) neo-Keynesians 14
scenarios 188t neo-liberalism 3
m i n i m u m wage 51, 69, 129, 233 neo-Malthusianism 195
MM (mark-up model) 230f, 235f, neoclassical economics 167,172,
236f, 242-3, 244(n8) 181, 189, 194, 195, 199, 204(n8),
'denotes profit-rate equation' 230 206, 212, 215
equilibrium schedule 234 hypotheses 6
monetarism 11 long-term growth theory 14

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Index 267

neoclassical economics - continued oligopoly 74, 173


substitution mechanisms 93, open economies 72, 202, 208,
110(nl) 221(nl)
synthesis 70 'opening clauses' 83, 88(n22)
theory 8 7 ( n l l ) openness 16, 29, 46, 77
two-factor models 151 Organisation for Economic

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
neoliberalism 156, 164, 172, 174, Co-operation and Development
206, 207, 220, 221 see OECD
distinguishing features 13-15 organizational learning 130
net capital transfers 134n Origo, F. 114(n29)
net equity issues 125t, 125n output 16, 53, 55, 70, 181, 194-6,
net transfers 133 200, 238, 244(nl4), 249
Netherlands 51, 52, 80, 86, 176t, average hourly real wage per unit
190t, 192t, 197t of 63
unemployment: causes 23, full capacity 71, 247
24t-26t, 37t, 37, 41t full employment level 224, 226, 227
'new consensus models' 70, per capita 198
86-7(n5) real wage per unit of 63
new growth theories 2, 7(n2) social 202
new Keynesianism 68, 70-3 output gaps 68, 87(n5), 213, 221 (n2)
New Zealand 23-7, 37t, 41t, 45, output growth 6, 22, 201
46, 47(n7), 52, 176t, 190t, 192t, per capita 202
197t real 35
Non-Accelerating Inflation Rate of output prices 74
Unemployment (NAIRU) (Layard, outsourcing 103, 128, 130
Nickell and Jackman) 4, 11, 49, overtime 88(n21)
50, 52, 53, 55-7, 59, 64, 70, 72, ownership 232-3, 237, 238-40,
73, 87(n9), 93, 110(nl) 244(nl4-15)
non-agricultural sector 215 manager-capitalists' share 232
workers' share 232
OECD (Organisation for Economic
Co-operation and Development) Pareto-optimality 15
22, 4 6 ( n l , n4), 182 pay-as-you-go (PAYG) 6, 194, 202
OECD countries 11, 51, 52, 56, 57, ageing and cost of 196-201,
64(nl), 130, 177(nlO), 190t, 191, 204(nl0-ll)
192t, 193, 197t, 198 financial imbalances 199
see also developed countries social sustainability 181
OECD economists 189, 203(n4) pension funds 18, 128
OECD u n e m p l o y m e n t (evidence on pension plans 237
causes) 23-35, 4 6 - 7 ( n l - 7 ) pension reforms 191, 196, 198
data 23-7, 45, 46(nl) pensions
oil price 10, 55, 58, 59, 103 abolition of real indexation to
Okun's law/coefficient 3 5 , 2 3 1 wages 196
old age 1 7 , 1 8 2 , 1 9 5 fully funded (FF) schemes 181
old-age dependency ratio 186, 187, public 204(nl0)
193, 198-201, 202 pensions and distribution in an
'overall economic dependency ratio' ageing society 6, 181-205
200 ageing and cost of PAYG 196-201,
see also dependency-ratio scenarios 204(nl0-ll)

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
268 Index

pensions and distribution in an post-war era (1945-) 9, 13, 44, 98,


ageing society - continued 108, 109, 112(nl2)
ageing and labour market 187, Post-Washington Consensus (PWC)
189-95, 203-4(n3-9) 5, 153, 154, 156, 165
conventional wisdom 182, 196, 'second-generation reforms' 152,
199, 201 171, 174

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
demographic trends 182-7, poverty 195
203(n2) absolute 172
'economic impact of ageing on 'exit' from 129-30
labour market and viability of poverty reduction 151
PAYG' 201 price competition 71, 87(n 13)
Phillips curve 86-7(n5), 109 price competitiveness 86
physical decline 182 price cuts 68, 72
Pigou effect 8 7 ( n l l ) price equation 112(nl2)
plant closures 130 price inflation 104
plant and equipment 55 price level 2, 71, 72, 77, 109, 228,
Poland 176t, 190t, 192t, 197t, 249
204(nll) price mechanism
political economists 87(n8) imperfect functioning 14
political economy 182, 223 price rigidities 72
monetary policy 3 8 - 4 0 price stability 1, 69
political trade cycles 11 price system 223, 224
population 182, 193, 194, price-indexation 114(n25)
203(nl) price-indexation clauses 106
active 200 price-setting 75
age structure 182 prices 4, 5, 18, 56, 58, 62t, 68, 7 0 - 3 ,
aged over fifty-five years 81t, 82t, 86-7(n5-12), 107, 108,
196, 198 151, 229
dependent 200 competitive 11
inactive 186 domestic 72
native 186 falling 8 7 ( n l l )
trends (global scenarios) 182-5, pricing 74, 93, 230
203 (n2) primary sector 5, 154
scenarios (2000-50) 190t private sector 13, 17, 57, 100, 106
see also working-age population privatization 129, 151, 173
population growth 17 7 (n 12) product market 14, 224, 231-2,
population replacement level 183-5 248-9
Portugal 51-2, 176t, 190t, 192t closure 244(n8)
unemployment: causes 23, product-market competition 228,
24t-26t, 37t, 41t, 46 229, 233, 234, 240, 241
post-industrial societies 1 7 , 1 8 product-market conflict
post-Keynesian models/approach 4, channel 240
70-3, 77 production 101, 108, 128, 173
open economy 208, 209t industrial 103
price theories 87(n6) labour-intensity 217
wages 87(n8) social act of 203 (nl)
Post-Keynesian-Kaleckian growth technical conditions 194
model 230 production costs 106
post-Keynesians 68, 69, 87(n9), 206 production process 194

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Index 269

production workers 95, 111 (n3-4) profits 56, 105, 118, 125, 127, 128,
hourly wages (rates of change) 95f 198, 202, 225f, 239-41,
productivity 1-4, 11, 12, 17, 18, 53, 244(nl4-15)
55, 56, 58, 59, 68, 84, 86, after-tax 125n
88(nl9), 97, 103, 105, 107, attributable to manager-capitalists
lll(n8), 112(nl2), 121, 127, 232, 238

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
131, 172,208,209,213,214, attributable to workers 232, 238
219, 220 corporate-retained 232
per capita 196, 198, 204(nll) distributed 234
coefficient 99 expected rate 72
per employee 8It, 82t retained 233, 234, 237
'and unemployment' 2111, 217 retained (as share of GDP) 237
rates of change (1979-2000) 107t total 237
productivity growth 6,71,75,79, profits: effect on investment and
80, 81t, 82t, 96, 98-9, 100, 106, employment (Bhaduri-Marglin
108, 117, 118, 126t, 196, model) 6,206-22
198-201, 202, 209t propensity to consume 195,
per capita 201 244(nl4), 247
rate 8In, 82n propensity to save 224, 241, 244(n6)
profit expectations 72 capitalists 230, 231, 235, 239-40
profit income 208, 237, 239, 241 capitalists (out of profit income)
profit levels 71 241
profit margin 3, 16, 18, 75, 76, 106, capitalists (out of wage income)
247, 249 241
per unit of sale 247 capitalists versus workers 227,
profit rate 99, 112(nl3), 130, 220, 244(n5)
226, 227, 229-31, 234-7, 239, profit income 226, 227, 247
241, 242f, 243f wages 226
current expected 208 workers 228, 233, 235, 244(nll)
decomposition 208 workers (out of profit income) 241
equilibrium 231 workers (out of wage income) 241
profit retention 238 property prices 68
profit share 6, 7, 54, 75, 76, 118, proprietors' income 136, 146(n9)
119t, 130,208-10,213-18, 'pseudo-curves' 93
221(n2), 225, 226, 228, 230, 232, pseudo demand and supply curves
234-41, 243(n3), 247-9 114(n28)
appropriateness as measure for 'public choice theory' 11
income distribution 218 public expenditure/government
effect on demand 215-16 spending 16, 104, 153, 219,
higher 'increases international 244(nl4)
competitiveness' 21 It allocation 17
Kaldor model 227f, 227 cuts 85
'profit squeeze' 249, 250, ratio of GDP 157, 158, 170
251t, 252
profit-led regimes 6-7, 207, 208, real estate 27, 28t, 3 It, 32-3, 38-9,
210, 21 It, 215, 218, 225-6, 234, 46-7(n4), 134n, 137, 141
236, 240, 242, 242f, 243(n2-3), real-wage labour market closure
247, 250-2 244(n8)
profitability 13, 57, 213, 218 recessions 20, 21, 55, 73, 75, 86, 127

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
270 Index

redundancy 101, l l l ( n 7 ) , 112(nl5) shareholders/equity ownership


regulatory quality 157 225, 238
remuneration 83, 88(nl8) shocks 4, 10, 29, 37, 50, 55, 57, 68,
growth rate 79 103, 209, 213-15
rent 118, 128 skills 120, 131, 172
replacement migration 186-7, social capital 152

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
188-9t social insurance system 129
'reservation wage' 14 social mobility 123-4, 129
reserve wage 204(n 7) social output 6, 200
reserve-army effect (hypothesis) 208, social pacts, national 79
209, 21 It social programmes 129
'no evidence' (developed countries) social protections 22
213, 214t social safety net 14, 233
'significant' (developing countries) social security 6, 70, 181
216t, 217 social spending/welfare expenditure
resource allocation 172, 174 l l l ( n 5 ) , 114(n26), 172, 173,
corporate 128 181, 199
restructuring 128 social stability 223
retirement 182, 186, 189, 198 social transfers 202-3
early 105, 196 socialism 3, 10
later 6 socio-political norms 152,
retirement age 182, 191, 195, 201 153, 154
retirement savings plans ('401(k)') Spain 23-9, 32-4, 37t, 37, 41t, 43,
128, 132 46(n2), 51-2, 59, 60t-61t, 80,
rule of law 157 176t, 190t, 192t, 197t
speed of adjustment 63, 251-2
safety nets 152 spillovers 3, 23, 52
salaries 79n, 87(nl3), 88(nl8), stability 248, 250, 2 5 I t
113(n23), 118 stability analysis 234-8, 241-3,
sales 3, 16, 118 2 4 4 ( n l l - 1 3 ) , 251
sales workers 131 Stability and Growth Pact 51, 85
savings 2, 7, 128, 196, 208, 218, 237, 'stable growth' mechanism 55
239, 247-9, 252 stagflation 10, 130
aggregate 231, 234 stagnation 4, 85, 86, 88(n25),
corporate 237, 239 98, 195
domestic 209t, 210 stagnationist regimes 210,216,
excess 238 243(n2), 244(nl2), 248, 250
foreign 210 'standard labour units' 111 (n3-4)
Kaldor-Kalecki approach 241 standard of living 10, 193
personal 124, 124t state l l l ( n 7 ) , 129
savings incentives 132 distribution claims 73
services 18, 56, 57 re-empowerment 152, 157,
Settlement countries 189, 190t, 191, 170, 171
192t, 201 state intervention/government
severance pay 103 intervention 2, 14-15
share options 225 state policies 218, 219, 220
share prices/price of equities 68, state role 13, 154, 174, 176(nl)
128, 238 state size 153, 157
share repurchases 1 1 8 , 1 2 5 , 1 2 8 stockmarkets 135, 245(nl6)

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Index 271

Stolper-Samuelson theorem 151,167 technological u n e m p l o y m e n t 2111


strikes 58, 59, 60t, 62t, 63, 79, 104 hypothesis confirmed 213, 214t
structural adjustment 206 technology 233, 244(nl0)
structural reforms 2 'neutral' 18
subsidies 105, 172, 219 new 229
substitution (of labour with capital) terms of trade 99, lOOt, 100,

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
211t 112(nl3)
hypothesis rejected 214t, 214 trade 5, 12, 129, 151-4, 157, 171,
supply and demand 9 4 , 1 1 0 , 1 1 8 , 174, 176(nl), 208, 209t, 212
127, 223 distributional impact 127
supply shocks 68 effect o n growth and employment
supply side 2, 49, 196, 252 214
Sweden 103, 176t, 190t, 192t, 196, international 9, 3 2 - 3 , 215
197t, 2 0 4 ( n l l ) source of international
unemployment: causes 23, competitiveness 214
24t-26t, 32, 35, 37t, 37, 41t US 127-8
Switzerland 23, 24t-26t, 37t, 4It, trade deficits 13, 127
190t, 192t trade intensity 127
'system of industrial relations' 34 trade liberalization 5, 162
trade openness 167
tax base 1 7 2 , 1 7 3 trade theory 127
tax burden 3, 10, 22, 35, 43, 173, trade u n i o n density 3, 22, 23, 25t,
182, 196 27, 28t, 30t, 32, 33, 35, 36, 38-43,
tax competition 173,174, 45, 47(n8), 50, 233
199, 202 trade union representation 128
tax coordination 202 trade unions 5, 20, 54, 71, 84,
tax incentives 172 103-5, 108, 110(n2), l l l ( n 3 ) ,
tax policy 130 112(nl0), 113(nl8-19, n 2 1 ,
tax rates n24), 114(n25-6), 127, 130,
low 27 206, 240
marginal 133 bargaining 4, 44, 79, 80, 83, 102
net 74 central bank waging war o n 3 8 - 4 0
overall 26t coordination initiatives 85-6,
tax reduction/cuts 2 1 , 35, 50, 130, 88(n27)
172 loss of political power 83
Tax Reform Act (USA, 1986) 133 membership 79, 104
tax returns 122 militancy 57, 101, 102, 104, 233
tax system l l l ( n 5 ) wage policy 69
tax wedge 57 weakness 95
taxation/taxes 58, l l l ( n 4 ) , 134n, trade volume 156, 170, 177(nlO)
198, 199, 218 changes 165-7
corporate and capital 173 training 17, 35, 121
corporation and income 172 Turkey 176t, 190t, 192t, 207,
general 199 214-17, 218, 219
Taylor rule 87(n5)
technical capital productivity 208 underemployment 55
technical/technological progress 6, unemployed 199
20, 94, 101, 112(nl2-13), 172, number 60t, 62t
201, 208, 209, 219 temporary 194

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
272 Index

unemployment 1, 6, 9-17, 75, empirical findings 29-34,


87(nl4), 120, 128, 129, 157, 46-7(n3-7)
203(n6), 206, 208, 214, empirical model 27-9, 46(n2)
217,223 employment (fair and full) 4 3 - 5 ,
adjusted standard 45-6 47(n9)
capital stock and 4, 4 9 - 6 6 features 33-4

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
cause 2 further interpreting the results 34-5
central cause 14 high employment (debate over
c o m m o n 46 causes) 2 0 - 3
conventional wisdom 43 importance of macroeconomic
definition 101, 105n, 112-13(nl6) forces 34
disguised l l l ( n 6 ) , 215 labour-market institutions and
effect of demand on 63 income inequality 40-3,
emanates from 'self-inflicted 47(n8)
dysfunctional macroeconomic macroeconomic variables 28, 32,
policy' 52 33,34
'equilibrium' rate 56, 58 microeconomic variables 28,
high 85 33,36
hysteresis effect 209 OECD u n e m p l o y m e n t (evidence o n
investment-enhancing policies 'can causes) 23-35, 4 6 - 7 ( n l - 7 )
lead the way in reducing' 64 political economy of monetary
juvenile 102 policy (central bankers versus
long-term 2, 57, 59, 60t, 62t, 63 unions) 3 8 - 4 0
'lowers wages' 211 principal contribution of chapter
mass 14, 79 22, 23, 29
m o n o t o n i c negative relationship quantifying the cause of changed
with capacity utilization 231 u n e m p l o y m e n t rates 3 6 - 8
'natural rate' (Friedman) 11, 21 time-series u n e m p l o y m e n t rate
OECD (evidence o n causes) 23-35, regression 3 0 - l t
46-7(nl-7) u n e m p l o y m e n t benefits 3, 14,
'reduced fear of job-loss' 10 20,21
'reserve army of labour' 11 duration 22, 27
'significantly determined by capital real 60t, 63
shortages' 56-7 real level 58, 59, 64(n3)
standardized 46(nl) value and duration 50
structural 172 u n e m p l o y m e n t insurance
total 51 compensation 233
voluntary/involuntary 14, 194, unemployment, productivity, and
204(n7) institutions: influence o n real
unemployment: causes (sclerosis wage trends (Italy, 1970-2000)
versus macroeconomic policy) 4_5 ; 93-116
3, 2 0 - 4 8 cross-regional studies 110,
appendix 4 5 - 6 114(n29)
'caused by macroeconomic external constraints,
factors' 44 macroeconomic policies, and
country-specific effects 28-9, 32, institutional changes 102-8,
43, 47(n4) 113-14(nl8-26)
data 23-7, 46(nl) political and institutional factors
economic policy m e n u 44f 110

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Index 273

unemployment, productivity, and structural changes in economy


institutions - continued 129
predictive purposes (down-played) tax changes 237
109 unemployment: causes (sclerosis
socio-political context 102, 106, versus macroeconomic policy)
108, 110 20, 2 1 , 23-9, 35, 37t, 37, 41t,

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
wages 'not affected by 44f; 44 ; 46 ; 47(n5, n8)
u n e m p l o y m e n t levels' inequality 127-45, 145-6(n4-9)
(contradicted) 110 inequality trends 118-24,
unemployment rate 4, 20-3, 29, 35-8, 145(nl-2)
45, 46(n3), 55, 62t, 70-2, 87(n9), macro-economic background
96, 97, 99-101, 108, l l l ( n 6 ) , 124-7, 145, 145(n3)
112(nl0), 112-13(nl6), 142-3, 191, national income 119t
209t, 209, 213, 221(n2), 231 non-financial corporate resources
union wage coverage 26t, 27 (1953-2004) 125t
unionization 128, 129 savings and consumer debt
unit costs 11, 16, 18 (1949-2004) 124t
unit labour costs 4, 68-73, 106, US economy 226, 236-7, 240,
113(n25), 208 243(n3)
causes of observed trends 70, US Federal Reserve 2 1 , 38
79-84, 88(nl8-23) Board of Governors (BOG) 69,
growth rate 79, 85 135, 141, 143
increases (explanation for increases
in inflation) 76 wage anchor 69, 73
lower 'increase international wage bargaining 22, 103, 104, 106,
competitiveness' 21 I t 219
stable 71 coordination 3, 26t, 27, 71, 87(n8)
trends 86(n4) decentralization 2
unit-labour-costs growth, deflation risks in Germany and
inflation and 73-8, 8 7 - 8 ( n l 3 - 1 8 ) EMU 4, 6 7 - 9 2
United Kingdom 23-7, 37t, 41t, see also bargaining
47(n7), 51-3, 104, 113(n22), 176t, wage bill 7, 236f, 236, 237, 239, 240
188t-190t, 192t, 197t, 198, distribution 225
2 0 4 ( n l 0 - l l ) , 207, 213, 214t manager capitalists 232-3
unemployment 'significantly workers 232-3
determined by capital wage claims 10, 18
shortages' 56-7 wage concessions 97
United Nations 183, 185 'wage curve' 94, 112(nl2)
United States of America 7, 12, 13, wage cuts 172
56, 57, 104, 108, 176t, 186, wage demands 97
188t-190t, 191, 192t, 197t, wage determination 93, 94, 97, 110,
203(n4), 207, 213-14, 242 112(nl2), 226, 232, 240
inequality of household income 5 wage distribution channel 225
'more wage and income inequality 'wage drift' 81t, 82t, 113(n24)
t h a n any other OECD country' negative 83, 84, 88(n21)
130 wage flexibility 11
productivity, labour costs, exchange insufficient downward nominal 69
rates (1979-2000) 107t wage growth 71, 108, 130
real GDP growth rate 46 wage income 208, 232, 241

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
274 Index

wage increases 80, 88(nl9), 98 decrease 'increases employment'


margins 99, 112(nl3) 21 It (hypothesis rejected, 214t,
wage indexation 103, 105, 113(n20) 217,220)
imperfect (hypothesis supported) deflation risks in Germany and
211t, 213, 214t EMU 4, 6 7 - 9 2
wage inequality 118, 120, 123, 138, dual function 207

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
146(n9) effect of demand o n 63
wage labour 203 (nl) falling 72
wage led regimes 6-7, 12, 207, 208, final 196
210, 21 It, 215, 218, 225-6, 234, flexibility 21, 201
236, 240, 243ff, 243(n2), 247, gender inequality 120
250-2 gross 96
wage level 193, 244(nl4) growth patterns 131
wage moderation 103, 113(nl8-19) hourly 80, 88(nl9), 95f, 121
wage negotiations 113 (n20) increase 'leads to substitution of
wage policies, dilemma in Germany labour with capital which in
and Europe 84-6, 88(n24-7) turn increases productivity'
wage-price spiral 107 21 It (hypothesis rejected, 214t,
wage rate 193, 194, 204(n7) 217)
direct and social 195 increases 103
flexibility 195 industrial 97, 98t, 99f, 108
real 18 laggard 118
wage-replacement rate 27 low 118, 130, 217
wage restraint 12, 18, 104 manager-capitalists 225f, 225,
excessive 69, 85, 86 243(nl)
wage rigidities 73 net 96f
nominal 68, 69 nominal 58, 68-75, 107, 229
wage-setting procedures 107, 108 pivot variable 97
wage share 54, 215, 217, 219, 236, rates of change 97, 98t, 99f, lOOt,
240 112(nl0)
wage trends real 10, 11, 18, 53-4, 58, 68-70,
collectively agreed 83 94, 97-103, 105-6, 108, 110,
influence of unemployment, 112(nl2-13), 114(n25), 172,
productivity, and institutions 199, 206, 209, 218-19, 228,
(Italy, 1970-2000) 4-5, 93-116 231, 244(n8), 247, 249
wages 1-8, 14, 56, 57, 60t-61t, 7 0 - 3 , real (changes in characteristic
121, 127, 129, 198, 202, 213, 227, behaviour) 109
238, 241 real (per unit of output) 58
actual 84 real (Phillips curve) 109,
average 196 114(n28)
average hourly real (per unit of real (rates of change) 111 (n8)
output) 63 real gross 96f
check o n 194 relation to capital stock 58
collectively agreed 8It, 82t, 84 rigid nominal 69
'compensation' 87(nl5), 117, 118, short-term variations (not of
119t, 128 interest) l l l ( n 9 )
consumption assumption 247 stagnation 108
contractual 95, 97, 98, 102, 'too high' 206
lll(n3) upward pressure 54

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger
Index 275

wages - continued bargaining power 100, 106,


workers 225f, 225 113(n20, n22), 236
see also earnings blue-collar (bias against) 128
wages, employment, distribution and 'forced saving' 249
growth 1-8 immigrant 102, 113(nl7)
thesis of book 2 mature 200

Copyright material from www.palgraveconnect.com - licensed to University of Newcastle, Australia - PalgraveConnect - 2014-03-15
wages policy 68-71, 73, 79, 80, median 120
85,86 old/older 195, 201
anchor for combating inflationary/ ownership share of capital stock
deflationary tendencies 68 244(n6)
Washington Consensus 22, 45, problem of definition 243 (nl)
152, 156 skilled 64, 113(n20)
'first-generation reforms' 174 white-collar lll(n3), 113(n20)
wealth 2, 72, 73, 87(nll), 113(n20), see also employees
124, 135, 237 working hours/working time 83,
wealth tax 237 88(n21), 101, lll(n3), 123, 124,
welfare capitalism 3, 10 201
welfare economics 15 short-time 88(n21)
welfare state 1, 10, 14, 130, 181, 199 total lll(n7)
Western Europe 3, 9, 20, 22, 43 working-age population 10, 12,
whites (USA) 120-1 182, 187, 189, 191-6, 198,
women 120, 121, 123, 124, 183, 191, 203(n3), 221 (n2)
194, 195, 200 ageing (developed countries)
worker rights 22 185-6
worker spending out of wages World Bank 156
244(nl4) data-set 155, 175t
worker spending out of worker homepage 45
income 244(nl4) website 157
workers 3, 10, 44, 96, 104, 112(nl5), World War 11,13
121, 130, 193, 198, 202, 208,
224-7, 232, 236f, 237, 238, 240, younger generations 187,191
244(n5, nl5), 250, 251t youth 195, 199, 201, 202

10.1057/9780230371781 - Wages, Employment, Distribution and Growth, Edited by Eckhard Hein, Arne Heise and Achim
Truger

Potrebbero piacerti anche