Economic Policy Proposals for Germany and Europe

Economic Policy Proposals for Germany and Europe Despite exporting more good and services than any other country in th...

Author: Ronald Schettkat | Jochem Langkau

23 downloads 1061 Views 861KB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

Economic Policy Proposals for Germany and Europe

Despite exporting more good and services than any other country in the world, economic growth in Germany has been slow through the nineties and the early twenty-ﬁrst century with low wage growth, rising unemployment and increasing public deﬁcits. German unemployment was traditionally diagnosed as structural, neglecting macroeconomic causes of economic stagnancy in the economic policy debate. This book offers a fresh, innovative analysis of the German economic policy debate, containing essays from eight distinguished international economists. These essays tackle various aspects of the German and European market, ranging from theoretical issues criticizing the narrowness of the debate, analyses of the real effects of monetary policies in the short and long run, ﬁscal policy contributions, wage policies, to family policies, arguing for a more expansionary macroeconomic policy to counteract economic stagnancy and improve prosperity in Germany and Europe. This book will be of great interest to students and researchers engaged with macroeconomic policy and monetary economics, as well as policy makers within the European Union. Ronald Schettkat is Professor of Economics at the Bergische Universität Wuppertal in Germany. Jochem Langkau recently retired as the Director of the Economics Department in the Friedrich-Ebert-Stiftung, Bonn, Germany. Financial support by the Friedrich-Ebert-Stiftung, Bonn, Germany, is gratefully acknowledged.

Routledge studies in the European economy

1 Growth and Crisis in the Spanish Economy, 1940–1993 Sima Lieberman 2 Work and Employment in Europe A new convergence? Edited by Peter Cressey and Bryn Jones 3 Trans-European Telecommunication Networks The challenges for industrial policy Colin Turner 4 European Union – European Industrial Relations? Global challenges, national developments and transnational dynamics Edited by Wolfgang E. Lecher and Hans-Wolfgang Platzer 5 Governance, Industry and Labour Markets in Britain and France The modernizing state in the mid-twentieth century Edited by Noel Whiteside and Robert Salais

6 Labour Market Efﬁciency in the European Union Employment protection and ﬁxed-term contracts Klaus Schömann, Ralf Rogowski and Thomas Kruppe 7 The Enlargement of the European Union Issues and strategies Edited by Victoria Curzon-Price, Alice Landau and Richard Whitman 8 European Trade Unions Change and response Edited by Mike Rigby, Roger Smith and Teresa Lawlor 9 Fiscal Federalism in the European Union Edited by Amedeo Fossati and Giorgio Panella 10 European Telecommunications Liberalisation Edited by Kjell A. Eliassen and Marit Sjøvaag

11 Integration and Transition in Europe The economic geography of interaction Edited by George Petrakos, Gunther Maier and Grzegorz Gorzelak 12 SMEs and European Integration Internationalisation strategies Birgit Hegge

15 Russian Path Dependence Stefan Hedlund 16 The Impact of European Integration on Regional Structural Change and Cohesion Edited by Christiane KriegerBoden, Edgar Morgenroth and George Petrakos

13 Fiscal Federalism and European Economic Integration Edited by Mark Baimbridge and Philip Whyman

17 Macroeconomic Policy in the European Monetary Union From the old to the new stability and growth pact Edited by Francesco Farina and Roberto Tamborini

14 Financial Markets in Central and Eastern Europe Stability and efﬁciency Edited by Morten Balling, Frank Lierman and Andy Mullineux

18 Economic Policy Proposals for Germany and Europe Edited by Ronald Schettkat and Jochem Langkau

Economic Policy Proposals for Germany and Europe

Edited by Ronald Schettkat and Jochem Langkau

First published 2008 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN Simultaneously published in the USA and Canada by Routledge 270 Madison Ave, New York, NY 10016 This edition published in the Taylor & Francis e-Library, 2008. “To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.” Routledge is an imprint of the Taylor & Francis Group, an informa business © 2008 Selection and editorial matter, Ronald Schettkat and Jochem Langkau; individual chapters, the contributors All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data Economic policy proposals for Germany and Europe/edited by Ronald Schettkat and Jochem Langkau. p. cm. – (Routledge studies in the European economy; 18) Includes bibliographical references and index. Germany–Economic policy. 2. Europe–Economic policy. I. Schettkat, Ronald. II. Langkau, Jochem. HC286.8.E263 2008 339.50973–dc22 2007041760 ISBN 0-203-92853-9 Master e-book ISBN ISBN10: 0-415-46084-0 (hbk) ISBN10: 0-203-92853-9 (ebk) ISBN13: 978-0-415-46084-2 (hbk) ISBN13: 978-0-203-92853-0 (ebk)

Contents

Notes on contributors

1

Introductory summary: prosperity for Germany and Europe

viii

1

RONALD SCHETTKAT

2

Broadening the discussion of macroeconomic policy

20

ROBERT M. SOLOW

3

Monetary policy and the real economy

29

PAUL DE GRAUWE AND CLÁUDIA COSTA STORTI

4

Germany in the monetary union

54

CHARLES WYPLOSZ

5

Reforms, macroeconomic policy and economic performance in Germany

72

WENDY CARLIN AND DAVID SOSKICE

6

Exportweltmeister – so what? Better goals for German foreign economic policy

119

ADAM S. POSEN

7

Wanted: a new German Wirtschaftswunder

144

RICHARD B. FREEMAN

Index

167

Contributors

Wendy Carlin is Professor at the University College London and was born in Australia. Together with Philippe Aghion, she is the managing editor of “Economics in Transition”. She is also a member of the scientiﬁc panel of the journal Economic Policy. In 2005, Professor Carlin has been elected to the Council of the Royal Economic Society and was appointed as a member of the Academic Advisory Board for the Anglo-German Foundation. Wendy Carlin is also a Research Fellow of the Centre for Economic Policy Research, London. Cláudia Costa Storti is Senior Economist at the Central Bank of Portugal in Lisbon. Previously, she held research positions at the University of Lisbon and the Catholic University Leuven, Belgium. She worked for the European Commission and hosted a program on the Economy for the Portuguese television. She has published numerous articles on monetary policy. Richard B. Freeman holds the Herbert Ascherman Chair for Economics at Harvard University. He is also the director of the Labor Studies Program of the National Bureau of Economic Research (NBER). Professor Freeman is a member of the American Academy of Arts and Sciences and was, among others, the Okun Lecturer at Yale University. Also, he was a Fellow of the Russel Sage Foundation, New York. Professor Freeman is known for his innovative analyses of labor markets and his detailed knowledge thereof. He has published more than 300 articles and numerous books. Paul de Grauwe is Professor at the Belgium Centre for Economic Policy of the Catholic University Leuven, Belgium, and a Research Fellow at the Centre for Economic Policy Research, London. He has been a visiting professor at numerous European and US universities, among others at Wharton School, Philadelphia, Institut für Weltwirtschaft, Kiel, International Monetary Fund, Bank of Japan, Free University Berlin, Humboldt University Berlin. Also, Professor de Grauwe served as a Belgian Parliament Senator, Parliament Member of the House and Chairman of the Economic and Finance Committee. He is a member of the group of Economic Policy Analysts advising the European Commission President Barroso.

Contributors ix Jochem Langkau was until 2006 the Director of the Department for Economic and Social Policy Research of the Friedrich Ebert Foundation. He has published numerous articles on economic, employment and labor market policy. Adam S. Posen is Deputy Director of the Peterson Institute for International Economics in Washington, DC, where he has been a Senior Fellow since 1997. His research covers mainly macroeconomic policy under globalization, G-3 (the United States–EU–Japan) economic relations and central banking issues. Previously, Adam Posen worked as an economist at the Federal Reserve Bank, New York, was Okun Memorial Fellow at the Brookings Institution and a Bosch Foundation Fellow in Germany. He is a member of the American Council on Germany and the cofounder of the refereed journal International Finance and an executive editor of The International Economy magazine. Ronald Schettkat is Professor of Economics at the University of Wuppertal, Germany. Before, he held a chair at Utrecht University, the Netherlands, and was a Fellow at the Russel Sage Foundation, New York. Professor Schettkat was also a visiting professor at numerous European and American universities, among others, at Stanford University, University of California at Berkeley, Netherlands Institute for Advanced Study, Tinbergen Institut, Universities of Bologna and Modena, as well as at the Wissenschaftszentrum Berlin. He also served as an advisor for governments and international organizations. Robert M. Solow is Institute Professor Emeritus at MIT and Nobel Laureate in economics. He holds honorary PhDs of renowned American and European Universities. Professor Solow was a member of the Council of Economic Advisors (along with Walter Heller (director), James Tobin, Kermit Gordon and Arthur Okun) advising US President John F. Kennedy. Currently, he is the Foundation Fellow of the Russel Sage Foundation, New York. In addition to his research on growth theory, his works on labor markets, environmental issues and natural resources have also strongly inﬂuenced economic thinking. David Soskice is School Centennial Professor of European Political Economy at the London School of Economics, Professor of Political Science at Duke University and was Research Professor at the Wissenschaftszentrum für Sozialforschung in Berlin (WZB). He has been visiting professor at numerous universities, among them the University of California at Berkeley, Yale, and Harvard University. His ﬁeld of research concentrates especially on varieties of capitalism, comparative political economy and macroeconomics and labor markets. Charles Wyplosz is Professor of International Economics at the Graduate Institute for International Studies in Geneva where he is Director of the International Centre for Money and Banking Studies. Previously, he has served as Associate Dean for Research and Development at INSEAD and Director of

x

Contributors the PhD program in Economics at the Ecole des Hautes Etudes et Science Sociales in Paris. He has also been the Director of the International Macroeconomics Program of CEPR, London. His main research areas include ﬁnancial crises, European monetary integration, ﬁscal policy, and economic transition. Currently, he is also a member of the Group of Independent Economic Advisors to the President of the European Commission.

1

Introductory summary Prosperity for Germany and Europe1 Ronald Schettkat

Unnecessarily narrow economic policy debate in Germany No country in the world exports more goods and services than Germany. Success in international markets is a concrete indicator of the competitiveness of the German economy, especially when prices of German products are comparatively high. At the same time, however, German GDP per capita growth was the lowest in the EU in the 1990s and at only 1.4 percent in the period 1991–2004 was close to that of the long stagnant Japanese economy (where the average annual GDP per capita growth was 1.1 percent in that period). In the years since 2000, the performance of the German economy has not been any better (see Table 1.1): its tremendous success in world markets has been accompanied by poor growth in domestic markets, stagnating wages, rising unemployment and increasing public deﬁcits. Why did the German economy stagnate for such a long time? Why did its enormous success in world markets not translate into strong domestic growth? The most popular hypothesis for Germany’s economic problems – strongly inﬂuenced by the Bundesbank and the ECB, business lobby groups like the “Initiative für Soziale Marktwirtschaft” and almost universally accepted among German politicians – states that the German labor market is overly rigid. “Labor market rigidities” or “structural rigidities” prevent the economy from growing and force the country into stagnation and high unemployment. According to this hypothesis, the economic problem of Germany is structural. The structure of the economy and its institutional framework are causing the slack. To bring Germany back on an upward growth path requires radical and painful changes: labor markets need to be made more ﬂexible, taxes need to be reduced and transfer payments need to be lowered (for recent elaborations in this direction, see Siebert 2005; Sinn 2007). Given this inﬂuential diagnosis, the medicine prescribed by many German economists, the OECD’s jobs study, IMF and others consisted of supply-side treatments only. It is not that German governments ignored these recommendations, but until 2006 “reforms were visible everywhere but not in the economics statistics”. Although almost universally accepted, the “structural rigidities” hypothesis is based more on theoretical assumptions and anecdotal evidence rather than on

2

R. Schettkat

Table 1.1 Average annual growth rates, different periods Germany GDP GDP per capita (15–64) Contributions Productivity (hour) E-pop (15–64) Working time GDP GDP per capita (15–64) Contributions Productivity (hour) E-pop (15–64) Working time GDP GDP per capita (15–64) Contributions Productivity (hour) E-pop (15–64) Working time

1991–1995 1.4 1.4

USA

UK

France

Sweden

Japan

2.5 1.3

1.7 1.6

1.2 0.9

0.7 0.3

1.5 1.2

1.4 –0.7 –0.3 1996–2000 2.0 2.0

1.3 –0.2 0.1

1.6 –0.9 –0.4

0.9 –0.3 –0.6

0.3 –2.7 0.8

1.2 0.4 –1.5

4.1 2.7

3.2 2.8

2.8 2.4

3.2 2.9

1.0 1.2

2.0 0.8 –0.8 2001–2004 0.6 0.9

2.4 0.4 –0.2

2.4 0.8 –0.4

2.0 1.2 –0.8

2.4 0.5 0.0

1.7 0.2 –0.7

2.3 1.0

2.5 1.8

1.6 1.1

2.1 1.4

1.1 1.5

1.1 0.1 –0.4

2.3 –0.8 –0.4

2.0 0.2 –0.5

1.9 0.1 –0.9

2.4 –0.3 –0.6

2.1 –0.1 –0.4

Source OECD 2003 (Sources of Economic Growth: 36).

hard empirical facts, and it cannot explain the recent recovery of German growth rates. In the summer of 2006, the German economy showed signs of recovery: GDP growth reached about 2.5 percent on an annual basis, even full-time employment covered by social security increased and the situation of public budgets relaxed. Also, a 3 percentage point increase in the German VAT did not slow economic growth in 2007. For the ﬁrst time in ﬁve years, the ﬁnance minister could inform Brussels that public deﬁcit in 2006 was below the 3 percent threshold of the European Stability and Growth Pact (the so-called Maastricht criteria) and core inﬂation2 is, at 0.6 percent, low and not showing any sign of acceleration. In 2007, the EU Commission formally released the deﬁcit procedure against Germany, and predictions for annual growth of the German economy cluster around 3 percent. How can this be if “structural rigidities” prevented the economy from growing? One answer is, of course, that past reforms – the marginal tax rate has been lowered substantially, unemployment beneﬁts for those unemployed for a year or longer have been lowered, job-acceptance criteria have been sharpened and now any job is declared to be an acceptable job, unions have lost inﬂuence and are engaged in concession bargaining, public budgets have been reduced, etc. – have made the economy more ﬂexible. Of course, the government has a preference for this view because it implies that their policy was on the right track. But timing is a problem with this answer. Why did growth and employ-

Introductory summary 3 ment pick up now and not earlier? Why did economic growth recover ﬁrst and only later employment? Alleged time lags between treatment and effect can solve the puzzle formally, but they are a weak argument because the right time lag allows any current outcome to be “explained” by past events. Unless one has a good explanation for time lags that have different lengths for different measures, only an already strong believer in these policies will be satisﬁed with this answer. Another explanation of the upswing in 2006 is that the grand coalition changed the gear with respect to ﬁscal policy. Courageously, the new ﬁnance minister, Peer Steinbrück, an economist, openly gave up the “savings strategy” and argued that the federal budget of 2006 should be expansionary in order to stimulate the economy, which would consolidate public budgets and enable Germany to meet the 3 percent deﬁcit criteria in the future. Although the conventional economic litany started immediately, ranging from the “straw-ﬁre argument” to allegations that the 2006 federal budget is unconstitutional, Steinbrück was counting on an expansionary push that would be strong enough to overturn the contracting effects of the 3 percentage point rise of the value added tax in 2007. In the event, GDP growth for 2006 was 2.5 percent, tax revenues are rising, employment covered by social security is growing by about 600.000 in 2006/2007 and unemployment has dropped substantially below the 10 percent threshold (6.4 percent according to ILO deﬁnitions; see Bundesagentur für Arbeit 2007). Could it be that the German economic stagnation was not only caused by “structural rigidities” but that it was also caused by contracting macroeconomic policy? Was the savings strategy of Finance Minister Hans Eichel (minister for ﬁnance 1999–2005) the wrong approach? Could an expansionary ﬁscal and monetary policy have improved the economic situation? Such questions seem to be rare among German economists, where the common view seems to be based on the assumption that the control of inﬂation is not only necessary but also sufﬁcient to achieve favorable macroeconomic outcomes. The analyses in this volume, however, show that past ﬁscal policy in Germany was clearly contracting and drove the German economy even deeper into stagnation. Robert Solow (“Broadening the discussion of macroeconomic policy”, this volume) argues that part of the explanation for misguided economic policy in Germany is the intellectual debate, which disregards macroeconomics and the demand side almost entirely. Being a Keynesian is a synonym for being a species soon extinguished in the world of German economists. Macroeconomic policy is discarded as wrong-footed; big spender, “big state” economics cannot deliver a useful contribution in the new world of global capitalism. According to the common view among German economists, politicians and laymen, expansionary economic policies, either ﬁscal or monetary, are ineffective to stimulate economic growth and employment even in the short run. All expansionary economic policy can achieve is a straw-ﬁre, expansionary impulses will soon end in accelerated inﬂation without substantial effects in the real economy. This view, however, is based on the assumption that the economy

4

R. Schettkat

is always in Walrasian equilibrium, i.e. in a situation in which all individuals optimize their behavior and maximize individual utility, and markets clear. Then, the employed are best off when working and the unemployed are best off when unemployed. To get the unemployed into jobs requires different labor supply incentives. A theoretical position developed by Milton Friedman (1968) and Edmund Phelps (1968), which became known as the natural rate of unemployment and later as the NAIRU, posits a non-accelerating inﬂation rate of unemployment or structural unemployment rate to which the economy always returns in the long run. Accordingly, there is a strong belief that structural unemployment rates are very high in Germany (see OECD estimates displayed in Table 1.3 below) and that it hence requires structural reforms to improve the functioning of labor markets. The major if not the only impediment for higher employment lies in labor market institutions. Attempts to stimulate economic activity through ﬁscal or monetary policy would fail according to this view, because labor market rigidities will prevent quantities from reacting, i.e. unemployment is diagnosed as structural in the sense that everybody is optimizing within the existing institutional framework. In other words, they hold that deviations of actual from potential output, even fairly large ones, are quickly and automatically self-correcting if only inﬂation is kept low and steady, and wages are ﬂexible. Inﬂation is a purely monetary phenomenon and unemployment results only from excessively high real wages. So there is no stabilization function for ﬁscal policy to perform. (Solow, this volume) Was the German economic policy debate with its focus on structural reforms misguided; did it follow the wrong track? Was it a mistake to disregard macroeconomic policy options and to emphasize structural reforms only? The unanimous answer of the distinguished economists who contributed to this volume is a clear YES: a more expansionary economic policy could have improved the economic situation in Germany and Europe substantially. None of the analysts, however, take the position that supply-side measures improving the functioning of the economy are unnecessary; rather, by themselves they are unlikely to create economic growth – they may even produce adverse effects. Robert Solow, winner of the Nobel Prize for his contribution to growth theory, reminds us of the fundamental distinction between potential and actual economic output. Whereas potential output depends on supply-side variables, actual growth depends largely on demand. Markets are like scissors, they cut with two blades, supply and demand (Marshall 1920). If demand falls short of potential output, actual growth will be below potential growth, and thus some production capacity will be idle – Okun’s law – reﬂected in unused capital and unemployment. If there is idle capacity, the economy can expand without substantial upward pressure on prices. “Clear thinking requires a reminder: the reward from more efﬁcient markets is higher potential output. There is no guarantee that an increase in

Introductory summary 5 Table 1.2 Components of growth (annual averages, excluding stockbuilding, %) GDP 1991–1995 1.6 2.4 1.5 1.2 0.6 1.6 1996–2000 Germany 2.1 United States 4.0 United Kingdom 3.3 France 2.6 Sweden 3.2 Japan 1.0 2001–2005 Germany 0.7 United States 2.6 United Kingdom 2.4 France 1.5 Sweden 2.5 Japan 1.5 Germany United States United Kingdom France Sweden Japan

I-Gov. I private C-Gov. C private Exports Imports 0.0 0.0 0.1 0.0 0.2 0.4

0.2 0.6 –0.1 –0.2 –1.0 –0.6

0.5 0.1 0.2 0.6 0.3 0.5

1.2 1.8 0.9 0.4 –0.1 1.3

0.4 0.6 1.1 1.0 2.0 0.3

0.7 0.6 0.7 0.6 0.8 0.3

0.0 0.1 –0.1 0.1 –0.1 –0.2

0.5 1.3 1.0 0.7 1.0 0.0

0.3 0.3 0.4 0.4 0.2 0.4

1.1 3.0 2.6 1.4 1.6 0.5

2.5 0.7 1.6 2.0 3.6 0.5

2.2 1.4 2.2 2.0 3.0 0.3

–0.1 0.1 0.2 0.1 0.0 –0.4

–0.4 0.4 0.2 0.3 0.3 0.3

0.0 0.4 0.7 0.5 0.3 0.4

0.2 2.2 1.9 1.2 0.8 0.8

2.1 0.2 0.8 0.4 2.2 0.7

1.2 0.7 1.3 1.0 1.1 0.4

Source: OECD Economic Outlook Database. Note GDP and private Investment exclude stock building.

potential output will be quickly translated into higher current output and employment” (Solow, this volume). If the hypothesis of high structural unemployment is valid, German industry should regard it as a disaster if foreign demand for German products were to increase substantially. If foreign demand is happily satisﬁed, why not domestic demand, is a bold question Robert Solow asks. It seems implausible that the German economy would not respond with higher output to higher demand, given that actual output is currently below potential. The “labor market rigidities as the root of European unemployment hypothesis” ﬁts the cross-country data well, but it fails to explain why unemployment rose substantially in Germany. “The claim that institutional change would resolve European unemployment problems did not take account of the longitudinal data. It rested entirely on cross-section comparisons of institutions and outcomes” (Richard Freeman, “Wanted: a new German Wirtschaftswunder”, this volume).

This book: a fresh view from international economists This book presents fresh, innovative analyses of the German economic policy debate from different angles. All papers are written by distinguished

6

R. Schettkat

international economists with the highest reputation. None has German nationality. Three authors are US citizens – Robert Solow (MIT and Russell Sage Foundation), Richard Freeman (Harvard University and NBER) and Adam Posen (Institute for International Economics, Washington); one author, Wendy Carlin, is Australian and teaches at the University College in London; and the other authors are from neighboring countries: Charles Wyplosz, France (Graduate Institute of International Studies, Geneva and CEPR); David Soskice, Britain (London School of Economics and Duke University); Paul de Grauwe, Belgium (University of Leuven); and Cláudia Costa Storti, Portugal (Banco de Portugal). We invited these prominent international economists to comment on the German economic policy debate because external observers are not partisan and have an unbiased view on the German debate. Many arguments have been repeated so many times in the German debate that they are not questioned any more. Often these arguments are regarded as facts even though there is sometimes a complete lack of evidence. Often it seems that careful empirical research is ignored if it does not pass the ﬁlter of prejudices which strong theories seem to produce. Just like anybody else, economists seem to be more attracted by “facts” which ﬁt their views than by contradicting results. But in the German case, the debate has become overly narrow, focusing on simple cost-reducing supply-side measures only and excluding a whole set of policy options which can fruitfully be used to improve economic prosperity. Many participants in the German economic policy debate have forwarded speciﬁc arguments so rigorously that it is hard to discard them. Furthermore, being in the middle of the mist, it is often more difﬁcult to see the major lines than with a clear view from above. Foreign economists not too deeply involved in the German debates, working at ﬁrst-class universities, and being very knowledgeable about the German economy, offer fresh, non-partisan views. Nobel Laureate Robert Solow (this volume) sees a lack of a serious macroeconomic policy debate in Germany. Although reforms to improve the functioning of labor markets (and also product markets) are desirable, Solow argues that the debate is far too narrow if it stops there. Ignoring macroeconomic factors is symptomatic of a misunderstanding of both the German situation and economic theory. Slow and inadequate economic growth and unemployment are speciﬁc macroeconomic problems. When even economics professors (e.g. Eekhoff in Berliner Gespräch Phoenix March 2005) declare that public budgets in Germany were expansionary in the past because they were in deﬁcit (actually they were highly pro-cyclical, see Wyplosz, this volume) and that public budgets are like budgets of a family where the father cannot reduce a deﬁcit by increased spending, it shows a deep misunderstanding of macroeconomics (see also Baumol 2005). The ignorance of macroeconomic insights is somewhat unique to the German economic policy debate, and it makes many foreign economists wonder why Germans are so dogmatic. It seems that “Deutsche Gründlichkeit” leads many to ignore fundamental insights of economics, producing unnecessary and counterproductive restrictions of thinking and potential policy instruments, as all authors in this volume seem

Introductory summary 7 to agree. “I think there are purely intellectual sources of the narrowness that seems to have characterized it” (Solow, this volume). However, theories which discard a whole set of policy options are very costly, reducing the wealth of Germans unnecessarily if they are wrong. The international community of economists does not ﬁt in the tight theoretical straightjacket of the German economic policy debate and uses much broader theoretical perspectives. Calling someone a Keynesian is not showing any disrespect among US economists. Keynesian economics is also not a matter of left and right political views as the former chief economic advisor to US President Bush, Gregory Mankiw (2006), argues. Keynesian economists, among the most respected and honored American economists, have been advisors to democratic as well as republican administrations. And this has a long tradition from Robert Solow advising John F. Kennedy, Joseph Stiglitz advising Bill Clinton, to Gregory Mankiw. In American politics, in the administration and in the FED, radical views on economic policy discarding macroeconomics as an important tool to improve the nation’s wealth never became dominant (Blinder 1998). Keynes probably helped capitalism to survive more than any free market proponent, stated Joseph Stiglitz (2006). We are convinced that the contributions in this volume will stimulate and improve the German economic policy debate. The analyses of leading international economists will help to get a clearer view on unnecessary economic policy constraints, and they may help to overcome self-imposed restrictions.

The analyses forwarded in this book Is monetary policy ineffective? Monetary policy option unused When the European Monetary System was created in the 1990s, Germany insisted on designing the European institution to guarantee a high degree of price stability. Germans feared that other EU countries would not follow the “culture of stability”. Charles Wyplosz (“Germany in the monetary union”, this volume) reminds us that the introduction of the euro brought a loss of independent monetary policy for the Bundesbank – then the de facto central bank of Europe – but not for the other European central banks, whose monetary policy was often explicitly or implicitly delegated to the Bundesbank. These countries (and their respective central banks) gained inﬂuence over the relevant monetary policy in Europe. For example, the Dutch central bank had previously followed all decisions of the Bundesbank in due course in order to stabilize the nominal exchange rate of the guilder to the DM. Thus, the Netherlands gave up a national monetary policy when pegging the guilder to the DM, which was actually the basis for the Dutch wage policy (Schettkat 2003). With respect to price stability, the euro performance – the ECB performance – is even superior to that of the DM – the Bundesbank. “The Eurosystem’s institutional set-up – considerable central bank independence and the priority given to price stability – is stronger in this respect than the Bundesbank’s and, for all

8

R. Schettkat

we know, institutions deliver when they are well designed” (Wyplosz, this volume). The euro kept the “culture of price stability”, yet it performed even better than the DM in this respect. The economic philosophy of the ECB seems to be one where monetary policy is regarded as neutral to output and employment. As Professor Otmar Issing, the chief economist of the Bundesbank and later of the ECB, described the economic philosophy of the ECB, monetary variables do not directly affect production or employment (for a description see, e.g., Issing 2000). All monetary policy does is to inﬂuence the price level. Improving economic growth and employment requires structural policies, e.g. deregulation of labor markets, which are commonly identiﬁed as the main cause for Europe’s and especially Germany’s economic problems. The best economic policy the ECB can pursue is, according to Otmar Issing, a policy of low and stable inﬂation. This way, expectations can be stabilized, which enhances investment because it reduces uncertainty. Thus, it is best for monetary policy to keep inﬂation stable at a low rate. Monetary policy may have output and employment effects in the short run, but it is neutral in the long run. Thus, in the ECB model, it does not make sense to pursue an expansionary economic policy path, which would result in rising prices alone but which would not affect the real economy. This is a one-sided view, already pursued by the Bundesbank, which was strongly criticized in an open letter from former chancellor Helmut Schmidt (1996) to the president of the Bundesbank. Does the ECB-policy model stand on solid ground? Do the predictions – mainly price but no quantity effects of monetary policy – ﬁt the facts? Paul de Grauwe and Cláudia Costa Storti (“Monetary policy and the real economy”, this volume) performed a meta-analysis of 83 studies on the effects of monetary policy and found that monetary policy affects economic activity in the short run but also in the long run. Those studies which do not ﬁnd long-run effects of monetary policy are based on the so-called structural VAR (vector autoregressive) models, i.e. they use theory to constrain parameters. These studies impose rather than test empirically that monetary policy is neutral to real economic activity in the long run. Methods which do not impose this constraint and “allow the data to speak” reject the long-run neutrality of money on economic activity (de Grauwe and Costa Sorti, this volume). Monetary policy seems to have real effects on output and hence employment in the short and in the long run. Probably most importantly, the EU and the US economies seem to show very similar responses to monetary policy. In other words, the de Grauwe and Costa Storti analysis suggests that the ECB could do exactly as the FED does to stimulate economic activity. Monetary policy can be used to improve the living conditions of Europeans. The conclusion that monetary policies should only be used to stabilize price levels and that it cannot be used for other policy purposes “have more to do with theoretical convictions about how the world should work, than with hard empirical evidence of how the world actually works” (de Grauwe and Costa Storti, this volume). The FED, many economists argue, can follow a more active monetary policy

Introductory summary 9 because the US labor market is much more ﬂexible than the German labor market or the EU labor markets in general. Labor market rigidities, so the argument goes, distinguish Europe, and especially Germany, from the United States, where ﬂexible labor markets allow the government and the FED to pursue counter-cyclical macroeconomic policies. Here, however, the conventional economic litany runs into a theoretical contradiction: in perfectly ﬂexible markets, monetary policy should not have any effect. For the nominal sphere to inﬂuence the real sphere, it needs “rigidities”. Only if a time lag between the monetary impulse and price effects exists, can monetary policy affect output. In other words, conventional theory should predict monetary policy to have stronger effects in Europe, where labor markets are alleged to be rigid, as compared with the United States, where ﬂexibility is regarded as high. In their meta-analysis, Paul de Grauwe and Cláudia Costa Storti (this volume), however, do not ﬁnd any evidence that the effects of monetary policy differ between Europe and the United States. In both regions, monetary policy affects output and employment. In the United States and in Europe, monetary policy affects economic activity. In other words, labor markets on both sides of the Atlantic seem to be similarly “rigid”. The common assertion that monetary policy is effective in the United States but ineffective in the allegedly more rigid Europe (Wendy Carlin and David Soskice, “Reforms, macroeconomic policy and economic performance in Germany”, this volume; Freeman, this volume) seems not to hold. A more practical issue is that the ECB has to design monetary policy for the entire Eurozone and not just for Germany, which creates contractionary monetary effects in low-inﬂation countries, because under the condition of identical nominal interest rates, real interest rates are higher in low-inﬂation countries. However, not only such “practical” matters but rather economic theory applied by the ECB prevents the bank from pursuing a less contractionary monetary policy. Maybe the ECB needed to build a good reputation in its ﬁrst years of existence, but it actually got more discretionary freedom than the Bundesbank ever had. If monetary policy affects the real economy and not only monetary variables, the sole focus of the ECB on price stability is mistaken and President Sarkozy’s proposals for the central banks’ support of European economic policy goals should be discussed seriously German ﬁscal policy is pro-cyclical The change in the cyclically adjusted primary budget balance (difference in the annual public budget balance adjusted for cyclical ﬂuctuations minus interest expenditures), which is a measure of the macroeconomic effects of ﬁscal policy (see the contributions of Solow and Wyplosz in this volume), was positive since 2002 (OECD 2006), i.e. German ﬁscal policy was pro-cyclical. Fiscal policy in Germany was reinforcing the output gap (the percentage point difference between actual and potential GDP; see Charles Wyplosz, Figures 4.4 and 4.5, this volume). No doubt, German ﬁscal policy has been contracting especially during the stagnation in the 2000s. That is, the efforts of the government to

10

R. Schettkat

reduce the deﬁcit through budget cuts had the reverse of the intended effects; they pushed Germany further away from fulﬁlling the so-called 3 percent Maastricht criteria. German ﬁscal policy was worsening the economic situation instead of improving it. Germany was living below rather than above its potential, as many commentators like to argue. The 3 percent limit of new public net debt is a constraint for Germany’s and other EU countries’ ﬁscal policy. Solow reminds us that the Maastricht criteria were introduced to enforce the necessary convergence of the EU economies before the introduction of the common currency in 1999. They are not a blueprint for optimal macroeconomic policy conduct. The European Stability and Growth Pact in which the 3 percent criterion has been made permanent is not derived from the old testament but rather the result of a speciﬁc economic theory which denies a stabilizing role for ﬁscal policy. In this theory, potential output and current output are always identical, and thus there is no role for stabilizing policy. But, as Robert Solow (this volume) argues, this theory is based on a priori assumptions rather than on careful observations of economic behavior. By any standard, German ﬁscal policy has been pro-cyclical, i.e. worsening the problems rather than helping to solve them, Solow concludes in line with the Wyplosz analysis. Charles Wyplosz reminds us that the European macroeconomic institutions are constraining economic policy severely. With a common currency, ﬁscal policy is the only tool left to adjust the macro-economy to country-speciﬁc problems. But the use of this tool has been restricted by the Stability and Growth Pact (the 3 percent criteria). “Thus demand has been subdued and did not even keep up with a slowly rising supply. Both monetary and ﬁscal policies have been insufﬁciently supportive” (Wyplosz, this volume). Alan Blinder, former vice chairman of the FED and professor at Princeton, puts it this way: So, there is much greater need for nation-speciﬁc discretionary ﬁscal policy in Europe than there is for state-speciﬁc discretionary ﬁscal policy in the United States. When it comes to dealing with country-speciﬁc shocks, the monetary union tied one hand behind policymakers’ backs, perhaps for good reasons. The Stability and Growth Pact then tied the other. Try riding a horse with both hands tied behind your back. (Blinder 2002: 393) Wage restraint: foreign versus domestic demand effects “Exportweltmeister” and taillight in growth performance seem to be contradicting trends at ﬁrst glance, but they may actually have common roots. To the extent that exports are driven by price competitiveness, achieved by declining relative unit-labor costs, this may lower innovative activity and may dampen domestic demand (Adam Posen, “Exportweltmeister – so what? better goals for German foreign economic policy”, this volume). A policy of wage restraint can have very different effects in large economies from those in small economies, as

Introductory summary 11 Wendy Carlin and David Soskice (this volume) emphasize. Wage restraint promotes foreign demand, but at the same time it reduces domestic demand. In small economies, foreign demand naturally has a big weight, but for larger economies domestic demand may be more important. That is, the net effect of wage restraint may be negative (see Figure 5.3 in Carlin and Soskice, this volume). Furthermore, German products are to a large extent shipped to other EU countries dampening demand for domestic products there and probably slowing growth. An economy may solve overcapacity problems through exports, but that is, as Robert Solow (this volume) writes, “not very neighborly”. Low domestic demand not keeping up with potential growth is quite unique for Germany. And low demand is related to the slow growth of wages, which are of course improving the price competitiveness of German products, but “lower wages is not a recipe for increased consumption” (Solow, this volume). Solow shows that the export dependence of the German economy increased in the period when labor markets were deregulated, so that inﬂexibility can hardly be blamed for the slack. The loss of well-being in periods with unused capacity, when unemployment is high, is substantial. Material output is below its potential level, and the unemployed are suffering not only from material losses but also from losses in wellbeing. Unemployment is the single best predictor of lower happiness (Layard 2005). This should not be the case if the new classical hypothesis of structural unemployment holds since being without a job should maximize the utility of the unemployed. That is actually why they are unemployed. Involuntary unemployment would not exist or would be very short-lived in new classical model. In reality, unemployment represents real cost and is not the outcome of optimizing behavior of the unemployed. “Money doesn’t bring happiness, but lack of money brings misery” (Krugman 2000). Adam Posen (this volume) goes beyond the short-run effects of export promotion through relative deﬂation and argues that German economic policy is favoring high exports in a “near mercantilist” fashion. He criticizes the exportled growth strategy based on relative deﬂation of German products, which can create positive effects in the short run but which will damage competitiveness in the longer run because it reduces the incentives to improve productivity. Improving price competitiveness is also the wrong answer to new upcoming low-cost competition from giants like China and India, Posen argues. The appearance of these new players in world markets requires more than ever an innovation strategy based on the novelty and quality of products and not on price competitiveness. Wage restraint and concession bargaining is as limited a strategy as nominal exchange rate adoptions are. The German economy needs restructuring, it must produce new high value-added products and services rather than trying to keep production in industries which are already moving down the value-added chain and which will face strong low-cost competition (Posen, this volume). Even the “sacred cow” of German economic policy, the “Mittelstand”, does not escape Posen’s critique. Any pampering of companies, so his argument goes,

12 R. Schettkat reduces the incentive to improve productivity and foster restructuring to more efﬁcient units. “Mittelstand-companies” drive down German productivity, which is the critical factor for wealth creation. “Arrested development” is the Posen term for German policies that protect inefﬁcient ﬁrms from world market competition but which will backﬁre in the long run. Components of growth Especially in the period since the beginning of this century, all components of domestic demand have contributed little or negatively to GDP growth in Germany (see Table 1.2). Private consumption in particular contributed substantially to GDP growth in other countries (see Carlin and Soskice, this volume for an analysis). Private investment was low in Germany, which was, as the business associations usually argue, due to low proﬁt rates, but low demand is for sure also an impediment. Solow reminds us that a long tradition in macroeconomic research has found that the margin of excess capacity is an important negative inﬂuence on business investment: “If there is an Okun gap, closing it would clearly provide a stimulus to private investment” (Solow, this volume). In other words, actual and potential output are not independent; the former may affect the latter. The only driving force for German growth recently has been exports. However, the expansionary effect of foreign demand was partly compensated by domestic demand components. Without exporting to the rest of the world, Germany’s overcapacity, the Okun gap, would have been even bigger. Strikingly, public investment growth did not contribute at all to the demand expansion and was even negative in the period 2000–2005. At the same time, it is obvious that the German public infrastructure of streets, universities, libraries, schools, etc. is deteriorating. The public capital stock is shrinking, which is, of course, the direct result of constrained public budgets. Another big promise of the tax cuts, the stimulation of private expenditures obviously failed. The government has had many opportunities for useful investment, but the curbing Table 1.3 OECD-estimates of structural unemployment rates (NAIRUs) Country

1990

1995

2000

2003

Germanya UK USA France Japan Netherlands Sweden

6.5 8.0 5.9 9.3 2.8 7.5 2.2

6.4 7.2 5.3 10.4 3 5.3 4.8

7.6 5.6 4.9 9.2 3.6 4.5 4.8

7.7 5.4 4.8 9.1 3.9 3.3 4.7

Source: OECD, 2005a. Note a Germany refers to 1991.

Introductory summary 13 4 3.5 3 2.5 2 1.5 1

Germany

France

United Kingdom

2000

1990

1980

0.5

Sweden

United States

Figure 1.1 Public investment as a share of GDP (constant prices) Germany, USA, UK, Sweden and France (%) (source: computations on OECD Economic Outlook database).

of public investment in Germany went further and even below the US level of the 1980s (see Figure 1.1) when the whole world wondered why the Reagan administration allowed for an deterioration of public infrastructure. There are many areas for proﬁtable public investment. One such type of “infrastructure” unanimously ranked most important is education. “The OECD Growth Project estimated that in the OECD area, the longterm effect on output of one additional year of education in the adult population generally falls between 3 and 6%” (OECD 2005b: 150). The contribution of human capital to labor productivity growth is, according to the OECD estimates, close to zero for Germany in the period 1991–2000. For other countries (here the reference period is 1990–1999), changes in human capital components contributed from about one-third of 1 percent in Sweden and the United States to 0.85 percent in the United Kingdom to annual labor productivity growth (OECD 2005b: 150). Investment in education seems to be very proﬁtable for individuals and society (Krueger and Lindahl 2001). Although the economic literature on the relation between expenditures on education and performance often argues against a strong positive correlation, it is striking that Germany raised the expenditures on educational institutions (primary, secondary and tertiary) substantially less than other OECD countries (OECD 2005b: 182). Entry rates into tertiary education are much lower in Germany than in other OECD countries (OECD 2005b: 249). As Manfred Schmidt (2003) comments on the basis of his international comparison of expenditures on education, it is quite surprising that a country whose economy is based on the high skills of their workforce is risking losing this advantage in international competitiveness by low investments and a strong deterioration of its relative skill position.

14

R. Schettkat

7.5 7 6.5

Public Private

6 5.5 5 4.5 4

United States

United Kingdom

Sweden

Netherlands

Japan

Germany

3

France

3.5

Figure 1.2 Public and private expenditures on education, 2002 (source: OECD 2005: 176, StatLink).

Reformed institutions Reformed labor market institutions make German labor markets more ﬂexible, restrain wages, lower tax rates and reduce public expenditures; “follow the ten commandments of the OECD Jobs Study” was the prescription by the orthodox economic doctors (Freeman, this volume), and politicians of all parties swallowed these bitter pills, even social democrats. “There is no other way; we have to do it” became commonly acceptable. This assertion is also based on the rumor that the foundations of the economic success of Blair’s Labor government in Britain have been laid by Lady Thatcher’s harsh reforms. Strong Thatcher-type medicine is effective, so the message goes, it just takes some time to show up in the economic statistics. Despite the fact that German institutions always allowed for a smooth adjustment of the economy (Schettkat 1992), bitter pills have been swallowed for quite some time in Germany. Union density was declining as fast as in the United Kingdom under Thatcherism and at a higher rate than in the United States. Wage inequality was rising substantially (e.g. Burkhauser and Rovba 2006). What is the impact of structural reform on unemployment? Based on studies by Steve Nickell et al. (2005) and Bassanini and Duval (2006), Wendy Carlin and David Soskice simulate the predicted impact of the institutional changes implemented by German governments on the unemployment rate. Both, the Nickell et al. (2005) and Bassani and Duval studies, found no effect for employment protection, signiﬁcant positive effects for unemployment beneﬁt variables, signiﬁcant negative effects for the coordination of wage bargaining, a signiﬁcant positive effect for the tax wedge and a signiﬁcantly positive effect for product market regulations and for the real interest rate. The

Introductory summary 15 Carlin and Soskice simulations show that the UK unemployment path is tracked quite well by the coefﬁcients estimated by Nickell et al. and Bassani and Duval. But for Germany, these simulations fail. “Although changing institutions predict a modest part of the rise in unemployment in the 1960s and 1970s, they predict nothing thereafter” (Carlin and Soskice, this volume). For the period 1982–2003, the institutional change in Germany should have resulted in falling unemployment according to the estimates of Nickell et al. and Bassani and Duval analyses. “Combining the impact of changes in policy and in the output gap variable, unemployment should have been 2.5 points lower in 2003 than in 1982, whereas it was 2.9 points higher” (Carlin and Soskice, this volume). In other words, given the combined effect of German institutions and the output gap, the unemployment rate in 2003 should have been 2.5 percentage points lower than in 1982, whereas it actually was 2.9 percentage points higher. These estimates explicitly based on the institutional setting of the economies contrasts with aggregate estimates of structural unemployment rates (NAIRUs) published by the OECD (2005a). The OECD estimates falling NAIRUs for the United Kingdom and the United States but a rising rate for Germany, which is estimated to be 7.7 percent in 2003. Is it plausible that institutional change since the mid-1990s raised structural unemployment in Germany by 1.3 percentage points? Reforms in Germany have been in the directions recommended by others (Carlin and Soskice, this volume), but they did not show the predicted effect; they did not reduce unemployment. The Carlin and Soskice simulations are another crack in the “structural rigidities story” of high and rising German unemployment (see also Baker et al. 2005). There are good empirical reasons for rejecting this convenient belief that the labor market by itself provides an adequate account of the sad story of European unemployment. At the crudest level, the timing is wrong. One of the two big increases in unemployment took place in the early 1980s, although there was no change in labor-market regulation to account for it. (Solow 2000) Marketization and fertility Richard Freeman ﬁrst examines the substantial difference between labor force participation in the United States and in Germany, which is concentrated among the young, the elderly and women. Although recent ﬁgures show a rise in the female employment-population rates for mothers with children below three – where the published numbers for Germany even reach US employmentpopulation rates – this is related to very short working hours. Female labor force participation, Freeman argues, is not only a source of supply but also of demand: if more hours are spent in market work, less hours are available for traditional household production, which will then be substituted by market products. This way, “marketization” shifts both the supply and the demand curve. Because

16 R. Schettkat productivity levels in many European countries approached that of the United States, the difference in per capita income between the United States and Europe is nowadays mainly caused by the difference in working hours per head of the population (Blanchard 2004). This in turn is roughly caused 50 percent by shorter working hours and the other 50 percent by lower participation rates (Freeman and Schettkat 2005). Freeman regards higher female labor participation as a major source for the long-run growth in demand of the German economy. Freeman shows graphs in which the traditionally negative relationship between female labor force participation and fertility rates have turned positive: nowadays countries with high female employment have higher birth rates than countries with low female participation in market work. Of course, the relationship could only turn positive because supporting systems have been established in those countries which achieve high female employment. To increase employment and working hours of German women and to allow them to reconcile jobs and family without permanently endangering careers or discouraging child-bearing requires moving family policy from the periphery of policy discourse to center stage. There are family policies that can contribute to the goal of increased employment, and that can do so without lowering the rate of fertility. (Freeman, this volume) To stimulate female labor supply and to raise birth rates, Freeman argues, the supporting system (especially day care) needs to be improved in Western Germany, but incentives also need to be adjusted. Reducing marginal tax rates is a key variable in the Freeman approach. Freeman also draws a parallel between Puerto Rico and East Germany. In both cases, an economy with low productivity levels has been integrated into a highly productive, high income economy requiring substantial transfers to the less productive regions (“the rich uncle syndrome”, Freeman, this volume). However, Puerto Rico is tiny compared to the United States, but transfers to East Germany are a substantial burden for the West German economy and have especially raised labor costs (mainly through the shift of costs to the social security system), which affects marginal “taxes” on second earners in Germany substantially.

Conclusions The following contributions written by distinguished international economists all argue that improvements on the supply side are a necessary condition for smooth, long-run growth. But as demand side policies can fail if they do not meet the right supply-side conditions, so can supply-side policies. One-sided policy debates and proposals are not useful. Both sides of the market need to ﬁt to achieve improvements in economic activity. Supply-side variables affect potential output, but the production potential is not always and not automatically

Introductory summary 17 fully used, as Robert Solow reminds us. To assume that the economy always operates at full capacity (full employment) is not adequate, it assumes the problems away. Macroeconomic policy – ﬁscal and monetary – can effectively improve demand and should be used to enhance welfare in the German economy and in Europe, as the authors of this volume convincingly demonstrate. From inside the German economic policy debate, many hypotheses gained the status of facts not because they have been rigorously proven but because they have been repeated so many times that they have become the common view. Therefore, the well-informed outside analyses presented in this volume can open the eyes to better distinguish between hypotheses and facts. For example, it became the common view that monetary policy cannot stimulate economic activity and that therefore the tight monetary policy of the ECB is not harming economic development in Europe. Yet the contributions in this volume clearly show that monetary policy is effective, that it affects the real economy and that it thus can raise economic activity and wealth. Overly tight monetary policy is reducing economic activity; overly ambitious inﬂation goals are overly costly. The ECB’s primary objective is price stability, but the self-chosen limit of 2 percent annual inﬂation may be too low to allow the European economy to ﬂourish. Especially in periods of stagnation and unused capacity, the risk of rising inﬂation as response to an economic stimulus is low. Fiscal policy also should have been more expansionary in Germany. All indicators show that during the years of economic stagnation, ﬁscal policy was worsening rather than improving the situation. German ﬁscal policy kept the economy in stagnation rather than helping to overcome it. With monetary policy left to the ECB, ﬁscal policy is the only effective policy to overcome Okun gaps. Economic policy should not dismiss this measure. The so-called Maastricht criteria were designed for the convergence of national inﬂation and interest rates of the European Monetary. “It was understood from the start that these rules could under other circumstances force governments into perverse actions, like contractionary policies in a time of recession” (Solow, this volume). Labor market rigidities in Germany were never a convincing explanation for rising and high German unemployment as Carlin and Soskice (this volume) show, and a NAIRU of about 8 percent seems to be implausible especially against the background of their simulations. Politics has focused too much on labor market reforms in the strong belief that the root of the German economic problem lies there. This also shifted the emphasis to improvements in Germany’s price competitiveness, which has been improved substantially as the huge net export surplus shows. However, the exceptionally low wage increases, well below productivity growth and well below wage trends in other countries (see OECD 2006), affected German domestic demand negatively. At the same time, however, the long-lasting improvements on the supply side have not been adequately developed. Germany’s competitive advantages are certainly the skilled labor force and the infrastructure, but these have been neglected over the last years. Germany successfully caught up to US productivity levels until the mid-1990s. In this period, Germany invested heavily both in production

18 R. Schettkat equipment and in public infrastructure and human resources. This is no longer the case: investment – private and public – declined in the ﬁrst years of this century and human resource investments are particularly low compared with other advanced countries. Ironically, the underdevelopment of supply-side variables lowered domestic demand directly and indirectly. Underinvestment in education, for example, reduces aggregate demand directly, and resulting unused capacity is hardly stimulating private investment. Richard Freeman identiﬁes another source of long-run employment growth: the marketization of household services much more advanced in other countries. The traditionally negative correlation between fertility and women’s employment has turned positive showing that child care supporting systems can help to combine parental duties and work. Overcoming traditional patterns of work will shift labor demand and supply to higher levels and will thus give a boost to employment and income. The distinguished international economists contributing to this volume unanimously argue that a more expansionary macroeconomic policy would have helped to overcome the long stagnation Germany was suffering from during the last years. Their contributions clearly show that the view of impotent macroeconomic policy is mistaken. It is mistaken with respect to ﬁscal policy and with respect to monetary policy. Their arguments are against “common ideas” as they have been developed in Europe and Germany over the last 20 years or so. Therefore, a rethinking of the effectiveness of macroeconomic policy is required. A broader perspective on economic policy which does not neglect effective policies is needed in the debate in order to overcome unnecessary policy constraints and to improve prosperity in Germany and Europe.

Notes 1 The concept of this volume resulted from discussions in a group of economists consisting of Dr Michael Dauderstädt, Dr Gustav Horn, Prof. Peter Kalmbach, Prof. Jürgen Kromphardt and Dr Jochem Langkau. 2 “Core inﬂation” is the harmonized index of consumer prices excluding food, energy, alcohol and tobacco (data from OECD 2006: 64). The deﬂator for private consumption is 1.4 percent in 2006 and GDP deﬂator is 0.5.

References Baker, D., Glyn, A., Howell, D.R. and Schmitt, J. (2005) “Labor market institutions and unemployment: a critical assessment of the cross-country evidence”, in Howell, D.R. (ed.) Fighting Unemployment. The Limits of Free Market Orthodoxy, New York: Oxford University Press: 72–118. Bassanini, A. and Duval, R. (2006) “Employment patterns in OECD countries: reassessing the role of policies and institutions”, OECD Working Paper Series, Working paper No 486. Baumol, W.J. (2005) “Errors in economics and their consequences”, Social Research, 72(1): 1–26.

Introductory summary 19 Blanchard, O. (2004) “The economic future of Europe”, Journal of Economic Perspectives, 18(4): 3–26. Blinder, A.S. (1998) Central Banking in Theory and Practice, Cambridge, MA: The MIT Press. Blinder, A.S. (2002) “Commentary: should the European Central Bank and the Federal Reserve be concerned about ﬁscal policy?” in The Federal Reserve Bank of Kansas City (ed.) Rethinking Stabilization Policy, Wyoming: Jackson Hole. Bundesagentur für Arbeit (2007) Arbeits- und Ausbildungsmarkt in Deutschland, Nürnberg: Juni. Burkhauser, R. and Rovba, L. (2006) “Income inequality in the 1990s: comparing the United States, Great Britain, and Germany”, DIW Discussion Paper 576, Berlin. Freeman, R.B. and Schettkat, R. (2005) “Marketization of household production and the US-EU gap in work”, Economic Policy, 20(41): 6–50. Friedman, M. (1968) “The role of monetary policy”, American Economic Review, 58(1): 1–17. Issing, O. (2000) “The European Monetary Union”, CESifo Forum (A Quarterly Journal on European Issues), Summer: 3–9. Krueger, A. and Lindahl, M. (2001) “Education for growth: why, and for whom?” Journal of Economic Literature, 39(4): 1101–1136. Krugman, P. (2000). “Purchasing happiness”, New York Times, 29 March. Layard, R. (2005) Happiness, New York: The Penguin Press. Mankiw, G. (2006) “The macroeconomist as scientist and engineer”, NBER Working Paper Series, Working Paper 12349. Marshall, A. (1920) Principles of Economics, London: Macmillan. Nickell, S., Nunziata, L. and Ochel, W. (2005) “Unemployment in the OECD since the 1960s. What do we know?” Economic Journal, 115(500): 1–27. OECD (2003) The Sources of Economic Growth in OECD Countries, Paris: OECD Publishing. OECD (2005a) Economic Policy Reforms. Going for Growth 2005, Paris: OECD Publishing. OECD (2005b) Education at a Glance. OECD Indicators 2005, Paris: OECD Publishing. OECD (2006) OECD Economic Outlook, Paris: OECD Publishing. Phelps, E. (1968) “Money-wage dynamics and labor-market equilibrium”, Journal of Political Economy, 76: 678–711. Schettkat, R. (1992) The Labor Market Dynamics of Economic Restructuring: The United States and Germany in Transition, New York: Praeger. Schettkat, R. (2003) “Are institutional rigidities at the root of European unemployment?” Cambridge Journal of Economics, 27: 771–787. Schmidt, H. (1996) “Offener Brief an Bundesbankpräsident Hans Tietmeyer. Die Bundesbank – kein Staat im Staate”, Die Zeit, Nr. 46. Schmidt, M. (2003) “Ausgaben für Bildung im internationalen Vergleich”, Politik und Zeitgeschichte, B-21–22: 6–11. Siebert, H. (2005) The German Economy: Beyond the Social Market, Princeton, NJ: Princeton University Press. Sinn, H.W. (2007) Can Germany Be Saved? The Malaise of the World’s First Welfare State, Cambridge, MA: MIT Press. Solow, R.M. (2000) “Unemployment in the United States and in Europe: a contrast and the reasons”, CESifo Working Paper 231, Munich. Stiglitz, J. (2006) “Chancen der Globalisierung”, Financial Times Deutschland, 23 September.

2

Broadening the discussion of macroeconomic policy Robert M. Solow

It is not my purpose in this short chapter to prescribe a macroeconomic policy for Germany. That is better left to those who know the German background far better than I do. All I want to do, as an outside observer, is to call attention to the extremely and unnecessarily narrow focus of the current discussion of macro-policy in expert circles in Germany. It is not too much to say, in my view, that there is almost no proper discussion of speciﬁcally macroeconomic problems and remedies. Instead, there is always talk of labor-market reform and, only very recently, occasionally product-market reform. These are no doubt important issues, and reforms are surely desirable. But stopping with them is exactly what I mean by a narrow focus and an evasion of macroeconomic factors. I think that this limitation is symptomatic of a misunderstanding of both the German situation and macroeconomic theory. The German economy has many problems. (Every advanced industrial economy has many problems.) Two of them are speciﬁcally macroeconomic: slow, or at least inadequate, economic growth, and high and persistent unemployment. It is important to keep in mind that unemployment stands here for more than the unavailability of appropriate jobs for those who are seeking work. We use the unemployment rate as an indicator of unused productive capacity in the economy as a whole. That is because unemployment is easier to measure than idle machinery, ofﬁces, shops and so on. Sometimes the difference between unused labor and unused capital can be important, but that does not seem to be the case in Germany today. These two big problems have been well documented, so I can describe them very brieﬂy. On the growth side: during the 30 years after 1950, say, as the German economy was rebuilt and modernized, productivity (output per hour worked) grew very fast and almost caught up with the United States. But then the growth of productivity slowed and the gap with the United States even began to widen again. German labor productivity grew at an annual rate of 5.3 percent in the 1960s, 3.7 percent in the 1970s, 2.6 percent in the 1980s, 2.1 percent in the 1990s and only 1.2 percent from 2000 to 2003. Some of that slowing down was probably inevitable as the catching-up process proceeded, but not all of it. The weakening growth of productivity is naturally reﬂected in weakening growth of real national income per person. Germany is still a rich economy, but

Broadening the macroeconomic policy 21 there is a pervasive feeling that it could and should be growing faster. (Much the same could be said of other large European economies, like France.) The other big problem is unemployment. (Here, I will refer to West Germany, merely to maintain comparability over time. The reuniﬁcation, as carried out, was a major shock to the German economy, and its effects remain. I leave them aside, despite their importance, so that I can focus on the central issue of macroeconomic policy.) In the 1950s and 1960s, unemployment was very low, never rising as high as 2 percent. But there was a jump to 4 percent in the 1970s, and even in the next boom, the unemployment rate stayed well above 3 percent. Another enormous jump in the 1980s brought the unemployment rate to 8 percent. There have been ﬂuctuations since then, ranging between 6 and 10 percent, but nowadays something slightly above 8 percent seems to be the norm. As I write, the standard forecasts do not suggest any noticeable reduction in unemployment in the near future. I realize, though I do not intend to discuss it further, that the annual hours of work (for a full-time worker) have been falling in Germany and are now much lower than in the United States. This means that output and income per person will grow more slowly than output per hour worked. Some observers think that all or most of this difference reﬂects a “cultural” preference in Germany and continental Europe for enjoying the beneﬁts of increasing productivity in the form of more and better leisure rather than more and better goods and services; Americans presumably make the opposite choice. Other observers believe that the higher marginal tax rate on labor income in Germany is enough to account for all or most of the difference in hours worked, without any appeal to cultural differences; Europeans work less because the “price of leisure” is much lower for them than for Americans. The answer to this question is not especially relevant to the issues I want to discuss, except to the extent that policies to encourage shorter hours are a response to high unemployment, an attempt to share an inadequate number of jobs among a larger number of people. This was pretty clearly the motivation for the French 35-hour Law. In that case, an increase in the demand for labor might partially reverse the reduction in hours worked. A long and respectable tradition in macroeconomics distinguishes carefully between these two problems, growth and unemployment. It begins by deﬁning the “potential output” of an economy as the real GDP that it can produce with the labor (and skills), capital and technology available to it, but without putting so much sustained pressure on the factors of production that serious inﬂationary pressure is created. Potential output increases as the labor force increases, and its skills are improved by education and training as technological progress occurs and as investment adds to the stock of capital and enables it to absorb new technology. The increase of potential output is what we mean by economic growth. It is a medium-to-long-run process, mostly concerned with the “supply side” of the economy, the forces that make for productive capacity. There is already something to be learned from this deﬁnition. It has been found that the main sources of sustained growth per person are continuing technological progress and the continuing improvements in human skills that

22 R.M. Solow usually go along with it. On the other hand, any one-time improvement in economic efﬁciency, the sort of gain that comes from smoother factor mobility, better ﬁnancial intermediation, the removal of a monopoly-enhancing regulation, and indeed tougher competition generally, generates a one-time increase in the level of potential output. If the economy is able to take advantage of this increase, it will experience a period of temporarily faster growth of real GDP, but not a permanent increase in the long-run growth rate. Of course even onetime – but sustained – increases in potential output are very valuable, but the distinction is worth keeping straight, to avoid confusion about what some particular policy proposal can hope to deliver. If current real GDP were always equal to potential GDP, we would not have sudden unexpected ﬂuctuations in output, because potential GDP grows rather smoothly. But current output is not always equal to potential output. In a recession, output falls below potential. (The gap between them is sometimes called an “Okun gap” after the economist who ﬁrst tried seriously to measure it.) An overheated economy can drive current output above potential, with inﬂationary consequences, but that is not the problem right now. The visible symptoms of such a gap are excess unemployment and idle or partially idle capital. Potential output is not observed; it has to be estimated. Part of the difﬁculty about estimating it is to know how much “excess” unemployment there is, and why it occurs. In a commonsense way, we understand that aggregate output is sometimes noticeably less than potential output. When that happens, it is usually because the economy’s demand for goods and services falls short of its capacity to produce. If there were more willing buyers in shops and in factory showrooms, sellers would hire some of the unemployed labor and use some of the idle capital to meet their demand. The incomes earned in this extra production might add further to total demand. It is possible that willing buyers are there, but sellers do not want to satisfy them because costs are too high. I do not think that all of Germany’s high unemployment can be explained that way – where are all those unsatisﬁed buyers? – but that is a subject for discussion. Generally, when output falls short of potential, it is because aggregate demand is insufﬁcient to buy back potential output. I have treated actual and potential output as if they are independent of each other, one governed by demand-side factors and the other by supply-side factors. That is not literally true, of course. Strong demand will usually include, and stimulate, business investment, and this in turn will certainly generate some extra growth of potential. Conversely, fast growth of potential through technological progress will usually stimulate investment; that constitutes an increment to demand, supplemented by the spending of additional incomes earned during this process. These interactions can also work in reverse in periods of recession. More generally, the history of industrial capitalism tells us that very large, very persistent Okun gaps are rare. When they do happen, as in the Great Depression of the 1930s or perhaps the Japanese slump of the 1990s, they are memorable events. The rarity of great depressions is unlikely to be simple good luck. It may come about because deliberate policy eventually stabilizes the economy or because of the natural stability of a market economy, or for both

Broadening the macroeconomic policy 23 reasons. Nevertheless, it is clear that periods of recession and stagnation do occur, when actual output is noticeably less than potential for more than just a few months. The loss of well-being in such periods can be substantial. Even if one were conﬁdent that they would not become enormous or last forever, there is every reason to ask what public policy can do to ﬁll those Okun gaps. Perhaps I should mention here that there is an important academic school of thought that in effect denies the distinction between potential and actual output. It prefers to think of observed economic ﬂuctuations as the “optimal” forwardlooking response of the economy to unforeseen and unforeseeable shocks. Unless the government has better information about current and future events than the private sector has, any attempt at stabilization through public policy can only make things worse. After all, according to the theory, things are already as good as they can be. And if the government does have superior information, its best bet is simply to inform the private sector. I do not ﬁnd this approach to macroeconomics at all convincing, for many reasons. In any case, this does not seem to be the kind of thinking that underlies German policy discussion today. If it is, it should be actively debated by economists and others. So I return to the earlier train of thought. The ﬁrst basic question to ask about Germany today is the size of the gap between current output and potential output. If it is at all substantial, if there is indeed a shortage of aggregate demand, then it takes a great leap of faith to believe that merely lowering real wages by deregulating labor markets will by itself (or even accompanied by some freeing of product markets) always generate the demand necessary to close the gap and utilize fully the current productive capacity of the German economy. Lower wages is not a recipe for increased consumption. Investment is a trickier question. Lower costs and higher proﬁtability might indeed provide some incentive for business ﬁrms to add new capacity. But the existence of excess capacity – and that is what an Okun gap means – must work in the opposite direction, weakening the responsiveness of investment. A broader-based macroeconomic policy would face this issue directly. It is worth remembering that an investment-based recovery would also contribute to solving the German slow-growth problem. A more modern stock of capital would be a more productive stock of capital. So, a faster rate of capital investment would, at least for a while, allow faster growth from the supply side, even while contributing to it from the demand side. Even the one-time efﬁciency gains discussed earlier are easier to achieve in a buoyant economy. I now want to argue that there is good reason to believe that a substantial part of German unemployment reﬂects an Okun gap and would respond to a policy of expanding aggregate demand. It seems to me to be an intellectual failure that this side of the problem gets little or no attention in the policy debate. There may even be an element of paradox: one reason why the project of labor-market reform makes so little headway is that it is seen as putting downward pressure on real wages without offering any realistic compensation in the form of higher employment. Reasonable reform would stand a better political chance in an expanding economy.

24 R.M. Solow The German economy is able to reach its current level only by virtue of its export success. Between 2001 and 2004, according to the European Commission, domestic demand in Germany fell by more than 1 percent. Among the EU12, only Portugal shared that experience; all the other countries had rising domestic demand. In contrast, Germany had the largest percentage increase in exports among the EU-12; Portugal was second, and only Italy had falling exports. There would seem to be at least two suggestions implicit in this pattern. First, it suggests, in a general way, that internal demand in Germany has been suspiciously weak, surely not keeping up with the growth of potential output. Second, a nation’s export success is not usually a signal of excessively high wages. The two can be made compatible, of course, by very high productivity in the export sectors of the economy and that is undoubtedly part of the German picture. Nevertheless, it is pretty clear that the conventional account of the German problem deserves more scrutiny than it gets. Along the same lines, it is interesting that the export dependence of the German economy was much higher in the ten years from 1995 to 2004 than it had been in the preceding decade. Between 1985 and 1994, when real GDP grew at an average annual rate of 2.8 percent and unemployment averaged 6.1 percent, the contribution of domestic demand was 2.8 percent while increased exports added 0.2 percent to the overall growth rate. Between 1995 and 2004, this pattern changed: real GDP grew by only 1.3 percent annually and unemployment averaged 8.8 percent, with internal demand accounting for 0.8 percent and the foreign balance for 0.5 percent. One expects changes of this amount in a dynamic economy. The interesting point is that this shift occurred while the degree of regulation of the German labor market was diminishing, in most although not all respects, both absolutely and probably relative to the rest of Europe. (“Labor-market rigidity” is a hard thing to measure, but that is what makeshift indicators show.) So, the failure of internal demand cannot plausibly be blamed on the inﬂexibility of the labor market. It is true that an increasing part of German exports consists of re-exports rather than original (and more labor-intensive) manufactures. That, too, is expectable in a world in which low-wage, low-cost manufacturing in Asia and elsewhere increasingly appears on the world market. In a well-functioning economy, labor displaced in manufacturing would shift to other sectors, perhaps especially services. That process would go more smoothly if internal demand were stronger. Exports could continue to carry the burden, but that is neither good for Germany nor is it neighborly. In fact, the recent behavior of wages in Germany has been at worst moderate. In the decade from 1995 to 2004, employee compensation per hour (wages plus beneﬁts) rose a little faster than productivity. As a result, over those ten years, labor cost per unit of output rose by about 6 percent. During the same period, prices rose by almost 14 percent (an inﬂation rate of less than 1.4 percent per year). Excessive wage increases were not enough to create noticeable inﬂationary pressure. More important, even while prices were rising so gently – visibly

Broadening the macroeconomic policy 25 less than in the European Monetary Union as a whole – the margin of proﬁtability above labor cost was getting wider. To attribute continued high unemployment exclusively to the pressure of labor costs, as so much of the German conventional wisdom seems to do, strikes me as a considerable stretch beyond the evidence. It is much more plausible that there is indeed an Okun gap in Germany today, that more than 8 percent is not “the natural” rate of unemployment and that, although the rate of investment has been very low, enough capacity exists to support a higher level of production. I could not guess at the size of the gap. The standard estimate is that one percentage point of unemployment corresponds (conservatively) to 1.5–2 percent of real GDP, but I do not know if that applies well to Germany. Could one imagine a two-point reduction in average unemployment? Here is a thought experiment: imagine that the rest of the world’s demand for German goods was spontaneously to increase by 5 percent, spread over various goods and services that are currently exported. Would German industry regard this as a disaster? Or would German industry welcome it and respond by some normal combination of a small price increase (small out of respect for the competition) and an increase of employment and output? The evidence of the past decade or more suggests the second answer. Such a stimulus to demand might even induce some badly needed investment. (Some of that investment might be located outside of Germany; that is yet another problem, but not directly relevant to my point, although it is of course related to the level of wages in Germany.) If additional external demand would be welcome, why not additional internal demand? Indeed, a natural question to ask is, Why has there not been more internal demand in Germany? I suspect that the answer might be that monetary and ﬁscal policies have been excessively contractionary during the past decade. For much of that time, of course, since the advent of the EMU, monetary policy has been in the hands of the European Central Bank although the Bundesbank had considerable discretion in the years before EMU. For that reason, I will not say much about monetary policy, except to remind that the ECB can only make one monetary policy for all of EMU although macroeconomic conditions in the various countries may differ considerably. Since there are national differences in the rate of inﬂation, real interest rates can also be different even if nominal rates are similar. In the late 1980s and early 1990s, real (short-term) interest rates in Germany were lower than the average in the EMU countries; in the following decade, they were, if anything, a little higher in Germany, despite the fact that the German economy was weaker in the second period than in the ﬁrst, compared with other European countries. (Was there a causal connection between the real rate of interest and the strength of the economy? It would take a much deeper investigation to provide an answer.) To my mind, neither the Bundesbank nor the ECB has deserved any prizes for the conduct of an active monetary policy sensitive to the needs of the economy. For anyone interested in looking ahead, ﬁscal policy is a more relevant topic

26 R.M. Solow for discussion. There Germany has rather more discretion. (I will come to the Maastricht criteria and the Stability and Growth Pact (SGP) in a moment.) The ﬁrst question is whether ﬁscal policy has contributed to the recent macroeconomic stagnation (or, to use an American cliche, whether ﬁscal policy has been part of the problem or part of the solution). The cyclically adjusted (or standardized) budget balance is a reasonable, if imperfect, measure of the force exerted by ﬁscal policy on aggregate demand. (Some economists prefer the adjusted primary surplus or deﬁcit.) Measurement problems aside, a minimal requirement for ﬁscal policy to be a stabilizing inﬂuence is that the adjusted balance (relative to GDP, say) should fall when the Okun gap rises and vice versa. In simpler language, ﬁscal policy should become more expansionary or less contractionary when the economy weakens. One could ask for more, and once upon a time, one did. But German ﬁscal policy does not even pass the weaker test. During the past decade, it seems to have been pro-cyclical more often than it has been counter-cyclical. And the record has not been getting better. There are, no doubt, several reasons for this dismal performance. I am not well enough informed to make any judgment about their relative importance; that requires an intimate knowledge of the German intellectual and political scene. But I will make some comments anyway, because my concern is not so much with policy decisions themselves as with the narrowness of the public debate. The Maastricht criteria and the SGP cannot be taken as a general statement of the “right” way to conduct macroeconomic policy. They were designed for a speciﬁc occasion: to help in establishing the minimal convergence of national inﬂation and interest rates that was necessary for the creation of the European Monetary Union and the single currency. Some countries were clearly engaged in chronic ﬁscal excess; in many cases, their governments must have welcomed the compulsion as providing a political justiﬁcation for what they already knew to be a sensible economic policy. It was understood from the start that these rules could under other circumstances force governments into perverse actions, like contractionary policies in a time of recession. Those circumstances arrived in due course. When they did, the rules with respect to deﬁcits or debt were violated, either explicitly or implicitly through “creative accounting”. It is not for me to judge whether such rules – the same ones or modiﬁed versions – are still needed to restrain governments from acts of genuine ﬁscal irresponsibility. (In any case, I am skeptical whether they can be made to work sensibly.) It is possible, however, that the mere presence of the SGP inhibited serious public debate about ﬁscal policy. It is one thing to disregard an obligation, but perhaps harder to argue explicitly that it is the right thing to disregard an obligation. This could explain the attractiveness of focusing exclusively on labor-market rigidities. In the German case, though not elsewhere, an important special factor was the mismanagement of the economic aspects of the reuniﬁcation. The need for large transfers from West to East put an unusual strain on ﬁscal policy. The cir-

Broadening the macroeconomic policy 27 cumstances were also unusual, and standard arguments about ﬁscal policy needed to take account of them. That is a topic by itself, and it is not one about which I have a strong opinion. But, again, this factor could be one explanation for the inability of German discussion to focus on the possibility of an Okun gap and the desirability of compensatory policy. In addition, the costs of reuniﬁcation have left Germany with a large public debt to manage. That is something of an excuse, but it is also worth mentioning that continued macroeconomic stagnation does not make the debt easier to manage. These are accidental, almost exogenous, constraints on the content of debate about macroeconomic policy in Germany. I think there are also purely intellectual sources of the narrowness that seems to have characterized it. (What I have described as “purely intellectual” sources may in turn have roots in the twentieth-century history of Germany, but that is well beyond my competence.) I have already mentioned the existence – mainly in the United States but also elsewhere – of an important school of academic macroeconomic thought for which the whole concept of “inadequate internal demand” makes no sense. That is not the result of any careful observation of economic life, but rather because it is ruled out by deﬁnition: what is individually optimal is automatically socially optimal because there is, in effect, only one individual. I do not think that this rather artiﬁcial picture of a modern economy is widely accepted among German economists. Instead, it is an older habit of thought that seems to lead many German economists to be blind to genuinely macroeconomic pathology. So far as my limited acquaintance goes, many of the best and most signiﬁcant German economists seem to believe that control of inﬂation is not only necessary for a modern industrial economy to achieve a favorable macroeconomic outcome – a negligible Okun gap – but also sufﬁcient. In other words, they hold that deviations of actual output from potential output, even fairly large ones, are quickly and automatically self-correcting if only inﬂation is kept low and steady, and wages are ﬂexible. Inﬂation is a purely monetary phenomenon, and unemployment results only from excessively high real wages. So there is no stabilization function for ﬁscal policy to perform. That is why there is nothing to talk about except reform of the labor market. I have been arguing, on the contrary, that it is empirically unconvincing to attribute the whole rise and persistence of unemployment, from less than 2 percent in the 1960s to more than 8 percent currently, to excessive gains in real wages and worsened labor-market rigidity. The magnitudes are out of proportion; the timing is perverse and the strength of the export sector ﬁts very badly with the general hypothesis. In addition, cross-country studies linking unemployment with job protection and similar regulations have generally not been able to document an adequately strong effect in European countries. Perhaps it bears repeating that I have not tried to make a general case against reform of the labor market. That would have to be argued on different grounds. For example if, as I have read (CES-ifo Bulletin, January 2006), the marginal tax rate for a single person with average income was 68 percent in Germany, compared with 38 percent in the United States and 49 percent in the United

28 R.M. Solow Kingdom, normal wage incentives are likely to be blunted. Perhaps real wages are too high and some of the excess unemployment is “classical”. But it seems to me to be economically unconvincing (and politically counterproductive) to claim that nearly all of it is. I think it far more plausible that there is a layer of unemployment originating in weak internal demand induced by unnecessarily contractionary ﬁscal (and monetary) policy. Arthur Okun himself used to say that the belief that all unemployment is the result of high wages is like the belief that the hole in a ﬂat tire must be on the bottom, because that is where the tire is ﬂat. It is sobering that in this judgment, I seem to disagree with some of the eminent German economists I most respect. It should be sobering, on the other side, that the central research issue – the size of the Okun gap – gets little or no attention by the research community. Some years ago, I found a description of the method then used by the Bundesbank to estimate the level of potential output in Germany. It seemed to me to be defective, in the serious sense that it ruled out by deﬁnition any possibility of a persistent Okun gap. That should not be the end of the story. Finally, how does this view interact with Germany’s other macroeconomic problem of slow growth? A policy aimed at faster growth of potential output would focus on investment and innovation. (Labor-market ﬂexibility would enter here too, even if only temporarily, though it would probably become more important after the margin of excess unemployment had been absorbed.) A long tradition in macroeconomics has found that the margin of excess capacity is an important (negative) inﬂuence on business investment. If there is an Okun gap, closing it would clearly provide a stimulus to private investment. (Public investment has also been declining. If there is a need for improved infrastructure, ﬁscal policy should take account of it.) Innovation, including the absorption of new technology developed elsewhere, is a more complicated issue about which less is known. I am not aware that economic research has had anything to say about whether innovation ﬂourishes more in a high-utilization or a low-utilization economy. Many studies by the McKinsey Global Institute have found that exposure to competition from bestpractice ﬁrms around the world is an important determinant of an industry’s productivity performance. So, protective and restrictive regulation in product markets may inhibit growth. Reform of product markets probably deserves as much emphasis as reform of labor markets in the growth context and might also be useful when it comes to reducing unemployment. Clear thinking requires a reminder: the reward from more efﬁcient markets is higher potential output. There is no guarantee that an increase in potential output will be quickly translated into higher current output and employment. That may or may not happen. What Richard Musgrave called the “stabilization branch” of the government may have to help. That is unlikely to happen in the absence of serious everyday discussion of macroeconomic policy.

3

Monetary policy and the real economy Paul De Grauwe and Cláudia Costa Storti

Introduction How effective is monetary policy in inﬂuencing output and employment? How long do the effects of monetary policy last? These questions have been hotly debated. In the 1970s and 1980s, they led to a major schism in the economics profession between Monetarists and Keynesians. This schism seems to have been resolved now, and a mainstream view can be said to have emerged. The mainstream view today has been inﬂuenced by several theoretical developments. First, there is the real business cycle theory, which has introduced the idea that macroeconomic models should be based on sound micro-foundations in which individual agents continuously maximize their utilities in a dynamic framework and in which these agents understand and use the full complexity of the underlying model in forecasting the future (rational expectations). Second, building on this methodological innovation, macroeconomists have introduced price and wage rigidities into their models. This has led to the so-called NeoKeynesian models in which representative agents optimize their utilities and have rational expectations but face some constraints in that they cannot adjust prices and wages instantaneously (see e.g. Clarida et al. 1999; Christiano et al. 2001). This last feature provides the basis for monetary policy to affect output and employment. These theoretical developments have led to the consensus view, ﬁrst, that monetary policy has signiﬁcant short-term effects on output and employment; second, that these effects are temporary. This consensus can be represented by a hump-shaped curve showing the time proﬁle of the effect of an unexpected decline in the short-term interest rate on output. This curve is typically obtained from an impulse response analysis of the effect of an unanticipated decline in the short-term interest rate in an empirical version of Neo-Keynesian macroeconomic models (see e.g. Smets and Wouters 2003; Walsh 2003). Much of the discussion, today, among macroeconomists has shifted to the issue of whether this empirical regularity can be exploited by the monetary authorities and whether rules should govern the conduct of monetary policy. In this chapter, we provide additional evidence of the effects of monetary policies on output and on the timing of these effects. In order to do so, we will use

P. De Grauwe and C.C. Storti

Output

30

Time

Figure 3.1 Stylized impulse response function of output following an expansionary monetary policy.

meta-analysis as our methodology. We will focus on both the output and price effects of monetary policies, mainly because the issue of how effective monetary policy is in inﬂuencing output cannot be dissociated from the price effects of these policies.

A meta-analysis of the effects of monetary policy on output and prices The econometric analysis of the effect of monetary policies has changed considerably during the 1990s mainly as a result of the advance of econometric techniques and, in particular, as a result of the increasing use of VAR and structural VAR (SVAR) techniques. This has led to a proliferation of econometric evaluations of the effectiveness of monetary policies in many countries. In order to analyse the effectiveness of monetary policies in changing output, we will use a ‘meta-analysis’. This technique is frequently used in medical sciences and has sporadically been used in economics (see e.g. Knell and Stix 2003; Nijkamp and Poot 2004; Rose 2004).1 The objective of this analysis is ﬁrst to statistically analyse the estimated effects of monetary policy shocks on output and prices and second to identify the factors that can explain the differences in these estimated effects. The way to proceed is to ﬁrst collect data on the parameters that measure the effect of monetary policy on output and prices and that have been estimated in econometric studies. We will distinguish between the short-term and the longterm effects on output and price levels. The parameters collected from these

Monetary policy and real economy 31 studies will then be used as the dependent variable in an econometric analysis that aims at explaining the variation in these parameters. The data The source of the data we used is the empirical studies on the effects of monetary policies. We restricted the empirical studies to those published after 1990. The main reason is that during the 1990s, the new econometric technology using VARs came into use in studies evaluating monetary policies. Since this has become the new state-of-the-art econometric technology, we decided to restrict the analysis to a period in which this technology was introduced. We used a search of Econlit and also searched in well-known discussion paper series (NBER, CEPR, and CESifo) and the discussion paper series of central banks. We obtained 83 studies that report numbers on the effect of monetary policy. There are of course many more papers that analyse the transmission of monetary policies, but many of these papers provide no or incomplete quantitative evidence of the effects of monetary policy, or report results that cannot be made comparable to other results. We were interested in four different parameters measuring the effect of monetary policy. These are: 1 2 3 4

the short-term effect on output; the long-term effect on output; the short-term effect on the price level; the long-term effect on the price level.

We decided that the effects after one year measure the short run, while the effects obtained after ﬁve years measure the long run. In fact, all the long-run coefﬁcients reported in the studies relate to periods from ﬁve to seven years. We would have liked to use a longer time span, but no coefﬁcients measuring the effects after seven years were found. It is clear that this procedure introduces some arbitrariness in what we call short-run and long-run effects. In particular, one could argue that ﬁve to seven years is not really the long run. Yet, ﬁnding output effects of monetary policies lasting ﬁve to seven years comes close to what one could label the long run. The way the empirical results are reported is far from harmonized. The VAR and SVAR studies report impulse response functions that measure the impact of a monetary policy shock (typically a short-term unanticipated interest rate increase) on output and prices. We harmonized these numbers so that each number measures the effect of a 1 percent increase of the interest rate on output and the price level at the respective horizons.2 There are very few studies that use the money stock as the policy variable. Almost no VAR or SVAR studies use the money stock. As a result, we restricted the analysis to those studies that use the interest rate as the policy variable. There are also a number of studies using structural econometric models.

32

P. De Grauwe and C.C. Storti

These studies typically report the effect of a monetary policy shock on output (prices) as the difference between the simulated output (price) level obtained with and without the policy shock.3 We used these numbers and applied the same harmonization so that these parameters measure the effect of a shock in the interest rate (money stock) of 1 per cent. Many of the 83 studies selected report results for more than one country. As a result, we obtained 278 parameters measuring the short-term and long-term output effects of monetary policy shocks. For the effects of monetary policy on the price level, we only obtained 185 parameters because a number of studies focus only on the output effects of monetary policy. Some descriptive statistics Before engaging in the econometric analysis, it is useful to present some descriptive statistics of the different parameters measuring the effects of monetary policies. We do this in the form of histograms. We ﬁrst concentrate on the estimated output effects. In Figures 3.2 and 3.3, we show the histograms of the short-term and long-term effects of an interest rate increase of 1 percent obtained from our sample of econometric studies. We eliminated some outliers, i.e. in the case of the short-term effects, all the coefﬁcients lower than –1 and higher than +1, and in the case of the long-term effects, all the coefﬁcients lower than –1. However, for the sake of completeness, we present the full sample in Appendix 3.1. We focus ﬁrst on the short-term output effects in Figure 3.2 (effect after one year). We ﬁnd that the mean coefﬁcient is –0.23, i.e. after one year, a 1 percent (unanticipated) increase in the short-term interest rate leads to a decline in output of 0.23 percent. We also observe, however, that there is a large variance 30 Series: OUTPUTSTOUT Sample 1,278 Observations 253

25

0.232073 Mean 0.218000 Median 0.846154 Maximum 0.992908 Minimum 0.342539 Standard deviation 0.586143 Skewness 3.914012 Kurtosis

20 15 10 5

Jarque–Bera Probability

23.29361 0.000009

0.75

0.50

0.25

0.00

0.25

0.50

0.75

1.00

0

Figure 3.2 Frequency distribution of the estimated short-term output effect of a 1% increase in the short-term interest rate.

Monetary policy and real economy 33 of the estimated coefﬁcients. One of the purposes of the meta-analysis will be to identify the factors that explain this large variance. The long-term output effects shown in Figure 3.3 lead to similar observations. The mean long-term coefﬁcient is –0.16 which is lower than the shortterm coefﬁcient. This ﬁnding is in accordance with the hump-shaped time proﬁle of the effects of monetary policy shown in Figure 3.1, which suggests that the short-term effect of a monetary policy shock on output is larger than its longterm effect. We also ﬁnd that the variance in these estimated coefﬁcients is rather high. We perform a similar descriptive analysis of the price effects of monetary policy shocks. We show these in Figures 3.4 and 3.5. 60

Series: OUTPUTLTOUT Sample 1,278 Observations 221

50 40 30 20

Mean Median Maximum Minimum Standard deviation Skewness Kurtosis

0.157541 0.040000 0.811000 0.981432 0.300833 0.695962 3.700205

Jarque–Bera Probability

22.35543 0.000014

10

0.75

0.50

0.25

0.00

0.25

0.50

0.75

1.00

0

Figure 3.3 Frequency distribution of the estimated long-term output effect of a 1% increase in the short-term interest rate. 100

Series: SHORT Sample 1,185 Observations 183

80

Mean Median Maximum Minimum Standard deviation Skewness Kurtosis

60 40 20 0

Jarque–Bera Probability 3

2

1

0

1

2

0.043338 0.040000 3.232500 2.800000 0.425706 1.145268 30.58872 5843.677 0.000000

3

Figure 3.4 Frequency distribution of the estimated short-term price effect of a 1% increase in the short-term interest rate.

34

P. De Grauwe and C.C. Storti

60

Series: LONG Sample 1,185 Observations 177

50 40

Mean Median Maximum Minimum Standard deviaton Skewness Kurtosis

30 20

0.300804 0.110000 3.279000 6.000000 0.932080 2.530144 17.08711

10 Jarque–Bera Probability

0

1652.393 0.000000

5.00 3.752.501.25 0.00 1.25 2.50

Figure 3.5 Frequency distribution of the estimated long-term price effect of a 1% increase in the short-term interest rate.

From Figure 3.4, we observe that the mean coefﬁcient measuring the shortterm effect of an increase in the short-term interest rate is close to zero. The coefﬁcient measuring the long-run effect (Figure 3.5) is relatively large (in absolute value). Thus, the econometric studies conﬁrm that there is price stickiness. In the short run (one year), monetary policy shocks do not affect prices. These effects appear only in the long run. Note again that the variance around the mean coefﬁcients is very high. Econometric analysis: output effects In this section, we specify an econometric equation explaining the different parameters described in the previous section. The purpose is to control for a number of variables that can affect the size of the estimated coefﬁcients. This will allow us to explain part of the large variance in the estimated coefﬁcients. The econometric equation is speciﬁed as follows: PSi = a + ∑kk Dk + i

(1)

PLi = c + ∑kk Dk + i

(2)

where PSi and PLi are the observed short-term and long-term parameters measuring the effect of monetary policy. The variables Dk are variables expressing a particular characteristic of the study from which parameter i was obtained or from the country involved. We distinguish between the following characteristics: •

The countries analysed in the study: in this case, each country is represented by a separate dummy variable.

Monetary policy and real economy 35 •

•

•

•

The econometric technique used. We distinguish between ﬁve types of econometric methods. The ﬁrst one uses ‘plain vanilla’ VARs, i.e. the method used to impose identifying restrictions is based on imposing a recursiveness ordering (Choleski decomposition). The second one uses SVARs. This is a VAR method that relies on an economic theory to impose prior restrictions on (some) parameters of the model. Quite often, this method imposes a restriction on the long-term effect of monetary policy (e.g. a zero restriction on the long-term output effect). The third one, FAVAR, uses dynamic factor analysis, and the fourth one, MARKOV, uses switching in regimes. Finally, the ﬁfth technique relies on traditional econometric modelling.4 The variable used to measure output. We distinguish between GDP, industrial production (PROD), and output gap (GAP). Each of these measures is represented by a separate dummy. The sample period during which the studies were performed. We distinguish between studies in which the sample period starts in the 1960s, the 1970s and the 1980s. This distinction is introduced to ﬁnd out whether the coefﬁcients measuring the effectiveness of monetary policy have changed over time. We introduce three dummy variables: SIXTIES, SEVENTIES, and EIGHTIES. The exchange rate regime. We distinguish between two exchange rate regimes, ﬁxed and ﬂexible. The countries on a ﬂexible exchange rate regime are the United Kingdom, the United States, Japan and Germany. The others (EMS countries and emerging countries) are on a ﬁxed exchange rate regime.5

A ﬁnal issue concerns the weights given to the different publications. The quality of the different studies is not the same. One would therefore like to adjust for the quality of the studies. It is, however, very difﬁcult to do this without introducing subjective judgment. This could lead to the possibility of a selection bias, whereby the researcher gives a higher weight to those studies, which come close to his priors. We have not attempted to do this. The only quality criterion we have maintained is the length of the sample periods of the different studies.6 Thus, studies that use a longer sample period, and thus more information, receive a higher weight than studies using a shorter sample period. The way we do this is by weighting each study by the length of the sample period (expressed as a percent of the longest sample period). We will present results using both weighted and unweighted data. We show the results for the short-term output effects in Table 3.1.7 We have structured the model in such a way that we have one dummy variable for each country. For the other variables (econometric method, output measure, and sample period), we eliminate one of them, i.e. we eliminate VAR, GDP, and SIXTIES. As a result, the country coefﬁcients represent the effects of monetary policy in each country in studies using VAR as an econometric method, GDP as a measure of output, with a sample period starting in the 1960s. In this way, we eliminate differences between countries that have to do with the use of different econometric methods, output measures, and sample periods. The coefﬁcients of the remaining

36 P. De Grauwe and C.C. Storti variables then measure how the different econometric methods, the output measures, and the sample periods affect these country coefﬁcients on average. We observe that the country coefﬁcients are very similar and in most cases statistically different from zero. This suggests that studies estimating the effects of monetary policy using VARs, GDP as the measure of output, and samples starting in the 1960s (the benchmark case) ﬁnd signiﬁcant short-term effects on output in most countries in the sample. There are a few additional observations one can make from Table 3.1. First, Table 3.1 Regression results of equation (1): short-term output coefﬁcients

Variable Austria Belgium Denmark Eurozone Finland France Germany Ireland Italy Netherlands Portugal Spain Sweden Greece Luxembourg UK USA Japan Australia Canada Emerging IND GAP SVAR ECON FAVAR MARKOV SEVENTIES EIGHTIES MONEY FLOAT R-squared S.E. of regression Sum-squared resid. Log-likelihood

Unweighted regression

Weighted regression

Coefﬁcient

t-Statistic

Coefﬁcient

t-Statistic

–2.83 –2.09 –0.76 –0.84 –2.64 –2.94 –1.72 –1.02 –2.07 –2.13 –1.70 –2.03 –1.95 –0.93 –0.75 –1.38 –1.77 –0.27 –0.81 –0.99 0.25 –0.62 1.16 1.88 2.85 –1.96 0.09 –0.19 –1.10 1.54 0.05

–0.24 –0.22 –0.15 –0.22 –0.25 –0.22 –0.30 –0.18 –0.16 –0.20 –0.18 –0.18 –0.31 –0.24 –0.19 –0.27 –0.25 –0.16 –0.32 –0.31 –0.09 –0.03 0.06 0.11 0.07 –0.18 –0.06 0.03 0.07 0.13 0.05

–2.93 –2.46 –1.21 –1.34 –2.92 –3.42 –2.24 –1.61 –2.48 –2.39 –1.84 –2.50 –2.29 –1.61 –0.91 –1.97 –2.12 –1.08 –1.35 –1.75 –1.16 –0.67 0.86 2.02 2.16 –2.01 –0.55 0.52 1.38 2.14 0.46

–0.38 –0.31 –0.16 –0.22 –0.38 –0.32 –0.34 –0.19 –0.22 –0.29 –0.28 –0.25 –0.43 –0.23 –0.26 –0.28 –0.31 –0.06 –0.31 –0.28 0.03 –0.04 0.14 0.17 0.16 –0.29 0.02 –0.02 –0.08 0.16 0.01 0.18 0.33 23.30 –61.80

0.15 0.20 8.30 57.10

Monetary policy and real economy 37 the way output is measured does not seem to affect the size of the coefﬁcients. Second, the use of different econometric methods matters. Studies using SVARs produce short-term output coefﬁcients that on average are signiﬁcantly smaller (in absolute value) than the coefﬁcients obtained with VARs. The same is true for studies using econometric models. The opposite holds for studies using VARs that combine dynamic factor analysis. Third, there is some evidence (in the weighted regressions) that the coefﬁcients are smaller (in absolute value) in the studies with sample periods starting in the 1980s. The results of estimating Equation (2) for the long-term output coefﬁcients are shown in Table 3.2. A ﬁrst striking observation is that the long-term output coefﬁcients in the different countries are generally not zero. In many cases, they are statistically different from zero. This is surprising, as the consensus view described in the introduction tells us that in the long run the output effects of monetary policy shocks should be zero. This result, however, strongly depends on the econometric method that is used. As in Table 3.1, the country coefﬁcients in Table 3.2 represent the benchmark case, i.e. the use of VARs, GDP, and the 1960s as starting sample period. An analysis of the coefﬁcients of SVAR, ECON, FAVAR, and MARKOV in Table 3.2 reveals that in the studies that use SVARs and econometric models, the long-term output coefﬁcients are much lower (in absolute value) and are close to zero. The use of dynamic factor analysis, however, again leads to an increase in the long-term output coefﬁcients. Thus, we ﬁnd that econometric methods that use SVARs and econometric models produce results that are in accordance with the consensus view. This is not really surprising. These methods typically impose the long-term condition that the output effect is zero. In contrast, the econometric methods that do not impose such a long-run restriction, the ‘plain vanilla’ VAR, ﬁnd that in the long run (after ﬁve years), there are still signiﬁcant output effects of monetary policies in most countries. Put differently, if one ‘allows the data to speak’, the consensus view of monetary policy neutrality does not seem to hold. An important issue that arises here is whether ﬁve years can be considered the long run. One could argue that a fair test of the neutrality proposition should extend the time horizon beyond ﬁve years. We look into this problem in Section ‘The role of macroeconomic variables’. We note, however, that the ﬁnding that monetary policy effects on output last ﬁve years or more in many countries is troublesome for the consensus view described in the introduction. Finally, it is worth mentioning here that as in the case of the short-term coefﬁcients, the size of the long-term output coefﬁcient appears to have declined in the studies using more recent sample periods (after 1980). Econometric analysis: price effects In this section, we analyse the short-term and long-term price effects of monetary policies. We will proceed in the same way as in the previous section. We estimate the econometric model consisting of Equations (1) and (2), where PSi and PLi now represent the estimated short-term and long-term price effects of

38 P. De Grauwe and C.C. Storti Table 3.2 Regression results of equation (2): long-term output coefﬁcients

Variable Austria Belgium Denmark Finland France Germany Ireland Italy Netherlands Portugal Spain Greece Luxembourg Eurozone Sweden UK USA Japan Australia Canada Emerging IND GAP SVAR ECON FAVAR MARKOV SEVENTIES EIGHTIES FLOAT MONEY R-squared S.E. of regression Sum-squared resid. Log-likelihood

Unweighted regression

Weighted regression

Coefﬁcient

t-Statistic

Coefﬁcient

t-Statistic

–3.03 –1.42 –0.83 –1.13 –2.37 –1.17 –1.16 –0.96 –2.38 –2.19 –1.73 –2.18 –0.84 –0.93 –0.76 –0.63 –1.02 0.48 –0.75 –0.80 –1.35 –1.54 1.81 3.23 2.76 –3.35 0.85 –0.74 1.41 –0.77 –1.73

–0.24 –0.15 –0.14 –0.15 –0.16 –0.22 –0.18 –0.11 –0.21 –0.22 –0.18 –0.40 –0.21 –0.21 –0.17 –0.15 –0.22 –0.08 –0.28 –0.20 –0.14 –0.06 0.10 0.14 0.08 –0.14 0.08 0.01 0.13 0.01 –0.07

–3.18 –1.90 –1.35 –1.88 –2.64 –2.04 –1.95 –1.75 –2.65 –2.53 –2.44 –2.92 –1.13 –1.53 –1.43 –1.40 –2.38 –0.70 –1.34 –1.20 –1.88 –1.36 1.77 2.92 2.26 –2.06 0.85 0.30 2.85 0.13 –1.21

–0.36 –0.17 –0.13 –0.14 –0.23 –0.18 –0.17 –0.09 –0.29 –0.30 –0.20 –0.47 –0.25 –0.20 –0.14 –0.10 –0.14 0.09 –0.24 –0.21 –0.15 –0.10 0.17 0.24 0.16 –0.37 0.12 –0.05 0.10 –0.09 –0.16 0.26 0.28 16.08 –23.27

0.25 0.18 6.27 78.19

monetary policy shocks. A note of warning is necessary here. Because not all the empirical studies of the effect of monetary policies report results of the effects on the price level, we have fewer data points in the sample (185). As a result, the statistical quality of the econometric results is weaker than in the previous section. We ﬁrst concentrate on the short-term price effects (Equation (1)). We show the results of estimating Equation (1) in Table 3.3, both for the weighted and for the unweighted data. We ﬁnd that most of the country coefﬁcients are close to

Monetary policy and real economy 39 Table 3.3 Regression results of equation (1): short-term price coefﬁcients

Variable Austria Belgium Denmark Eurozone Finland France Germany Greece Ireland Italy Japan Luxembourg Netherlands Portugal Spain Sweden UK USA Australia Canada Emerging SVAR ECON FAVAR SEVENTIES EIGHTIES MONEY R-squared S.E. of regression Sum-squared resid. Log-likelihood

Unweighted regression

Weighted regression

Coefﬁcient

t-Statistic

Coefﬁcient

t-Statistic

–1.28 –0.94 0.01 0.16 –0.47 0.04 0.19 –0.34 –0.66 –0.33 2.67 –0.08 –1.64 –0.90 –0.41 –0.17 –0.20 –1.25 –0.32 –1.55 1.00 0.23 –0.31 0.01 0.28 –1.11 2.53

–0.05 –0.04 0.01 0.03 –0.06 –0.003 0.004 –0.04 –0.03 –0.02 0.15 0.02 –0.07 –0.04 –0.01 0.03 0.002 –0.03 –0.04 –0.19 0.020 –0.003 –0.04 0.01 –0.003 –0.02 0.42

–0.79 –0.62 0.04 0.24 –0.80 –0.07 0.07 –0.39 –0.41 –0.35 2.29 0.17 –0.99 –0.56 –0.17 0.41 0.05 –1.21 –0.26 –1.95 0.40 –0.11 –1.02 0.12 –0.08 –0.42 3.68

–0.15 –0.11 0.002 0.03 –0.06 0.003 0.02 –0.06 –0.09 –0.03 0.30 –0.02 –0.20 –0.13 –0.05 –0.03 –0.02 –0.06 –0.08 –0.27 0.08 0.01 –0.02 0.001 0.02 –0.07 0.50 0.18 0.24 8.59 10.18

0.17 0.14 2.81 99.64

zero. None is statistically different from zero. This contrasts with the short-term output coefﬁcients which were found to be statistically different from zero for most countries. These results are in line with the well-known empirical regularity discussed earlier, i.e. that prices are stickier than output. In the short run (i.e. after one year), prices do not react to monetary policy shocks. The next step in the analysis consists in performing the same analysis for the long-term price coefﬁcients. The results are shown in Table 3.4. We now ﬁnd country coefﬁcients that are statistically different from zero in almost all cases. Thus, in the long run (after ﬁve years or more), monetary policy shocks have signiﬁcant effects on the aggregate price levels in almost all countries. We also note that the statistical quality of the regression is higher when we use weighted

40

P. De Grauwe and C.C. Storti

Table 3.4 Regression results of equation (2): long-term price coefﬁcients

Variable Austria Belgium Denmark Eurozone Finland France Germany Greece Ireland Italy Japan Luxembourg Netherlands Portugal Spain Sweden UK USA Australia Canada Emerging SVAR ECON FAVAR SEVENTIES EIGHTIES MONEY R-squared S.E. of regression Sum-squared resid. Log-likelihood

Unweighted regression

Weighted regression

Coefﬁcient

t-Statistic

Coefﬁcient

t-Statistic

–2.13 –2.29 –0.98 –0.86 –1.83 –2.33 –2.47 –2.05 –1.62 –1.63 –0.46 –1.22 –2.34 –1.92 –3.40 –2.08 –2.84 –6.53 –3.40 –4.39 –1.36 1.41 1.96 –0.67 0.80 1.44 4.59

–0.24 –0.25 –0.22 –0.23 –0.22 –0.22 –0.24 –0.32 –0.22 –0.19 –0.16 –0.22 –0.25 –0.25 –0.33 –0.28 –0.26 –0.27 –0.56 –0.79 –0.16 0.06 0.03 –0.02 0.09 0.16 0.81

–2.87 –2.99 –1.80 –1.46 –2.47 –3.28 –3.81 –2.76 –2.49 –3.22 –2.24 –1.39 –3.15 –2.79 –4.45 –2.73 –3.89 –8.24 –3.42 –5.70 –2.79 1.47 0.72 –0.36 1.71 3.55 6.29

–0.34 –0.36 –0.22 –0.26 –0.30 –0.29 –0.29 –0.45 –0.27 –0.18 –0.06 –0.37 –0.34 –0.33 –0.47 –0.40 –0.35 –0.41 –1.03 –1.13 –0.14 0.11 0.16 –0.07 0.08 0.12 1.10 0.35 0.28 9.41 –8.40

0.43 0.15 2.68 79.21

data (higher R2 and more signiﬁcant coefﬁcients). In addition, the country coefﬁcients are more similar in the weighted regressions. The use of different econometric techniques does not affect the previous results very much. In particular, SVARs produce pretty much the same results as ‘plain vanilla’ VARs. This contrasts with the results obtained in the previous section dealing with the output effects. Similarly, there is little evidence that the price effects have changed over time. The coefﬁcients of the SEVENTIES and EIGHTIES dummies are not statistically different from zero. Finally, there is some evidence that the long-term price effects of monetary policies have tended to become smaller. This can be seen from the coefﬁcient of the EIGHTIES variable that is positive and signiﬁcant in the case of the

Monetary policy and real economy 41 weighted regression. This is probably related to the fact that since the 1980s, inﬂation has come down signiﬁcantly thus reducing its sensitivity to monetary shocks.

Are the effects of monetary policy different in the United States and the Eurozone? In this section, we analyse the issue of whether the effectiveness of monetary policies in the United States and in the Eurozone countries is different. A consensus seems to have emerged that because of the existence of labour market rigidities, monetary policies in the Eurozone are less effective in inﬂuencing output than is the case in the United States. The argument is quite often phrased as follows: ‘rigidities in the labour markets tend to limit the pace at which an economy can grow without fueling inﬂationary pressures’ (European Central Bank 2004: 21). Thus, when the ECB lowers the interest rate to stimulate the Eurozone economy, this will quickly be transmitted into higher prices with only limited effects on output. Since the US economy is less rigid, the Federal Reserve can more easily stimulate the economy without introducing inﬂationary pressures. An inﬂuential paper substantiating this view is Angeloni et al. (2003). These authors came to the conclusion that a one percentage point increase in the shortterm interest rate tends to have a substantially stronger output effect in the United States than in the Eurozone. In addition, they identiﬁed this difference to be due to a signiﬁcantly higher consumption effect of monetary policy changes in the United States as compared to the Eurozone. As an example, we show the effects of monetary policy on output, consumption, and investment in the United States and the Eurozone as obtained by Angeloni et al. (2003) in Table 3.5. It can be seen that the output effects of monetary policy changes are more than twice as strong in the United States than in the Eurozone and that most of this difference comes from much larger consumption effects in the United States than in the Eurozone. This evidence has strengthened the perception that while an activist monetary policy such as the one followed by the Federal Reserve during the last decade Table 3.5 Effect of one percentage point increase in the short-term interest rate on a number of macroeconomic variables in the US and in the Eurozone USA

CPI GDP Consumption Investment

Eurozone

1 year

2-years

3-years

–0.07 –0.35 –0.37 –0.31

–0.41 –1.28 –1.35 –1.79

–1.01 –1.37 –1.44 –3.16

Source: Angeloni et al. (2003).

■

1 year

2-years

3-years

–0.09 –0.22 –0.12 –0.34

–0.21 –0.38 –0.23 –1.04

–0.31 –0.31 –0.19 –1.22

42

P. De Grauwe and C.C. Storti

may be sensible for the United States, it is not appropriate for the Eurozone where as a result of rigidities, such an activist monetary policy would barely affect output, while it would mainly lead to more inﬂation. This view of the relative ineffectiveness of monetary policy in the Eurozone has now become the conventional one and is often formulated in the popular ﬁnancial press (see e.g. The Economist 2005: 20). There is a large literature on the relation between price and wage rigidities and the optimal design of monetary policies. This literature has led to a number of propositions that can be summarized as follows. First, in a world of perfectly ﬂexible wages and prices, monetary policy has no effect on output (Woodford 2003). In such a world, there is also no need for using monetary policy to stabilize output. Second, in a world characterized by price and wage rigidities, monetary policy can be quite potent in inﬂuencing output at least in the short run (see e.g. Fischer 1977; Taylor 1980; Clarida et al. 1999). In fact, it is only because prices and wages are rigid that monetary policy can affect output in the short run (see also our discussion in the introduction). In this sense, the view that monetary policies in the Eurozone are ineffective because of the existence of rigidities is surprising. Without rigidities, monetary policy cannot affect output. This leads to a third proposition. The effectiveness of monetary policy depends on the nature of these rigidities. The consensus today is that nominal wage rigidities increase the output effects of monetary policy shocks. In contrast, real wage rigidities reduce the effectiveness of monetary policies in affecting output (see Gylfason and Lyndbeck 1994; Soskice and Iversen 2000; Tabellini 2001). It follows that it is important to specify the nature of the structural rigidities to understand how these affect the transmission of monetary policies. Some rigidities increase the effectiveness of monetary policies in affecting output, others reduce this effectiveness. Thus, the issue of how rigidities affect the effectiveness of monetary policies is an empirical one to which we now turn. The way we proceed in our empirical analysis is to perform Wald tests for equality of the coefﬁcient of the United States and the Eurozone countries reported in the different tables of the previous section. We do this for the output and price coefﬁcients. Note that the statistical testing procedure is such that when we test for equality, we control for differences in econometric methodology, differences in the measurement of output, and differences in the sample period. We ﬁrst show the results for the output coefﬁcients in Table 3.6. The Wald test on the short-term coefﬁcients reveals that we cannot reject the hypothesis that these coefﬁcients are equal. The results concerning the long-term coefﬁcients are more subtle. We ﬁnd that when we apply the test on all these coefﬁcients, we should reject the hypothesis that these are equal. It turns out, however, that if we remove the outlier (Greece in the weighted regression), we cannot reject the hypothesis that the United States and Eurozone long-term output coefﬁcients are equal. From the preceding analysis, we can conclude that there is little evidence that the transmission of monetary shocks into output in the Eurozone is any different

Monetary policy and real economy 43 Table 3.6 Wald test: equality of the US and Eurozone output coefﬁcients Short-term coefﬁcients (Equation (1)) weighted regression F-statistic 0.255 Probability Chi-square 3.320 Probability

0.99 0.99

Long-term coefﬁcients (Equation (2)) weighted regression F-statistic 0.590 Probability Chi-square 7.679 Probability

0.85 0.86

Long-term coefﬁcient (Equation (2)) weighted regression, outlier Greece excluded F-statistic 0.395 Probability Chi-square 4.743 Probability

0.96 0.96

from the transmission in the United States. Both the short-term and the longterm output coefﬁcients in the Eurozone countries and in the United States appear to be of the same order of magnitude if we control for differences in econometric methodology, differences in the measurement of output, and differences in the sample periods. As argued earlier, the hypothesis that monetary policies in the Eurozone are ineffective in inﬂuencing output has a corollary as far as the transmission into prices is concerned. It implies that a monetary expansion in the Eurozone will be transmitted more quickly and more completely into price increases (see European Central Bank 2004: 21). We now test this corollary by applying similar Wald tests on the price coefﬁcients. We show the results in Table 3.7. We cannot reject the hypothesis that both the short-term and the long-term price coefﬁcients are equal. We conclude from this that there is no evidence that monetary policy shocks lead to a quicker and stronger transmission into prices in the Eurozone than in the United States. The hypothesis that Eurozone monetary policy is less effective than US monetary policy because of a quicker and stronger transmission of Eurozone monetary policies into prices has no empirical backing.

The role of macroeconomic variables Macroeconomic variables also matter in explaining the differences in the output and price coefﬁcients. In this section, we analyse the importance of macroeconomic variables. We do this by adding macroeconomic variables that describe Table 3.7 Wald test: equality of the US and Eurozone price coefﬁcients Short-term coefﬁcients (Equation (1)) weighted regression F-statistic 0.218 Probability Chi-square 2.837 Probability

0.99 0.99

Long-term coefﬁcients (Equation (2)) weighted regression F-statistic 0.435 Probability Chi-square 5.657 Probability

0.95 0.96

44

P. De Grauwe and C.C. Storti

the nature of the macroeconomic regime of the countries involved in the empirical studies to our econometric Equations (1) and (2). The econometric equation is now speciﬁed as follows: PSi = a + ∑kk Dk + ∑jj Mj + i

(3)

PLi = c + ∑kk Dk + ∑jj Mj + i

(4)

where PSi and PLi are the observed short-term and long-term parameters measuring the effect of monetary policy. As before, the variables Dk are dummy variables expressing a particular characteristic of the study from which parameter i was obtained. The variables Mj are macroeconomic variables associated with country i (that is represented by parameter i). We use the following variables: •

• •

•

•

The openness of the country involved, as measured by the ratio of its exports to its GDP. We expect that the output effects of domestic monetary policy shocks are smaller in relatively open countries than in relatively closed ones. The size of countries as measured by their GDP in dollars. The exchange rate regime. We distinguished between two exchange rate regimes, ﬁxed and ﬂexible. The countries on a ﬂexible exchange rate regime are the United Kingdom, the United States, Japan, and Germany. The others (EMS countries and emerging countries) were on a ﬁxed exchange rate regime.8 The rate of inﬂation that prevailed on average during the sample period over which the output coefﬁcient was estimated. There is a theoretical presumption that the effect of a monetary expansion on output declines with the level of inﬂation. Several theoretical models can be invoked to substantiate this. The most inﬂuential is Lucas (1973) ‘island model’. In this model, the aggregate supply equation depends on the relative variance of real and nominal disturbances. The implication is that in a regime of high nominal variability, an increase in prices is more likely to be interpreted as resulting from an aggregate price increase than in a regime of low nominal variability. As a result, the real effects of such an increase in prices will be reduced. A similar analysis can be performed using the Philips curve as a tool. In such a framework, the Philips curve is also non-linear in the rate of inﬂation. Thus, when inﬂation is high, one will need a stronger monetary surprise to generate a given increase in output (decline in unemployment) than when inﬂation is low.9 This proposition was also tested by Lucas (1973). The importance of the banking sector as measured by the ratio of the consolidated balance sheet of the banking sector over GDP. The theory is not clear about how this variable affects the output effects of monetary policy shocks. We introduce this variable here to ﬁnd out whether differences in the size of the banking sector can explain the differences in the estimated

Monetary policy and real economy 45 output coefﬁcients. Because of the limited availability of data, we had to estimate Equations (3) and (4) on a smaller sample of coefﬁcients than in the previous sections. In addition, the country dummies are now highly correlated with some of the macroeconomic variables (size, openness, and inﬂation). Using Wald tests, we found that the differences in the country dummy coefﬁcients are not statistically signiﬁcant. As a result, we estimated Equations (3) and (4) restricting the country coefﬁcients to be equal. The results are shown in Tables 3.8 and 3.9. The constant term in Tables 3.8 and 3.9 represents the effect of the omitted dummies. As before, the omitted dummies are VAR, GDP, and SIXTIES. Thus, the constant term measures the coefﬁcient of studies using VAR methods, using GDP as the measure of output, and using a sample period starting in the 1960s. The most important results can be summarized as follows. As the results of the study characteristics are similar to those obtained in Section ‘A meta-analysis of the effects of monetary policy on output and prices’, we concentrate our analysis on the coefﬁcients of the macroeconomic variables. Our major result is that the level of inﬂation matters. We ﬁnd that inﬂation tends to reduce the output effect of monetary policy, both in the short run and in the long run. From Table 3.8, we ﬁnd that for every percentage point increase in inﬂation the short-term output parameter declines (in absolute value) by approximately 0.04. This effect is signiﬁcant in the regression using unweighted data but is less so in the regression using weighted data. We ﬁnd a similar result for the long-term output coefﬁcients (Table 3.9). This result is conforming to economic theory. It is interesting to have an insight in the quantitative importance of the effect of inﬂation. The median inﬂation rate in the sample is 5.2 per cent. (In Appendix 3.1, we show the distribution of the inﬂation rates in the sample of countries.) Thus, for the median inﬂation rate, the short-term output coefﬁcient is reduced by 0.2. For the highest inﬂation country in the sample (16 per cent), the shortterm output coefﬁcient is reduced by 0.622. Thus, the short-term output effect is reduced by half compared to the benchmark in the highest inﬂation country. We obtain similar results for the long-term coefﬁcients. We ﬁnd that for the median inﬂation country, the long-term output coefﬁcient is reduced by 0.21, while for the highest inﬂation country, it is reduced by 0.64. As a result, for the highest inﬂation countries in the sample, the long-term output effects of monetary policies are close to zero. The interesting aspect of this result is that for the low inﬂation countries, these long-term output effects are strong and signiﬁcant. The other variables in the regression equation do not have a signiﬁcant effect on the output coefﬁcients. The macroeconomic variables such as openness, size of the countries, and the importance of the banking sector do not create signiﬁcant differences in the output effects of monetary policy shocks. This may seem surprising. For example, one may expect that openness and size matter. In particular, the output effects of monetary policies should be smaller in relatively small and open economies because much of the domestically generated

46 P. De Grauwe and C.C. Storti Table 3.8 Short-term output coefﬁcients (equation (3)). Included observations: 127 Unweighted data Weighted data Variable

Coefﬁcient

t-Statistic

Coefﬁcient

t-Statistic

C SVAR ECON IND GAP SEVENTIES EIGHTIES INFLATION FIX PUBDATE SIZE OPEN BANKING

–1.333 –0.388 –0.125 0.140 –0.174 0.066 –0.275 0.038 0.046 0.075 3.9E-05 0.141 0.067

–4.543 –2.555 –0.841 0.967 –0.853 0.415 –1.646 1.895 0.294 2.821 0.979 0.781 0.599

–1.067 –0.124 –0.012 0.061 –0.163 0.058 –0.213 0.023 0.064 0.048 5.02E-05 0.126 0.052

–4.536 –0.872 –0.104 0.469 –0.910 0.470 –1.415 1.372 0.489 2.197 1.308 0.816 0.711

R-squared Adjusted R-squared Mean-dependent VAR

0.194 0.117 –0.394

0.130 0.046 –0.366

Table 3.9 Long-term output coefﬁcients (equation (4)). Included observations: 122 Unweighted data Weighted data Variable

Coefﬁcient

t-Statistic

Coefﬁcient

t-Statistic

C SVAR ECON IND GAP SEVENTIES EIGHTIES INFLATION FIX PUBDATE SIZE OPEN BANKING

–0.769 0.282 0.378 0.208 0.467 –0.140 0.152 0.039 0.228 –0.007 4.73E-05 0.198 –0.023

–2.689 1.829 2.413 1.390 2.341 –0.879 0.901 1.909 1.447 –0.301 1.191 1.103 –0.213

–0.848 0.411 0.500 0.210 0.560 –0.260 0.118 0.038 0.273 –0.009 5.10E-05 0.283 –0.005

–2.528 1.977 2.882 1.087 2.182 –1.432 0.538 1.597 1.446 –0.309 0.927 1.263 –0.051

R-squared Adjusted R-squared Mean-dependent VAR

0.194 0.111 –0.252

0.256 0.181 –0.304

Monetary policy and real economy 47 monetary shocks spill over to the rest of the world. However, in small open economies, most of the monetary policy shocks are not generated by domestic monetary authorities. They are typically the result of monetary policy shocks originating in large countries. To give an example: when the German Bundesbank increased its interest rate, central banks of countries like the Netherlands and Belgium routinely increased their short-term interest rates a few minutes later. As a result, the monetary policy shock occurred in many countries at the same time. It is therefore not so surprising that monetary policy shocks can have similar effects in large and small countries. We conclude this section by discussing the results of estimating Equations (3) and (4) for the price effects, which are presented in Tables 3.10 and 3.11. We ﬁnd as before that in the short run, monetary policy shocks (increase in the interest rate) have no signiﬁcant effect on prices. In the long run, these price effects are strong and signiﬁcant. We also ﬁnd that the sign of the inﬂation variable in the short-run equation is correct, although its signiﬁcance is weak (see Table 3.10). Thus, in high inﬂation countries, the monetary policy shock appears to a stronger impact on prices than in the low inﬂation countries. This is the corollary of what we found for the short-term output effects, i.e. in high inﬂation countries, monetary policy shocks have a weaker effect on output than in low inﬂation countries. From Table 3.10, we conclude that in the long run, there is no difference in the price effects of monetary policies between low and high inﬂation countries.

Conclusion The issues of whether monetary policies affect output and, if so, how long this effect lasts have been hotly debated by economists. The consensus today is that Table 3.10 Short-term price coefﬁcients (equation (3)). Included observations: 86 Unweighted data Weighted data Variable

Coefﬁcient

t-Statistic

Coefﬁcient

t-Statistic

C SVAR ECON INFLATION SIZE OPEN BANKING PUBDATE FIX

0.216 –0.257 –0.089 –0.022 4.03E-06 –0.018 0.050 –0.021 0.134

0.942 –2.658 –0.767 –1.276 0.120 –0.145 0.642 –1.247 1.159

0.219 –0.364 –0.114 –0.025 3.46E-05 –0.051 0.064 –0.028 0.256

1.047 –3.224 –0.989 –1.395 0.893 –0.366 0.997 –1.662 1.984

R-squared Adjusted R-squared Mean-dependent VAR

0.116 0.024 –0.098

0.201 0.117 –0.096

48 P. De Grauwe and C.C. Storti Table 3.11 Long-term price coefﬁcients (equation (4)). Included observations: 84 Unweighted data Weighted data Variable

Coefﬁcient

t-Statistic

Coefﬁcient

t-Statistic

C SVAR ECON INFLATION SIZE OPEN BANKING PUBDATE FIX SEVENTIES EIGHTIES

–1.230 0.320 0.640 0.010 0.0001 0.216 0.034 0.022 0.624 –0.829 –0.013

–1.989 1.156 1.814 0.211 1.281 0.621 0.164 0.433 1.795 –2.182 –0.034

–1.154 0.292 0.908 0.005 0.0001 0.342 0.014 0.005 0.589 –0.996 0.015

–1.887 0.851 2.453 0.095 1.172 0.814 0.075 0.104 1.449 –2.593 0.035

R-squared Adjusted R-squared Mean-dependent VAR

0.195 0.085 –0.403

0.261 0.160 –0.493

monetary policies do affect output. What is less clear is how long these effects last. In this chapter, we have analysed these questions using a meta-analysis of the effects of monetary policies on output and prices. We can summarize the main results concerning the output effects of monetary policy as follows. First, there is a large variation in the reported output effects of monetary policies. This is the case both with the short-term and with the long-term effects. Second, we are able to explain part of these large variations by a number of variables, although much remains unexplained. Third, a signiﬁcant part of the wide variation in the long-term output effects is due to the use of different econometric techniques. In particular, the use of VARs produces long-term effects of monetary policies, while the use of SVARs leads to signiﬁcantly lower long-term effects. This result is not without importance. It suggests that techniques that use economic theory to constrain parameters lead to signiﬁcantly different effects from those techniques that ‘allow the data to speak’. More speciﬁcally, SVARs typically impose the condition that the long-term output effects of monetary policies should be zero. This condition is based on the now prevailing theoretical insights inﬂuenced by monetarism and the real business cycle theory that money is neutral in the long run. The worrisome aspect of these theories is that one has to assume this condition to hold in order to validate it empirically. Put differently, in those econometric studies that do not impose long-run neutrality, the long-run neutrality of money is rejected. This certainly calls into question one of the fundamental tenets of the new consensus in macroeconomics. It should be pointed out that there is still the issue of how long the long run is. In the meta-analysis, we have used as a cut-off point ﬁve years or more. It

Monetary policy and real economy 49 could be argued that the long run is longer than ﬁve years and that a fair test of the long-run neutrality proposition would be to analyse longer time horizons. This has not been done in this chapter, because the reported econometric studies typically do not report results that go beyond a time horizon of ﬁve to seven years. A fourth result of our analysis is that the level of inﬂation affects the effectiveness of monetary policies. More particularly, we found that in the countries which experienced low inﬂation, the output effects of monetary policy shocks are substantial. This is the case both for the short-term and for the long-term effects. In the high inﬂation countries of our sample, these output effects are much smaller. Moreover, the long-term output effects of monetary policies all but vanish for the highest inﬂation countries. This conﬁrms the theory, which suggests that in a low inﬂation environment, monetary policies are quite effective in inﬂuencing output, both in the short run and in the long run. In other words, like in the United States and the Eurozone countries, one should expect that monetary policies have relatively strong output effects in comparison to countries like Argentina with a history of high inﬂation. Fifth, we could not ﬁnd any signiﬁcant differences in the output and price effects of monetary policies in the United States and in the Eurozone countries. There is a popular view according to which monetary policies in the Eurozone are ineffective in boosting output because supply rigidities quickly lead to higher inﬂation, while in the United States, monetary policies are capable of boosting output without strong inﬂationary effects. The existing econometric estimates of the output and price effects of monetary policies in the United States and the Eurozone countries do not allow us to draw such a conclusion. Since the effectiveness of monetary policies very much depends on the nature of price and wage rigidities, these results suggest that the United States and the Eurozone are less different in terms of wage and price rigidities than is commonly thought. Much of the discussion today among macroeconomists has shifted to the issue of whether the output effects of monetary policies can be exploited by the monetary authorities and whether rules should govern the conduct of monetary policy. The consensus view today, based on the use of monetarist and real business cycle models, is that monetary authorities should not actively try to ﬁnetune output movements. Although the issues concerning monetary policy rules were not the focus of our research, our results allow us to shed some light on these questions. In particular, our ﬁnding that the long-run neutrality of money has a weak empirical basis calls into question the use of models whose central theoretical building block is the long-run neutrality of money. These models are now used for policy purposes, and they have led to the widespread view among policymakers that monetary policy should only be used to stabilize the price level and should not be employed for other purposes. It should be clear, however, that these conclusions have more to do with theoretical convictions about how the world should work, than with hard empirical evidence of how the world actually works.

50 P. De Grauwe and C.C. Storti

Appendix: Descriptive statistics, full sample 100

Series: OUTPUTST Sample 1,278 Observations 278

80

Mean 0.297346 Median 0.250000 Maximum 4.853800 Minimum 4.100000 Standard deviation 0.694908 Skewness 0.184213 Kurtosis 20.23044

60

40

Jarque–Bera Probability

20

0

3.75 2.50 1.25 0.00

1.25

2.50

3.75

3440.527 0.000000

5.00

Short-term output coefﬁcient (OUTPUTST), full sample.

100

Series: OUTPUTLT Sample 1,278 Observations 247

80

Mean Median Maximum Minimum Standard deviation Skewness Kurtosis

60

40

20

0

Jarque–Bera Probability 4

3

2

1

0

1

Long-term output coefﬁcients (OUTPUTLT), full sample.

0.237818 0.070000 1.106700 4.200000 0.531429 3.242688 20.70667

3659.574 0.000000

100 Series: SHORT Sample 1,278 Observations 278

80

Mean Median Maximum Minimum Standard deviation Skewness Kurtosis

60

40

Jarque–Bera Probability

20

0.043338 0.040000 3.232500 2.800000 0.425706 1.145268 30.58872

5843.677 0.000000

0 3

2

1

0

1

2

3

Short-term price coefﬁcients (SHORT), full sample. 60

Series: LONG Sample 1,185 Observations 177

50

Mean Median Maximum Minimum Standard deviation Skewness Kurtosis

40 30 20

Jarque–Bera Probability

10

0

0.300804 0.110000 3.279000 6.000000 0.932080 2.530144 17.08711

1652.393 0.000000

5.00 3.75 2.50 1.25 0.00 1.25 2.50

Long-term price coefﬁcient (LONG), full sample. 20

Series: INFLATION Sample 1,139 Observations 124

15

5

5.824748 Mean 5.209471 Median 16.05292 Maximum 1.925000 Minimum Standard deviation 2.847042 1.088077 Skewness 3.883654 Kurtosis

0

Jarque–Bera Probability

10

2

4

6

8

10

12

Frequency distribution of inﬂation rates.

14

16

28.50187 0.000001

52 P. De Grauwe and C.C. Storti

Notes 1 See Stanley (2001) for a critical analysis of the use of meta-analysis in economics. 2 Many VAR and SVAR studies only report the graphs of the impulse response functions. We therefore enlarged these graphs considerably allowing us to measure the coefﬁcients of the impulse response functions with great precision. 3 Thus, the parameter estimates obtained from econometric models do not distinguish between anticipated and unanticipated interest rate shocks. Typically, VAR-based estimates relate to unanticipated interest rate shocks. 4 There is, of course, scope for further distinctions in the econometric techniques. 5 One could introduce ﬁner distinctions between different exchange rate regimes. For example, one could use the IMF classiﬁcation of exchange rate regimes. This classiﬁcation has been criticized, however. See Calvo and Reinhart (2000). 6 Another possible quality criterion could be the signiﬁcance of the estimated coefﬁcients. The trouble with this is that many studies do not report conﬁdence levels of the estimated coefﬁcients. 7 The variable FLOAT was dropped. We did not ﬁnd any signiﬁcant difference between countries with ﬂoating and ﬁxed exchange rates. 8 One could clearly introduce ﬁner distinctions between different exchange rate regimes. For example, one could use the IMF classiﬁcation of exchange rate regimes. This classiﬁcation has been criticized, however. See Calvo and Reinhart (2000). 9 See Ball et al. (1988) and Wyplosz (2001).

References Angeloni, I., Kashyap, A., Mojon, B. and Terlizzese, D. (2003) ‘The Output Composition Puzzle: A Difference in the Monetary Transmission Mechanism in the Euro Area and the US’, European Central Bank Working Paper, 268, September. Ball, L., Mankiw, G. and Romer, D. (1988) ‘The New Keynesian Economics and the Output-Inﬂation Trade-Off’, Brookings Papers on Economic Activity, 1: 1–82. Calvo, G.A. and Reinhart, C.M. (2000) ‘Fear of Floating’, NBER Working Papers, 7993, National Bureau of Economic Research, Cambridge, MA. Christiano, L., Eichenbaum, M. and Evans, C. (2001) ‘Nominal Rigidities and the Dynamic Effects of a Shock to Monetary Policy’, Federal Reserve Bank of Cleveland Working Paper. Clarida, R., Gali, J. and Gertler, M. (1999) ‘The Science of Monetary Policy’, Journal of Economic Literature, 37 (4): 1661–1707. European Central Bank (2004) The Monetary Policy of the ECB, Frankfurt. Fischer, S. (1977) ‘Long-Term Contracts, Rational Expectations and Optimal Money Supply Rules’, Journal of Political Economy, 85: 191–206. Greene, W. (1997) Econometric Analysis, 3rd edn, Upper Saddle River, NJ: Prentice Hall. Gylfason, T. and Lyndbeck, A. (1994) ‘The Interaction of Monetary Policy and Wages’, Public Choice, 79: 33–46. Knell, M. and Stix, H. (2003) ‘How Robust Are Money Demand Estimations? A MetaAnalytic Approach’, Discussion Paper, 81, Austrian National Bank. Lucas, R. (1973) ‘Some International Evidence on Output-Inﬂation Tradeoffs’, The American Economic Review, 63 (3): 326–334. Nijkamp, P. and Poot, J. (2004) ‘Meta-Analysis of the Effect of Fiscal Policies on LongRun Growth’, European Journal of Political Economy, 20: 91–124. Rose, A. (2004) ‘The Effect of Common Currencies on International Trade: A Meta-

Monetary policy and real economy 53 Analysis’, in von Furstenberg, G.M. (ed.), Monetary Unions and Hard Pegs: Effects on Trade, Financial Development and Stability, Oxford: Oxford University Press. Smets, F. and Wouters, R. (2003) ‘An Estimated Dynamic Stochastic General Equilibrium Model of the Euro Area’, Journal of the European Economic Association, 1 (5): 1123–1175. Soskice, D. and Iversen, T. (2000) ‘The Non-Neutrality of Monetary Policy with Large Price or Wage Setters’, Quarterly Journal of Economics, 14: 110–124. Stanley, T. (2001) ‘Wheat from Chaff: Meta-Analysis as Quantitative Literature Review’, Journal of Economic Perspectives, 15 (3): 131–150. Tabellini, G. (2001) ‘Comment on “the Case for Price Stability” by Marvin Goodfriend and Robert King’, in Herrero, G., Hoogduin, M. and Winkler (eds), Why Price Stability? Proceedings of the First ECB Central Banking Conference, Frankfurt: ECB. Taylor, J. (1980) ‘Aggregate Dynamics and Staggered Contracts’, Journal of Political Economy, 88: 1–22. The Economist (2005) ‘The Great Thrift Shift. A Survey of the World Economy’, September 24, London. Walsh, C. (2003) Monetary Theory and Policy, 2nd edn, Cambridge, MA: MIT Press. Woodford, M. (2003) Interest and Prices: Foundations of a Theory of Monetary Policy, Princeton, NJ, and Oxford: Princeton University Press. Wyplosz , C. (ed.) (2001) The Impact of EMU on Europe and the Developing Countries, Oxford: Oxford University Press.

4

Germany in the monetary union Charles Wyplosz

Introduction By joining the monetary union, Germany did not just lose monetary independence like every other country. It gave up one of the world’s strongest currencies and relinquished the Bundesbank domination over monetary conditions in the European Union. Above all, this momentous step was seen a contribution to lasting peace in Europe,1 but it was also perceived in Germany as a sacriﬁce that entailed signiﬁcant risks. Six years on, public opinion polls (see Figure 4.3 below p. 60) show that the public opinion, initially skeptical, now holds benign views similar to those expressed elsewhere in the euro area. German citizens feared that price stability would be jeopardized and that the euro would not be as strong as the DM. Table 4.1 helps explain why German citizens now feel better about the euro: inﬂation has never been so low. The euro has ﬁrst depreciated vis-à-vis the dollar, but then it has recovered, and it is generally perceived as a strong currency in the face of large ﬂuctuations affecting the dollar, the yen and other major currencies. On that dimension, therefore, the record of ECB is at least as good as that of the Bundesbank. There is another side to the medal, however. This table also shows that the growth performance has deteriorated, both in the run-up to the launch of the euro and afterward. Domestic demand has now been ﬂat for several years running. The unemployment rate is high, although much of it is concentrated in the Eastern part of the country which is going through its transition from central planning. Paradoxically, therefore, the euro has delivered the opposite of what Germans feared at the outset. Table 4.1 Macroeconomic performance of Germany (average annual rate, %)

Growth Inﬂation

1963–1973

1975–1990

1992–1998

1999–2005

4.3 3.5

2.4 3.3

1.3 2.6

1.2 1.4

Source: Economic Outlook, December 2005, OECD. Note All time series face a break in 1991 as a consequence of German uniﬁcation.

Germany in the monetary union 55 This chapter examines how euro area membership has affected the macroeconomic situation in Germany. The loss of monetary independence has deeply altered the conduct of macroeconomic policies, especially since ﬁscal policy is now subject to the Stability and Growth Pact. How much of the poor economic performance of the last few years can be attributed to euro area membership? The introduction starts with monetary policy. It shows how the Eurosystem has adopted wholesale the magic Bundesbank recipe, carrying out its most and its least desirable practices. It also underlines that, so far, the common monetary policy has not been ideally adapted to German economic conditions. The introduction looks at ﬁscal policy. Like many other countries in Europe, Germany has a record of procyclical ﬁscal policy. The combination of adverse shocks and of the Stability and Growth Pact has only reinforced this tendency. With less support from monetary policy and a contractionary ﬁscal policy stance, growth has been subdued, which helps explain the low inﬂation rate. Poor macroeconomic management alone, however, does not fully explain slow growth. Structural rigidities, which have dented potential growth, require supply-side policies. The introduction examines the links between structural reforms and monetary union membership. While it is too early to reach strong conclusions, it seems safe to argue that the need for reforms does not impede better macroeconomic management. The concluding section summarizes the main points and asks, but does not seek to fully answer, a disquieting question: why has Germany long misused ﬁscal policy?

Monetary policy in Germany The Bundesbank’s magic By the mid-1980s, the Bundesbank was arguably perceived as one of the most successful central banks. Germany’s success in dealing with the oil shocks with limited inﬂation and no larger output cost than most other oil-importing countries was largely attributed to the Bundesbank’s independence and to its clearly stated priority on keeping inﬂation in check. In particular, the monetary targeting strategy has come to be seen as the lynchpin of the Bundesbank’s success. Yet, the Bundesbank’s endorsement of monetary targeting has been slow and cautious. Importantly, the Bundesbank’s focus on price stability only came about in the 1970s following the end of the Bretton Woods system. Until then, the exchange rate anchor was a central element of a strategy that was explicitly concerned with growth, employment and external balance. How then did the Bundesbank manage to establish the DM reputation as a strong currency even during the Bretton Woods era? Two, mutually supporting explanations seem to provide the answer (Richter, 1999): the bank was independent and the German public opinion was strongly in favor of price stability. The Bundesbank well knew that it could continue enjoying strong independence as long as it was delivering on price stability. Yet, in its pursuit of price stability, the Bundesbank always remained

56

C. Wyplosz

3.33

3.28

3.23

3.18

3.13

3.08

3.03 1988M1

1990M1

1992M1

1994M1

1996M1

1998M1

Figure 4.1 Bundesbank monetary targeting (source: Baltensperger (1999)).

sensitive to German public opinion. At times, this meant focusing on growth, and the Bundesbank dutifully obliged. Even during its monetary targeting years, the bank did not follow a rigid rule. This can be seen from Figure 4.1, which displays the annual target ranges and the actual evolution of the chosen monetary aggregate.2 The target was achieved only on 13 of the 24 years for which it was pre-announced – either as a single growth objective or as a target range. Importantly, the Bundesbank never attempted to make up for missed targets, in effect allowing for a drift in the rule. Truth is that the Bundesbank has always displayed considerable ﬂexibility in its strategy, as formally shown in Clarida et al. (1998). This observation stands in contrast with the currently popular view that the Bundesbank stellar reputation is based on a single-minded adherence to a rigid rule. This view has largely been created by the Bundesbank itself in the years preceding the adoption of the euro, largely as a precautionary measure as anxiety rose in Germany that the euro would be less strong that the DM. In doing so, the Bundesbank has reinforced its tendency to talk tough and act smoothly. This tendency has been made plain by the target overruns in the years that followed uniﬁcation. Figure 4.1 shows that money growth has signiﬁcantly exceeded the annual targets on every single year from 1991 to 1995, at a time when the bank constantly reiterated its determination to prevent

Germany in the monetary union 57 inﬂation from rising. Unsurprisingly, inﬂation rose to reach 5 percent in 1992, partly due to the quickly rising budget deﬁcit, partly because the Bundesbank decided not to tighten its stance, aiming instead at a gradual return of inﬂation toward a rate compatible with price stability. Yet, its public statements have allowed the Bundesbank to emerge from this difﬁcult period with its reputation untainted. Unsustainable Bundesbank dominance The Exchange Rate Mechanism (ERM) was meant to avoid exchange rate ﬂuctuations within the Common Market. Its careful design was also meant to be fully symmetric as there was no presumption that the weak-currency countries should undertake more adjustments than the strong-currency countries. Underpinned by a commitment to unlimited exchange market interventions by all member countries, the ERM was a highly cohesive effort to foster cooperation without coercion. Within a few years, it emerged that the ERM had become a ‘greater DM area’. De facto, the DM had become the reference currency (Giavazzi and Giovannini, 1992). Why it turned out this way is interesting. To start with, design symmetry did not prevent the situation from being asymmetric. The situation of weak-currency countries, when they face a hemorrhage of foreign exchange reserves, fundamentally differs from that of the strong-currency countries, which simply accumulate more reserves. In addition, realignments were decided by consensus; in practice, parity changes did not quite compensate for accumulated inﬂation differentials. As a result, weak-currency countries gradually lost competitiveness to the advantage of strong-currency countries. Furthermore, realignments were usually preceded by speculative attacks. Devaluations under pressure were politically costly, while revaluing countries were enjoying the prestige of being seen as successful. In the end, weak-currency countries decided to emulate the Bundesbank, which became the de facto central bank of all ERM countries. This is when Germany started to discover the unpleasant consequences of its monetary domination. The Bundesbank was asked to recognize its leadership position and was invited to take into account the impact of its decisions on the other countries. Tensions rose at the time of German uniﬁcation when its tight monetary policy stance collided with a downturn in most ERM member countries. The situation unavoidably led many countries to wonder why the de facto central bank of Europe should be exclusively in German hands. Suddenly, a monetary union became a viable option. With the exception of Germany, ERM member countries saw the monetary union as the way to recover some (shared) sovereignty on monetary policy. In other words, the Bundesbank successes were creating the conditions of its demise. As political pressure grew and the prospect of a monetary union rose, Germany focused on preserving the Bundesbank’s magic recipe and undertook to make the future European Central Bank a replica of the Bundesbank.

58 C. Wyplosz Design of the monetary union The introduction argues that the key success factors of the Bundesbank were its independence and deep popular support for price stability. Quite naturally, Germany insisted that these two characteristics be adopted in the monetary union. Central bank independence came naturally since it no longer was a controversial topic, although democratic accountability remains a live issue, especially in the case of the European monetary union. As for popular support for price stability, it could not be decreed. In the case of Germany, it is the result of traumatic historical events that have spared most of the other European countries. Knowing this full well, the German authorities devoted considerable efforts at enshrining into the Maastricht Treaty the need for a ‘culture of price stability’. The results of these efforts are multifaceted. At the formal level, they include the entry criteria, the hierarchy of objectives ascribed to the Eurosystem (price stability comes clearly as the top objective) and the Stability and Growth Pact. Informally, they also explain Germany’s insistence to base the ECB headquarters in Frankfurt, the choice of the ﬁrst chief economist and the sensitivity of the German media to the Eurosystem’s decisions. In many ways, these efforts have paid off. As Table 4.1 shows, inﬂation in the euro area has been lower than in Germany during the DM era. Some may see this outcome as the result of an unusual combination of events; this would be misleading. The Eurosystem’s institutional setup – considerable central bank independence and the priority given to price stability – is stronger in this respect than the Bundesbank’s, and for all we know, institutions deliver when they are well designed. Other aspects of German efforts at injecting a ‘culture of price stability’ are more questionable and can turn out to be self-defeating. The entry criteria failed to eliminate any country; they only delayed Greece entry by two years. They are now working against the new EU member countries, which differ in many important ways from the incumbents. For instance, since 1993, Estonia has adopted a currency board arrangement vis-à-vis the DM and then the euro. In effect, Estonia has long been part of the monetary union. Its inﬂation rate exceeds the acceptable limit, but it is beyond the control of the national authorities. Prices rise faster in Estonia because the country is catching up, a welcome consequence of the highly commendable rise of its standard of living. The fact that applying the criteria would keep Estonia out is proof that they are misguided. Similarly, the Stability and Growth Pact is justiﬁed by the correct observation that ﬁscal indiscipline has universally been the root cause of hyperinﬂations. The 3 percent budget deﬁcit limit, however, is on a scale of magnitude completely different from inﬂation-threatening deﬁcits. The 60 percent debt ceiling is so unrealistic that it has been ignored so far. As for diffuse fears that one country’s ﬁscal stress could affect the whole euro area, they fail to take into account the very explicit and unambiguous no-bailout clause included in the treaty. Advocates of the Stability and Growth Pact argue that prudence is a desirable feature,

Germany in the monetary union 59 but they fail to recognize that prudence can be very costly. The events of 2003 have amply demonstrated that the cost-beneﬁt balance was indefensible. The result of this lack of realism and formal rigidity has led to a modiﬁed Stability and Growth Pact that is likely to become powerless, as creative accounting and numerous exemptions make it difﬁcult to implement. The monetary union years As the biggest country, the home of the Bundesbank and now of the ECB as well, Germany could be expected to play an important role in the euro area. As the country that has made the most tangible sovereignty concession since the DM was previously acting as the de facto leading currency, one would expect German public opinion to be cool to the euro. These expectations are not borne out. With a GDP that represents 30 percent of the euro area GDP, Germany carries an important weight in the overall inﬂation index, but not an overwhelming one. Since the Eurosystem explicitly focuses its analysis on euro area data, ignoring country level information, Germany matters, but the situation is a far cry from earlier times when monetary policy in Europe was set by the Bundesbank, which explicitly ignored economic conditions outside of Germany. The unavoidable ‘one-size-ﬁts-all’ approach to monetary policy of the Eurosystem implies that the monetary stance is not always well adapted to any particular country’s economic conditions, and this applies to Germany. Figure 4.2 indeed shows that the long-term interest rate has been higher in Germany, where inﬂation has been lower, than in the euro area as a whole, even though the German output gap has been more negative, especially since 2001. Thus, the common monetary policy has been too tight. This is certainly not the situation most feared ex ante. As noted below (pp. 64–66) ‘Fiscal policy in Germany, the monetary union years’, this may have led the government to conduct a more expansionary ﬁscal policy that it would have wanted otherwise.

Long-term real interest rate

Output gap

4.00

2

3.50

1

3.00

0

2.50

1

2

2.00 Germany 1.50

Euroarea

1999 2000 2001 2002 2003 2004 2005

Germany 3

1999

2000

2001

Euroarea 2002

2003

2004

Figure 4.2 Monetary conditions 1999–2005 (source: OCDE, December 2005).

2005

60

C. Wyplosz

Indeed, many German observers have complained about both an excessively tight monetary policy and an excessively soft ﬁscal policy. Whether they are right or not, these are normal consequences of monetary union membership. The combination of the loss of the DM and of imperfectly adapted macroeconomic policies could have provoked a popular backlash against euro area membership. This has not been the case. Initially relatively warm to the monetary union,3 German public opinion then became noticeably cooler than the European average until the introduction of banknotes in 2002, as shown in Figure 4.3. Since then, there has been no signiﬁcant difference. In this sense, there is no more any ‘German problem’ with the monetary union. Part of the credit for soothing German anxieties undoubtedly goes to the ECB, which has been particularly attached to portray itself as a continuation of the Bundesbank. It did so in a number of ways. First, the ECB has adopted a monetary policy strategy that was strongly reminding of the Bundesbank’s monetary targeting rule. It adopted M3 growth as its ﬁrst pillar, publishing a ‘reference value’ that resembled in all but name a target. Second, it pursued the practice of talking soft and acting ﬂexibly, even if that meant consistently overshooting the money growth objective. While this approach did not play well with some public opinions attached to policy transparency, it was obviously appreciated in Germany. To this day, when criticized for the gap between words and deeds, the ECB frequently emphasizes the importance of reassuring German public opinion. Third, it adopted a very low inﬂation objective of ‘close to but less than 2%’. Even though the objective has been consistently missed at the whole euro area level, Table 4.1 shows that it has been achieved in Germany.

80 Germany

EU15

70

60

50

40

30

20

10

0

1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2002 2001 2003 2004 2005

Figure 4.3 Support for the Euro (source: Eurobarometer, various issues). Note: Percent responding ‘for the single currency’. Polls conducted in early part of the year, reported in the June or July issue. 1991–1995: EU 12 instead of EU15.

Germany in the monetary union 61 Finally, the ECB has been successful in convincing the German public opinion that the low growth and high unemployment performance of their country since the adoption of the euro is the responsibility of the federal government. In doing so, it successfully adopted the Bundesbank’s post-1970s tradition of only taking responsibility for inﬂation. It remains that the poor economic performance of Germany since 1999 is striking. A vast literature has documented the importance of numerous rigidities in the goods and labor markets, in Germany and in the other large monetary union member countries.4 Some reforms have been undertaken since the early 2000s, but they have been limited and their effects are only starting to be seen. A key question is whether reforms alone can bring back normal growth and lower unemployment and, if not, what the effect of monetary union is. ‘The monetary union years’ (pp. 64–66) argues that ﬁscal policy, under pressure from the Stability and Growth Pact, has failed to support activity. ‘Monetary union and structural reforms’ (pp. 66–67) examines the link between structural reforms and monetary union membership.

Fiscal policy in Germany Principles Having given up monetary policy independence, all euro area member countries are left with only one macroeconomic stabilization tool, ﬁscal policy. As has been noted for a very long time, the importance of ﬁscal policy rises considerably in a monetary union.5 Unfortunately, ﬁscal policy is a considerably more complex and less ﬂexible instrument than monetary policy for reasons that can be brieﬂy recalled. To start with, theory may cast doubts on the ability of ﬁscal policy to have any effect or even the desired effect. A good starting point is the principle of Ricardian equivalence. This principle asserts that any ﬁscal policy action is completely offset by private sector reactions in the opposite direction. For example, creating a budget deﬁcit by raising public spending to increase domestic demand is met by more saving and therefore less consumption on the part of households. The reason is simple: as they recognize that the resulting debt increase will have to be paid for later on through higher taxes, households save to face the future tax burden.6 In short, the government budget is a veil entirely pierced by clearsighted households. Of course, there are dozens of theoretical reason why households cannot pierce through the government veil but what is the situation in the real world? A large number of studies have explored whether Ricardian equivalence is met in practice, and the result is a resounding no.7 So we know that ﬁscal policy works under normal conditions. We return to this issue in ‘The monetary union years’ (pp. 64–66) which discusses non-Keynesian effects. Yet, it is not easy to see it working. If left untouched, ﬁscal policy is spontaneously doing its stabilizing job. This is the story of the automatic stabilizers, the fact that during a recession the budget surplus shrinks, or the deﬁcit deepens,

62

C. Wyplosz

mainly because tax payments decline, which means that ﬁscal policy becomes expansionary; the opposite occurs during a boom. Two aspects of the automatic stabilizers need to be highlighted. First, for this effect to occur, the government must remain passive so as not to frustrate the impact of the economic cycle on the budget outcome. Second, even if they are allowed to develop completely, the automatic stabilizers only partially cushion the cyclical movement, they cannot offset it completely. This means that if ﬁscal policy is to be used to replace monetary policy as a counter-cyclical tool, the automatic stabilizers are not enough, some additional discretionary action is called for. In a recession, the surplus must shrink further, or the deﬁcit must be allowed to deepen; in a boom, the surplus should increase, or the deﬁcit must be reduced. Discretionary policy, as this part of ﬁscal policy is called, must be used in addition to the automatic stabilizers. This crucial distinction is often lost in policy debates. A seriously complicating factor is the limited ﬂexibility of ﬁscal policy. Budgets are voted by parliaments and cannot be changed overnight. Several countries, including Germany, have some built-in ﬂexibility, but it is a limited one. If more needs to be done, the parliament must give its approval. This takes time – in comparison central banks can act any time they wish – and triggers political debates that may further lengthen the reaction time. It is feared, therefore, that by the time a decision is taken and implemented, it is not needed any more. Worse, it can actually come so late that its effects are the opposite of those intended and needed. Germany’s experience The existence of automatic stabilizers implies that the budget outcome in any year is partly determined by the conjuncture. In a boom year, for instance, tax revenues increase and the budget improves. This is why we need to compute cyclically adjusted budgets. The cyclically adjusted budget is an estimation of what the budget outcome would be if the economy were operating at its potential level. A further adjustment is needed to recognize that part of public spending is dedicated to the debt service, which is a contractual obligation of the government. Subtracting interest payments on the debt from the cyclically adjusted budget outcome provides the cyclically adjusted primary budget balance, a measure that is a good estimate of the government’s choices. An increase in the cyclically adjusted primary budget balance, for instance, means that the government has decided to reduce spending, cut taxes, or both; this is an indication that discretionary ﬁscal policy has been contractionary. Figure 4.4 displays the change in Germany’s cyclically adjusted primary budget balance between 1967 and 2005.8 An increase in this measure reveals contractionary ﬁscal policy actions – on top of the automatic stabilizers – while a decrease corresponds to an expansionary stance. This indicator is shown along with the output gap, which measures the cyclical situation of the economy. If the government uses ﬁscal appropriately, in a contractionary way during booms and

Germany in the monetary union 63 5 4 3 2 1 0 1 2 3 4 5 1967

1971

1975

Output gap

1979

1983

1987

1991

1995

1999

2003

Change in cyclically adjust primary budget balance

Figure 4.4 Cyclical behavior of German ﬁscal policy (1967–2005) (source: Economic Outlook, December 2005, OECD). Notes: The output gap measures the percentage difference between actual and potential real GDP. The change in the cyclically adjusted balance is the difference of the year’s cyclically adjusted balance, measured as a percentage of GDP, and its level the previous year. See the text for more details.

in an expansionary way during slowdowns, the two curves move in the same direction. The ﬁgure shows the opposite pattern: discretionary ﬁscal policy has been mostly procyclical in Germany, actually accentuating the economy’s cycles, and this situation has gotten worse after 1991.9 Germany is not alone in having conducted discretionary procyclical ﬁscal policies in the past. A likely reason is the misperception, common among governments, that the legitimate concern with public debt slippages requires preventing from quickly deepening. Since the automatic stabilizers are a source of deﬁcit deepening during cyclical slowdowns, there results a tendency to thwart their working. This is misguided since on average the automatic stabilizers are neutral as far as the debt is concerned. If the cyclically adjusted budget is balanced, deﬁcits in bad years are compensated by surpluses in good year leaving the debt unchanged over a completed cycle. If, however, the cyclically adjusted budget is in deﬁcit, the debt is indeed growing over the cycle; in this case, however, the problem lies with a cyclically adjusted deﬁcit, not with the actual deﬁcit deepening in bad years. Attempts at reducing the deﬁcit during a downswing are not just wrong, they are self-defeating because they further depress

64 C. Wyplosz growth, which means even bigger deﬁcits. The time to deal with a cyclically adjusted deﬁcit is when the economy is solidly growing, not when it is ﬂat. Germany seems to have fallen victim to this misguided policy stance. Figure 4.4 has shown that after 1991 ﬁscal policy has been systematically tightened in bad years and relaxed in good years, just the opposite of good practice. Figure 4.5 further shows that these procyclical policies have been futile as far as containing deﬁcits is concerned. This ﬁgure displays the actual budget balance along with the output gap. Had the German authorities been successful at blocking the automatic stabilizers, the actual budget balance would have remained unaffected by the cycles. Figure 4.5 shows that this has not been the case. The fact that the actual budget balance moves in parallel with the output gap implies that the procyclical use of discretionary policy has not been strong enough to fully offset the automatic stabilizers as far as containing the deﬁcit is concerned; the fact that cyclically adjusted primary balance has been improved in bad years mean that ﬁscal policy discretion has prevented the full use of the stabilizers. The monetary union years The track record of German ﬁscal policy since the adoption of the euro can be interpreted in two ways. One view is that, except for the brief relief from the 1999–2000 world-wide expansion, Germany has struggled to contain the budget deﬁcit within the limits imposed by the Stability and Growth Pact. As is well

4

2

0

2

4

6 1967

1971

1975

1979

1983

Output gap

1987

1991

1995

1999

2003

Budget balance

Figure 4.5 Cyclical behavior of German budget balances (1967–2005) (source: Economic Outlook, December 2005, OECD).

Germany in the monetary union 65 known, this effort has not been successful, and Germany only escaped sanctions when it succeeded in putting the pact in abeyance in November 2003. It then supported the watering down of new version adopted in 2005. Yet, as Figure 4.4 shows, it has conducted a contractionary ﬁscal policy during the long slowdown that started in 2001 and seems to be coming to an end at the time of writing (early 2006). The other view is that ﬁscal policy is a poor instrument. It is slow to put into action, it generates political pressure as interest groups see an opportunity to shift the use of public ﬁnances to their narrow beneﬁt and its overall effect on the economy is both limited and uncertain. Clearly, support for activist ﬁscal policy has waned in the face of considerable theory and empirical results. This has led some to argue that ﬁscal policy should be limited to the automatic stabilizers. This view has proven to be especially popular in Germany and is clearly stated by Issing (2005): An important stabilizing function of ﬁscal policy operates through the socalled ‘automatic ﬁscal stabilizers’. [. . .] In principle, stabilization can also result from discretionary ﬁscal policy-making, whereby governments actively decide to adjust spending or taxes in response to changes in economic activity. I shall argue, however, that discretionary ﬁscal policies are not normally suitable for demand management. As noted above, Germany has not just refrained from discretionary countercyclical ﬁscal policies, it has attempted to tighten up its budget deﬁcit during slowdowns. This is in line with the view that ﬁscal policy can have nonKeynesian features, namely that restrictive policy can have an expectation impact, and conversely. This view has been widely promoted as a justiﬁcation for procyclical policies, especially during downturns when the deﬁcit increases. It enjoys some currency in Germany and has been endorsed by Issing (2005): The experience of the industrialized countries in recent decades clearly shows that persistent ﬁscal imbalances limit the room for ﬁscal policy to stabilize the economy. Imbalances often necessitate tight ﬁscal policies during downturns to prevent unsustainable deﬁcits and debt developments. [. . .] Consolidation measures may then re-establish conﬁdence and improve expectations about the long-term outlook of public ﬁnances. These so-called ‘non-Keynesian’ effects may have the result that ﬁscal consolidation even has an expansionary impact on the economy. According to the ﬁrst view, the procyclical discretionary use of ﬁscal policy since 1999 has contributed, partly at least, to the unusual length of the 2001–2006 slowdown and to the attending continuously high unemployment rate. According to the second view, debt stabilization efforts, if pursued more forcefully, could have enhanced growth and may have limited the slowdown. Much research efforts will be needed to reach a verdict, and it is likely that

66

C. Wyplosz

70

68

66

64

62

60

58

1998

1999

2000

2001

2002

2003

2004

2005

2006

Figure 4.6 German public debt (Maastricht deﬁnition, % of GDP, 1998–2006) (source: OECD, December 2005).

controversies will not be resolved. Yet, the literature on non-Keynesian effects of ﬁscal policy has constantly indicated that these effects only apply in the presence of large ﬁscal actions (see the recent reappraisal in Giavazzi et al., 2005). In addition, non-Keynesian effects are expected to arise when, for example, a ﬁscal policy contraction is perceived to succeed in stabilizing or reducing the public debt. As Figure 4.6 readily conﬁrms, the public debt has increased during the slowdown years when discretionary ﬁscal policy has become contractionary. Monetary union and structural reforms Germany’s poor economic performance over the last six years is the result of a much reduced growth potential and of an increasingly negative output gap. Potential growth, estimated by the OECD at 3.1 percent annually in the 1970s and 2.2 percent in the 1980s, has fallen to 1.3 percent since 1995. In addition, as Figure 4.4 shows, the output gap has declined since the adoption of the euro and has even remained negative for several years running. Thus, demand has been subdued and did not even keep up with a slowly rising supply. Both monetary and ﬁscal policies have been insufﬁciently supportive. Figure 4.7 illustrates the relative contributions of the demand and supply sides. ‘Full employment GDP’ corresponds to a zero output gap; the difference between this path and the actual path shows the amount of loss output due to insufﬁcient demand. The highest path also assumes that the output gap is elimi-

Germany in the monetary union 67 120

110

100 Actual Higher growth potential and full employment Potential 90

1998

1999

2000

2001

2002

2003

2004

2005

Figure 4.7 Actual and simulated real GDP (Actual GDP in 1998 = 100) (source: author’s calculations).

nated, but, in addition, it allows for a trend growth rate of 2.2 percent instead of 1.3 percent. The difference with the previous path is illustrative of the gains to be reaped from supply-side policies, assuming that the policies succeed in raising the trend growth rate as postulated.10 Does monetary union membership affect the incentives to carry out reforms? There is by now a sizeable literature on the relationship between monetary union membership and supply-side policies; a good summary is in Duval and Elmeskov (2005). In brief, monetary union may encourage reforms for four main reasons. The ﬁrst reason follows from the fact that labor mobility – a key criterion of the optimum currency area theory – is low in Europe. The loss of the exchange rate instrument implies that adverse shocks must be absorbed internally, which is easier the less rigid are domestic economic institutions and practices. Second, the single currency is expected to increase competition within the euro area. A sound supply side becomes a key success factor. Third, as competition strengthens, rents shrink which should reduce opposition by interest groups. Finally, increased capital mobility implies that each country competes with other euro area members to attract foreign investments. Since ﬁrms are attracted by a friendly supply side, the incentive to reform is enhanced. Four other reasons have been advanced in the opposite direction. The ﬁrst one notes that additional demand for a higher supply requires lower relative

68 C. Wyplosz prices. In the absence of the exchange rate instrument, this requires lower absolute prices, which usually requires the pressure of a lasting period of slow growth. Second, the constraint imposed by the Stability and Growth Pact may deprive governments of the ‘oil’ that they may need to inject to compensate those that stand to lose from supply-side reforms. Third, one beneﬁt from supply-side reforms is to improve the inﬂation-output tradeoff. In the absence of a national monetary policy, this beneﬁt cannot be exploited unless countries in the euro area achieve the same progress on their supply sides. The last disincentive toward structural reform comes from the ﬁnancial side. Successful reforms are normally expected to lower the long-term risk premium because the economy becomes more resilient. This effect is lost inasmuch as long-term interest rates have converged in the euro area. We would need some empirical evidence to assess these opposing views. Unfortunately, structural reforms are low frequency events, so the six years of existence of the euro are far too short to offer a testing ground. At this stage, it may be prudent to assume that the pro and contra reform effects of the monetary union are balanced. It is also worth noting that there are very few cases of supply-side policies where externalities are signiﬁcant (Tabellini and Wyplosz, 2006). In the end, therefore, structural reforms are very much a national undertaking; they should be pursued on their own merits subject to local political constraints. Like demand-side policies, external conditions cannot be invoked as an explanation why too little is being done in the realm of supply-side policies. Should demand management support supply-side reforms? Another important question is whether demand-side policies should support supply-side policies. It is often argued that it makes little sense to raise the supply potential while muzzling demand. The argument is not very powerful, though. To start with, successful supply-side policies are likely to have favorable demand-side effects, for example as productivity and wealth rise. If the counter-argument is that the timing may be wrong, i.e. that demand will only grow with a lag, then one must recognize that supply-side effects appear with a considerable and variable delay. Contemporaneous demand-side policies are bound to kick in long before the potential rises, with an inﬂationary impact. Another argument is that it is easier to undertake painful reforms when the economy is going strong and potential losers do not feel at risk. This is a plausible view. Equally plausible is the opposite view that governments are unlikely to implement unpopular reforms and court political trouble unless they are pretty desperate. This view further argues that pressure groups that stand to lose from supply-side policies are only willing to make concessions when the consequences of status quo become very painful and very visible. In order to choose between these two views, we need to turn to the economic evidence. Recent work by Boeri (2004) and Duval and Elmeskov (2005) indicate that successful supply-side policies tend to be undertaken when growth is low, unemployment is high and the budget is not in serious deﬁcit.

Germany in the monetary union 69 Obviously, this does not mean either that governments should purposefully pursue contractionary policies in order to generate enough despair to then force through unpopular reforms. No democratically elected government is likely to adopt such a devious strategy. The most reasonable conclusion is that demandand supply-side policies ought to be kept separate. Good governments endeavor to keep actual GDP as close as possible to its potential level and to raise the potential level.

Conclusions Germany has sometimes been portrayed as the sick man of Europe. This kind of statement is bound to be too blunt to be entirely correct, but it captures the poor economic performance over the last decade. The slow growth record reﬂects a long-run phenomenon that calls for supply-side policies. It also reﬂects a poor demand management record. The primary aim of this chapter is to examine what effects the monetary union has had on the German economy. Assuming that the monetary union had limited effects on the supply side, this chapter focuses on the demand side. Simplifying somewhat, this chapter argues that the main change is the following: prior to monetary union, the Bundesbank used to conduct a fairly effective counter-cyclical monetary policy while the federal government mainly conducted a procyclical ﬁscal policy. Since the adoption of the euro, the ECB is structurally less sensitive to German conditions while the federal government continues to pursue a procyclical stance. The quality of demand management has deteriorated, precisely as potential growth slowed down. Structural reforms are not Pareto-improving, they raise overall welfare but leave losers behind. This may be why it appears that successful reforms are usually undertaken during difﬁcult periods, when governments feel that they need to take drastic actions and when pressure groups are more willing to abandon some of their rents. The poor economic performance of the last years may thus have a silver lining. Some limited reforms have been enacted, more are needed. Yet, this is no reason for continuing to carry out poor demand-side management. The federal government ought to carry a two-handed strategy, strengthening the supply side while closing the gap between actual and potential GDP. At any rate, there is no indication that poor demand-side management has been a conscious strategy to soften resistance to reform. Why has demand management been so poor remains an important and complex issue. A few hypotheses can be outlined. It can be that Germany was essentially preoccupied with the need to promote within the euro area a culture of price stability. This would have called for encouraging restraint in the Eurosystem and carrying out restrictive ﬁscal policies. There is little support for this hypothesis. The German authorities have not been seen as calling upon the Eurosystem to be more restrictive. Faced with a downturn, they have let the budget deﬁcit deepen to the point where they faced

70 C. Wyplosz the wrath of the European Commission for repeatedly violating the Stability and Growth Pact. It can be argued that, absent the pact, they might have adopted a counter-cyclical stance, but there is no evidence from past history to support this conjecture. It can be that the monetary union membership worsens the quality of demand management. Undoubtedly, the loss of domestic monetary policy affects the most versatile demand-side policy instrument. The Eurosystem cannot be faulted for having conducted an excessively tight monetary policy, but the common policy unavoidably implies that real interest rates are higher where inﬂation is lower. This well-anticipated effect calls for counter-cyclical ﬁscal policies, but the Stability and Growth Pact can stand in the way if the initial budget is in deﬁcit. This is precisely what happened to Germany, and the belated recognition of this design mistake has led to a modiﬁed pact. The new pact is very much open to interpretation, so it is impossible to predict how it will be used in practice. It can also be that the preoccupation with price stability of the largest shareholder of the ECB has weighed on the Eurosystem’s policy strategy. The Eurosystem has upheld the Bundesbank’s tradition of saying one thing and doing another one. Its inﬂation objective – ofﬁcially called its deﬁnition of price stability – is very low by international standards. One consequence has been lower inﬂation in Germany than at any time over the previous three decades. A related consequence is that the Eurosystem has been more careful in its moves than the other major central banks. In view of its size and its historical role, Germany might consider arguing for a less restrictive inﬂation objective. Finally, it can be that the view prevailing in German policy-making circles is deeply wed to a high degree of prudence in demand management. There is some evidence that this is the case. Debates often pit ‘monetarists’ against ‘Keynesians’, with the former exercising strong inﬂuence while the latter sometimes develop outdated views. Modern macroeconomic thinking has largely escaped this ideological dichotomy.11 Today’s well-managed countries are run by pragmatist economists.

Notes 1 It was also a political deal, a pre-condition set by French President Mitterand for his support of uniﬁcation. 2 In 1988, the bank shifted from central bank money to M3. 3 Chancellor Kohl actively presented the Maastricht Treaty as a step toward a political union, not just a monetary union. It progressively surfaced that the single currency would be the main practical consequence of the treaty. 4 Two excellent recent surveys are Nickell et al. (2005) and Blanchard (2006). 5 This remark is developed in Melitz (2000) and Wyplosz (1991). 6 An even worse case occurs when household worry that the deﬁcit will continue and save more than the government spends. In that case, an expansionary ﬁscal policy ends up being contractionary. This situation has been found to occur in a few exceptional cases. 7 A good overview of this literature is Gruen (1997).

Germany in the monetary union 71 8 All the budget ﬁgures concern the general government, i.e. the consolidated budgets of the federal government and of the Laender governments. 9 A formal statistical conﬁrmation is provided by Gali and Perotti (2003). A superﬁcial observation is that the correlation between the cyclically adjusted primary budget and the open gap, which should be positive, was – 0.25 over 1967–1991 and – 0.61 over 1992–2005. 10 In addition, supply-side policy effects are very slow to materialize, an issue that is ignored here. 11 See e.g. Blanchard (2000).

References Baltensperger, Ernst (1999) ‘Monetary Policy Under Conditions of Increasing Integration (1979–96)’, in: Deutsche Bundesbank (ed.) Fifty Years of the Deutsch Mark, Oxford University Press. Blanchard, Olivier (2000) ‘What Do We Know About Macroeconomics that Fisher and Wicksell Did Not?’ Quarterly Journal of Economics 463: 1375–1409. Blanchard, Olivier (2006) ‘European Unemployment: The Evolution of Facts and Ideas’, Economic Policy 45: 5–59. Boeri, Tito (2004) ‘Reforming Labour and Product Markets: Some Lessons from Two Decades of Experiments in Europe’, background paper for the IMF World Economic, IGIER, Milan. Clarida, Richard, Jordi Gali and Mark Gertler (1998) ‘Monetary Policy Rules in Practice: Some International Evidence’, European Economic Review 42: 1033–1067. Duval, Romain and Jørgen Elmeskov (2005) ‘The Effects of EMU on Structural Reforms in Labour and Product Markets’, OECD Working Paper. Gali, Jordí and Roberto Perotti (2003) ‘Fiscal Policy and Monetary Integration in Europe’, Economic Policy 37: 533–572. Giavazzi, Francesco and Alberto Giovannini (1992) Limiting Exchange Rate Variability: The European Monetary System, MIT Press. Giavazzi, Francesco, Tullio Jappelli, Marco Pagano and Marina Benedetti (2005) ‘Searching for Non-Monotonic Effects of Fiscal Policy: New Evidence’, unpublished paper, IGIER. Gruen, David (1997) ‘Ignorance and Ricardian Equivalence’, Economic Record 73: 35–44. Issing, Otmar (2005) ‘The Role of Fiscal and Monetary Policies in the Stabilization of the Economic Cycle’, speech at the international conference ‘Stability and Economic Growth: The Role of the Central Bank’, Mexico City. Online. Available at: www.ecb.int/press/key/date/2005/html/sp051114.en.html. Melitz, Jacques (2000) ‘Some Cross-Country Evidence About Fiscal Policy Behaviour and Consequences for EMU’, European Economy 2: 3–21. Nickell, Steve, Luca Nunziatta and Wolfgang Ochel (2005) ‘Unemployment in the OECD Since the 1960s. What Do We know?’ The Economic Journal 115: 1–27. Richter, Rudolf (1999) ‘German Monetary Policy as Reﬂected in the Academic Debate’, in: Deutsche Bundesbank (ed.) Fifty Years of the Deutsch Mark, Oxford University Press. Tabellini, Guido and Charles Wyplosz (2006) ‘Supply-Side Reforms in Europe: Can the Lisbon Strategy be Repaired?’ Swedish Economic Policy Review 13(1): 101–156. Wyplosz, Charles (1991) ‘Monetary Union and Fiscal Policy Discipline’, European Economy Special Edition No 1.

5

Reforms, macroeconomic policy and economic performance in Germany1 Wendy Carlin and David Soskice

Introduction This chapter maps out a set of broad hypotheses for understanding what are prima facie contradictory trends in the German economy over the last decade or so. By international standards, West Germany was a low unemployment country although there was an upward trend during the 1970s and 1980s. During the 1990s and 2000s, Germany implemented prescribed reforms (e.g. in line with OECD recommendations) to make its labour (and product) markets more ﬂexible and, according to OECD and many academic studies, would have expected to experience a fall in unemployment: yet it continued to rise. Moreover, we might have expected that low-skill service sector markets would be the beneﬁciaries of reform. While a full analysis of the data is needed before deﬁnitive conclusions can be drawn, it does not seem as though there has been a substantial expansion of employment in low-skill service sector markets. Unemployment is still low-skill intensive, and there seems to have been an increase in wage inequality and a rise in poverty.2 Worryingly, secondary education in Germany is no longer (if the PISA studies are accurate) as excellent as comparative studies in the 1980s showed it to have been, and there are also suggestions that the vocational training system no longer provides apprenticeships for many low-attaining children. These problems do not, by contrast, appear to have affected the high-valueadded export sectors of the German economy. On the contrary, in terms of world market shares of exports less imports of goods and services between 1995 and 2005,3 Germany outperformed all other G7 economies. The improvement by 5.3 per cent in their share of world exports less imports by the non-OECD countries was matched by an equal decline in the OECD’s share. The US share fell by 4.1 per cent, as did that of the United Kingdom, France, Italy and Japan by between 0.4 and 0.6 per cent. Germany’s share rose by 1.1 per cent. In terms of exports, Germany’s share of the world market for exports of goods and services was 8.8 per cent as compared with the US share of 10.3 per cent in 2005. This success is conﬁrmed by studies of comparative innovation performance, where the 2005 European Innovation Scoreboard (European Commission) put Germany in the leading group of European countries with Denmark, Sweden, Finland and

Reforms in Germany 73 Switzerland and ahead of both the United Kingdom and Ireland.4 What is perhaps surprising is that it is in these export sectors that the German model of employee representation and protection, the apprenticeship system and collective bargaining work most strongly. This German model has been reformed signiﬁcantly since 1990, largely by direct negotiation between business associations and unions at sectoral level and/or management and works council at company level – reforms in which the government has played little part. Perhaps the most notable change has been the extent to which works council chairs (elected by employees) have been brought into de facto co-management of large companies (e.g. Höpner, 2003). But these reforms have not been in the direction of ﬂexible labour markets as usually understood in the contemporary policy debate. Our ﬁrst hypothesis for explaining these contradictory patterns is that a major underlying cause of the high unemployment is an aggregate demand insufﬁciency. But here, there is a problem. While we subscribe to what is often now referred to as New Keynesian macroeconomics, its standard version offers little help in explaining why demand shocks are persistent. Section ‘Reinstating a role for aggregate demand in analysing German economic performance’ hypothesizes that there are ﬁve reasons why the standard New Keynesian model does not explain well what has happened in Germany in the last 15 years or so. Germany suffered a series of adverse shocks including monetary retrenchment in 1993–4, fulﬁlment of the Maastricht conditions in 1994–9 and the oil price increase and external deﬂationary pressures of 2001–3, and we identify a number of mechanisms that prevented stabilization from occurring in the way the New Keynesian model assumes it does. These ‘non-dampeners’ allowed the shocks to persist. Moreover, these mechanisms do not apply to the United Kingdom or the United States. They are: i the absence of a national central bank; ii the absence of a discretionary ﬁscal policy to cover situations in which monetary policy does not work; We argue that the nature of German industrial relations and coalition government structures make a discretionary policy difﬁcult to adopt. iii the existence in an open economy of multiple unemployment equilibria, where the equilibrium is determined by aggregate demand in the medium run; Thus, given (i)–(iii), adverse demand shocks unconstrained by dampening ﬁscal or monetary policy translate into higher equilibrium unemployment.5 This is ampliﬁed by two further factors that reﬂect the heterogeneity across countries in the nature of labour markets. iv the major unions in Germany as in other northern European economies are in the export sectors; Their response to uncertainty and export market weakness is wage restraint, especially given unresponsive monetary and ﬁscal policy. Unlike the case in more open small economies like Sweden, wage restraint in Germany holds

74

W. Carlin and D. Soskice

back domestic consumption counteracting the employment boost from the balance of trade. v the German workforce, with strong speciﬁc skills, is risk averse and attaches high importance to a strong welfare state in a period of rising unemployment. When it is under threat as in Germany over the past decade, the reaction of those in employment is to build up precautionary savings, thus amplifying the slow down in aggregate demand growth. Our ﬁrst hypothesis is therefore that aggregate demand insufﬁciency has been the underlying cause of persistent unemployment. Our second hypothesis relates to the implications of this for labour market ﬂexibility. As we have noted, the successful areas of the German economy are successful because they do not have ﬂexible labour markets in the conventional sense. We hypothesize that the internal political pressure to make a range of service sector labour markets more ﬂexible, in part by welfare state cuts, came from the rise in unemployment (a consequence of insufﬁcient aggregate demand) and the need to cut government expenditure because of the exigencies of the Maastricht Treaty and the Stability and Growth Pact. Almost inevitably, the rise in unemployment fell on lessskilled workers. We want to make four points: i Flexible labour markets do not magically create employment by themselves: they require an adequate level of aggregate demand, which in the canonical New Keynesian model, is ensured by active monetary policy. Legislation that creates ﬂexible labour markets in the context of inadequate demand leads to limited increases in employment and falling wages (as suggested, for example, by research on minimum wages in the United States). ii Thus, the danger of introducing ﬂexible labour market legislation in Germany is that income distribution becomes more inegalitarian and poverty increases. These effects are then ampliﬁed by welfare state cutbacks, which fall primarily on those with low skills and low incomes so that redistribution becomes more limited. iii The worsening income distribution and the rise in poverty may also lead to a worsening of educational outcomes. iv The apprenticeship system is put under severe pressure at its lower end as a result of: a the worsening of educational outcomes, imposing de facto higher training costs on employers; b labour market ﬂexibility, which means that employers can hire semiskilled labour more cheaply and hence not beneﬁt from using apprentices to do some of the semi-skilled work; c lack of aggregate demand, which implies a reduction in the projected size of skilled workforces in training companies. Our third hypothesis is that the failure to make it easy for women to re-enter the

Reforms in Germany 75 labour market, after a period of bringing up children, at the same wage and conditions as before reinforces and is reinforced by the ﬁrst two hypotheses. They are: i A system in which women can enter and re-enter public sector employment easily acts as a potentially strong additional insurance mechanism in an economy in which male workers with speciﬁc skills are risk-averse. ii It acts to prevent poverty especially for single mothers. iii It would give a powerful incentive to parents to increase the reproduction rate and hence to reduce the long-term pressure on the welfare state. In Section ‘Reinstating a role for aggregate demand in analysing German economic performance’, we set out the economic case for reinstating a leading role for aggregate demand alongside supply-side factors in the analysis of German economic performance. In Section ‘The United Kingdom and Germany: comparative performance, reforms and policies in the 1990s and 2000s’, we present a number of pieces of evidence that we hope will stimulate others to investigate our hypotheses in more depth. We concentrate on a comparison between the United Kingdom, where liberalizing economic reforms are widely believed to lie behind economic success, and Germany, where failure to reform is often blamed for its dismal economic performance. The role of positive aggregate demand shifts in the United Kingdom and negative ones in Germany is documented as candidates for explaining changes in unemployment. We investigate the extent to which Germany has implemented labour market reforms and the evidence that such reforms should have reduced unemployment. And we ask whether such reforms could themselves have depressed aggregate demand.

Reinstating a role for aggregate demand in analysing German economic performance This section is designed to show the potential importance of aggregate demand in explaining the persistent unemployment in the German economy. In addition, we aim to explain why there are major institutional barriers, themselves reﬂecting underlying power structures, which make the weak demand problem so difﬁcult to solve. In addition, given the speciﬁcity of skills in the workforce and the consequent fear of unemployment, we suggest why welfare state reform may have lead to a build up of precautionary savings and hence ampliﬁcation of the demand slowdown. While we hope this section may clarify how adverse demand shocks have at least in part lead to unemployment persistence (for nonhysteresis reasons), we also argue that institutional infeasibility rather than technical incompetence has ruled out simple demand-management solutions. The ﬁrst sub-section explains how in an open economy, a New Keynesian macro-model can imply multiple unemployment equilibria, in which the equilibrium is chosen by the level of aggregate demand.6 Hence, adverse demand shocks may lead to higher equilibrium unemployment. This is followed by an

76 W. Carlin and D. Soskice examination of potentially counteracting policy responses – monetary, ﬁscal and incomes policy. We show why each in turn is likely to be inadequate or even counterproductive. In the ﬁnal sub-section, we discuss why welfare state reforms may exacerbate the problem of depressed demand by stimulating household savings in economies in which speciﬁc skills are very important. Multiple unemployment equilibria and aggregate demand in open economies Much of the work applying New Keynesian economics to the issue of aggregate unemployment has focused on the closed economy. It is useful to start this section by setting out, using New Keynesian assumptions of imperfect product and labour markets, how the equilibrium rate of unemployment is derived in the closed economy and why it is unique. (The same is of course true of the New Classical model, but there union bargaining and wage restraint can have no role.) This will enable us to contrast the analysis for the open economy in which, using analogous assumptions, there may be multiple unemployment equilibria. In the New Keynesian model, equilibrium unemployment depends on the implications for real wages which come from wage-setting on the one hand and price-setting on the other. Both wage-setting and price-setting are nominal operations, but the goal in the ﬁrst is some level of the real wage and in the second of real proﬁts. Wage-setting can be carried out by employers alone (‘efﬁciency wages’) or as a result of bargaining. Here, we focus on wage bargaining: unions bargain for an expected real wage. The lower the level of aggregate unemployment, the higher the bargained expected real wage, wWS. But real wages are also set implicitly as a result of price-setting. Taking the simplest case of unit labour productivity, nominal unit cost is simply the money wage W. And suppose product markets enable businesses to set a markup of 1 + μ on unit costs (where the markup covers ﬁxed costs as well as a monopoly element), so that the price level is P = (1 + μ)W, which implies a (price-setting) real wage of wPS = W/P = 1/(1 + μ). Equilibrium is deﬁned as a constant rate of inﬂation. In equilibrium, the bargained real wage wWS must equal the price-setting real wage wPS. This because if, say wWS > w by 1 per cent, wage bargainers will believe they can raise the real wage by 1 per cent, so if pre-existing inﬂation is 2 per cent, this implies that unions will individually set nominal wage inﬂation at 3 per cent, in the belief that the real wage will increase by 1 per cent. However, this implies that businesses will raise prices by 3 per cent to restore their markup. And if this process continues, nominal wage inﬂation and price inﬂation will subsequently rise to 4 per cent, and so on. So for equilibrium, the bargainable real wage wWS needs to be equal to the price-setting real wage. If the unemployment rate is too low, wWS > w; if too high, wWS < w, so the equilibrium unemployment rate brings wWS into line with wPS. Some useful consequences are as follows. Union bargaining power tends to be higher the lower the elasticity of demand in the (aggregated) product markets

Reforms in Germany 77 which correspond to union bargaining coverage, since an increase in the bargained real wage implies a smaller fall in employment; hence, equilibrium unemployment needs to be higher to bring down wWS into line with wPS. Since the more aggregated the product market is (i.e. the more sub-sectors it covers), the more inelastic will be product demand, and at the same time, the smaller will be the overall number of unions: so, ceteris paribus, the smaller the number of unions, the higher the equilibrium unemployment rate. (There is a double whammy here, for the more inelastic product demand, the larger will be the pricing markup μ and, hence, the smaller the price-setting real wage – also pushing up the equilibrium rate of unemployment.) But cetera are not necessarily pares: wage restraint, the foregoing of bargaining power, allows the equilibrium unemployment rate to fall, and small numbers of unions may be better able than larger numbers to organize wage restraint. Unfortunately for comparative policy analysis, much less work has been done in open economies. There is, however, a persuasive argument that there may be many unemployment equilibria in open economies.7 In the open economy, equilibrium unemployment is determined as in the closed by the requirement that the bargained real wage, wWS, is equal to the real wage implied by price-setting, wPS. The bargained real wage is determined as before, and we will keep with the simple case in which it declines as unemployment rises. But wPS is different. This is because the price level is now a weighted average of domestic costs of production (W) and world prices (P*). World prices are important since they directly affect import costs and indirectly impose limits on markups as a consequence of potential competition. If P* is high relative to P – that reduces the real wage for two reasons – it implies a high real cost of imports and it enables domestic companies to set a high proﬁt markup. The ratio P/P* is the real exchange rate;8 a high real exchange rate means a high price-setting real wage (low cost of imports and low proﬁt markup) and vice versa. Thus, a high real exchange rate allows a high bargained real wage and hence a low equilibrium level of unemployment and vice versa. In Figure 5.1, the real wage is on the vertical axis, with unemployment on the horizontal. The downward sloping relationship shows how the bargained real wage, wWS, declines as aggregate unemployment increases. The real wage implied by price-setting is high when the real exchange rate is high, wPSHigh, since the real cost of imports is then low and the proﬁt markup is subdued because of the competitive effects of low world prices. So when the real exchange rate is high, the high price-setting real wage means that the bargained or wage-setting real wage will also be high in equilibrium and that allows equilibrium unemployment to be low, ULow, as shown by the intersection at A. By contrast, a low real exchange rate implies a low price-setting real wage, hence a low bargained real wage and a high unemployment rate in equilibrium, indicated by the intersection at B. How is the equilibrium rate of unemployment chosen? A central part of the analysis is that the equilibrium is determined by aggregate demand. Figure 5.2 shows how, in the face of a negative aggregate demand shock and in the absence

78

W. Carlin and D. Soskice

Real wage

Bargained real wage schedule WS W W

PS High

High real exchange rate

A

PS

W Low

B

Low real exchange rate

W

ULow

WS

UHigh Unemployment

Figure 5.1 The effect of real exchange rates in real wage – unemployment space.

of any counteracting policies, a new higher unemployment stable inﬂation equilibrium is established. Assume the economy is initially in equilibrium at A with low equilibrium unemployment and a high real wage, wWS = wPSHigh, with high aggregate demand implying low actual equal to low equilibrium unemployment. There is then an adverse demand shock, shown in Figure 5.2 by the rightward shift of the vertical aggregate demand dashed line. This pushes up unemployment and pushes down the bargained real, so that the actual real wage (wPSHigh) is now above the bargained real wage (say, by 2 per cent). The consequence is that employers bargaining with unions will be able to push down the real wage. Suppose initially that world inﬂation, domestic inﬂation and nominal wage inﬂation are all 3 per cent, implying a constant real exchange rate (since domestic inﬂation is equal to world inﬂation) and a constant real wage (since nominal wage inﬂation is equal to domestic inﬂation). Then, if wage bargainers in year one believe domestic inﬂation will continue at 3 per cent, money wage inﬂation will fall to 1 per cent to engineer an expected real wage cut of 2 per cent. What do price setters do in consequence of these 1 per cent money wage increases? Domestic inﬂation is a weighted average of money wage inﬂation and world price inﬂation, so the fall in money wage inﬂation will reduce domestic inﬂation relative to world inﬂation and this will reduce the real exchange rate. Thus, if, for example, domestic inﬂation () is composed half of money wage inﬂation (W) and half of world price inﬂation (*), domestic price inﬂation will fall to 2 per cent ( = 0.5W + 0.5* = 0.5 1% + 0.5 3% = 2%). Hence, the real exchange rate will fall by 1 per cent, since domestic inﬂation of 2 per cent is

Reforms in Germany 79 Low unemployment implied by high aggregate demand Bargained real wage schedule

Real wage

W W

PS High

High unemployment implied by low aggregate demand

Adverse demand shock

WS

B*

A

High real exchange rate Ê * > > W

PS

W Low

B

Low real exchange rate W WS

ULow

UHigh Unemployment

Figure 5.2 The effect of aggregate demand in real wage – unemployment space.

1 per cent below world price inﬂation of 3 per cent, and the real wage will have fallen by 1 per cent as well. So the economy will have moved half way down the vertical line between B* and B. The next year, a similar process will be repeated in wage bargaining (since the real wage is now 1 per cent above the bargained real wage) and price-setting, and it will continue until the economy is eventually at B. (In fact with rational expectations, this process could take place immediately, with W = –1 per cent and = 1 per cent for just one period while * remains at 3 per cent; with all three rates reverting back to 3 per cent the next period; so the real wage and the real exchange rate both fall by 2 per cent; the economy then moves at once to its new equilibrium at B.) Monetary policy and the ECB In the standard closed economy model with a well-behaved central bank, New Keynesian inﬂation-targeting monetary policy would respond to an adverse demand shock and the consequent fall in inﬂation below target by lowering interest rates and returning the economy to equilibrium within a reasonable period of time.9 It is useful to see why monetary policy does not work effectively in the case of an adverse demand shock in Germany. Since Germany constitutes about one-third of the EMU GDP, a rise in unemployment in the German economy does not by deﬁnition fail to trigger ECB monetary policy. Although as compared with having an independent monetary

80

W. Carlin and D. Soskice

policy, one issue is the partially asymmetric ECB response function. Even so, we might expect falling inﬂation in Germany to move the ECB eventually into action. The real problem is captured by the analysis in the previous paragraph: German inﬂation falls only temporarily because the German economy reaches a new higher unemployment equilibrium. Once inﬂation is constant, the monetary policy reaction function even of an independent German central bank would not lead to a further cut in the interest rate. This is then a situation in which counter-cyclical ﬁscal policy would be regarded as the appropriate move. It is to this that we now turn. Fiscal policy and institutional blockage There is something of a consensus today that monetary policy works more effectively than ﬁscal policy: central banks can change interest rates quickly, and they can avoid the inﬂationary bias associated with the desire of governments to choose below equilibrium unemployment rates. Fiscal policy takes longer to put into operation (because of planning lags, parliamentary time, and so on), and it is subject to the time inconsistency problem of macroeconomic policy since politicians are taking the decisions (see, for example, Allsopp and Vines, 2005). But, equally, most New Keynesians accept that there are lacunae in monetary policy, and then governments need to resort to ﬁscal policy. The two most discussed situations in which monetary policy has shown itself actually or potentially ineffective are Japan in the last decade and asymmetric demand shocks in EMU (Allsopp and Vines, 2005; Krugman, 2005). From a political economy point of view, there are two institutional situations that have typically led to the creation of conservative frameworks for the operation of ﬁscal policy – ruling out the possibility of the use of discretionary ﬁscal policy (except perhaps in extreme circumstances, as eventually in Japan). The ﬁrst situation is where there are a limited number of powerful wage bargainers who set the inﬂation rate and de facto the real exchange rate; we will refer to this as the Small-N case. These powerful unions, usually in industrial sectors but also in the public sector, themselves represent a spectrum ranging from core highly skilled members in no danger of losing their jobs even with sizeable real wage increases to more substitutable employees in more marginally proﬁtable companies or industries. Assume these unions were acting independently (only partly true), and assume that the government was pursuing a discretionary ﬁscal policy to respond to increased unemployment, then each union is big enough to believe that an increase in unemployment as a result of its own actions will lead to some response by the government of the order of 1/N. By abandoning discretionary ﬁscal policy and therefore removing this possibility, the government sharpens the incentives for wage moderation. This analysis certainly applies to Germany. The coordination of wage bargaining which we observe in Germany is an ex post phenomenon reﬂecting the understanding of the major unions that any one of them is in a position to gain serious real wage increases; that this in turn would put pressure on other unions

Reforms in Germany 81 to be more aggressive; and that the fear of sharp conservative responses to inﬂation has led to concerns across the major unions to coordinate restraint. Note that the prospect of discretionary ﬁscal policy has no effect when labour markets are ﬂexible, or if there are a great many independent bargaining units, since no wage-setter believes that the tiny increase in unemployment resulting from its own wage bargain will change government expenditure. So neither the United Kingdom nor the United States, with ﬂexible labour markets, should fear the effect of discretionary ﬁscal policy. Equally, countries with encompassing wage bargainers, or countries in which all the relevant wage bargainers sit around a table with the government and make a binding wage contract, should be happy to conduct a ﬂexible ﬁscal policy. Second, the structure of governments may affect the incentives for running a discretionary ﬁscal policy (Hallerberg et al., 2001). The basic argument is this: coalition government leads to a common pool problem if individual ministries with presumably different party preference functions can independently decide on expenditure while taxation is general. Expenditure will be too high with coalition government. Thus, it will be in the interest of all the members of the coalition to work out a binding contract regarding how much each ministry (or party) should be allowed to spend. For the binding contract to work, there needs to be a tough ﬁscal framework. This is likely to be at odds with a discretionary approach since there is a principal agent problem: whichever agent (say the minister of ﬁnance) is put in charge of the discretion will incite the concern of the principal (the other coalition members) that it is being used to the advantage of the ﬁnance minister’s party. Hence, coalition governments will be nervous about discretionary ﬁscal policy. Coalition versus single party government is largely a product of the electoral system. Majoritarian systems nearly always produce single party government; proportional representation systems nearly always produce coalitions (or, what are de facto the same, minority governments with agreed support from other parties). German governments therefore have fought shy of discretionary ﬁscal policy both because wage bargaining is carried out by a small number of powerful unions and because governments are nearly always coalitions. Wage restraint in small and large economies The above analysis is also a useful starting point for understanding the differential effects on unemployment of wage restraint in large and small open economies. Leading unions in northern European Small-N type systems typically represent export and industrial sectors. As noted in the previous subsection, their membership covers a spectrum ranging from core highly skilled members in no danger of losing their jobs even with sizeable real wage increases to more substitutable employees in more marginally proﬁtable companies or industries. Just focusing on the German case, unions have little to hope for from the government’s ﬁscal policy in the event of an adverse demand shock (and

82

W. Carlin and D. Soskice

even less from the ECB). Hence, if they are concerned about the number of their members, this constrained policy-making framework is likely to push them to respond to adverse shocks with wage restraint. Thus, with government and big business concerned about international competitiveness, a tight ﬁscal policy is attractive. Indeed, export sector union leaders themselves may not be too unhappy with constraints that enable them to resist demands to raise real wages and hence risk losing members. In this sub-section, we follow this general line of argument and assume that unions respond to an adverse demand shock by wage restraint. What we show is that, if unions in exposed sectors do respond in this way, the effects on a large economy like Germany may be as negative for the economy as a whole (but not for the export sectors) as they are positive in smaller open economies such as Sweden or the Netherlands. Here is the argument: Both large and small economies are in equilibrium initially at A in Figure 5.3 with w = wWS = w0PS and U = U0. The relevant bargained real wage schedule is labelled ‘Initial’. We focus on a likely difference between the large and the small economy: the slope of the aggregate demand curves. The aggregate demand schedule in the small economy is shown by the long dashed line and in the large economy, the short dashed line. A policy of wage restraint in the open economy leads to a new equilibrium with lower real wages. A cut in the real

Aggregate demand Aggregate demand small econ large econ

W

WS

Real wage

High real exchange rate

A

PS W0

PS

W Small

B C

PS

W Large

Initial bargained real wage schedule WS

W Restrained bargained real wage schedule U1smallU0

U1Large Unemployment

Figure 5.3 Aggregate demand in large and small economies in real wage – unemployment space.

Reforms in Germany 83 wage has two opposite effects on aggregate demand: on the one hand, it improves international competitiveness by lowering the real exchange rate, increasing exports and reducing imports and hence raising aggregate demand and lowering unemployment; on the other hand, it lowers domestic demand. If this was the only effect, the aggregate demand curve in the diagram would be upward sloping. But a cut in the real wage also lowers consumption assuming that liquidity constraints prevent workers from smoothing consumption in response to the fall in the real wage. In a small and therefore more open economy, we assume the ﬁrst effect dominates so that a lower real wage leads to an economy-wide improvement in unemployment (B) and vice versa in a large economy (C) (in both cases, employment in the export sectors should improve).10 Heterogeneous labour markets: speciﬁc skills, welfare state reform and precautionary savings We have hinted up to now at two of the major differences between the German political economy on the one hand and those of the United Kingdom and the United States on the other. The ﬁrst is in the organization of labour markets: in Germany, a small number of powerful unions and employer associations, mainly located in the export sector, play a leading role in wage-setting. This contrasts with ﬂexible labour markets in the Anglo-Saxon economies. The second difference is in the structure of political systems: Germany has what Lijphart (1984) called a consensus system, with proportional representation and consequent coalition governments, while the United Kingdom and the United States are majoritarian systems with ﬁrst-past-the-post elections and single party government (for all their other differences). There are two other important and well-known differences: Germany has a stronger welfare state (with a relatively high replacement rate) than the United Kingdom or the United States (continental rather than liberal in Esping-Andersen’s (1990) terminology). And Germany has as well a vocational training system in which companies play a leading role and in which young workers acquire relatively deep skills speciﬁc to a sector and to a considerable extent to a ﬁrm. By contrast, the United Kingdom and the United States place far more emphasis on general education and training. These differences in industrial relations, political systems, welfare states and education and training are not random but have co-evolved in ways which imply powerful complementarities. In the German (and northern European systems more generally), workers with company- and sector-speciﬁc skills are in a powerful position within their company – thus, paradoxically strong unions that keep wage bargaining outside the company are necessary to guarantee the longterm cooperative behaviour of the skilled insiders. Strong welfare states provide insurance for workers with speciﬁc skills in case they do lose their job, and behind the welfare state, the consensus political system underwrites their interests. Indeed, if the society wishes to encourage investments in speciﬁc skills and

84

W. Carlin and D. Soskice

if those with speciﬁc skills need the insurance of a strong welfare state, then the political system has to guarantee that institutional change requires wide agreement (Iversen, 2005). The complementarities operate in quite different ways in the United Kingdom and the United States. Hall and Soskice (2001) describe the two types of political economies as different varieties of capitalism – liberal market economies (LMEs) in the case of the Anglo-Saxon countries and coordinated market economies (CMEs) for the northern European. While its institutions are very different, Japan is a type of CME with recognizably functionally equivalent systems. Its problems in the past decade can be analysed in similar ways to those in Germany (Vogel, 2006). We suggest that the very different type of labour market and skill organization in Germany, in comparison with the United Kingdom and the United States, leads to problems for consumption behaviour. Employees with speciﬁc skills can be expected to react with particular concern to a slowdown in growth, a rise in unemployment and the fear of welfare state reforms to unemployment beneﬁts and to pensions. For workers with speciﬁc skills, it will generally be harder to ﬁnd appropriate re-employment if they lose their jobs. Moreover, there is a negative externality in a labour market dominated by speciﬁc skills. If most of the work force has long-term employment, the number of vacancies within a given category of employment is likely to be limited, and companies may anyway seek to ﬁll vacancies via apprenticeships. Thus, mid-career labour markets for many categories may be quite limited or ‘illiquid’. The most obvious comparative example of this is with life-time employment in Japan: in the relevant categories, mid-career labour markets do not exist, short of accepting a position in a subsidiary company. In Germany as in Japan, illiquidity of mid-career labour markets applies more to the relatively more highly skilled – since companies have already invested more in them, and since it pays companies to invest more in them, they have longer tenure. Less-skilled workers, but still with apprenticeship certiﬁcates, face more open occupational labour markets – which are what portable qualiﬁcations should equip them to do. Given serious concerns about unemployment and with governmental pressure for welfare state reform in unemployment beneﬁts and pensions (in Japan, equivalently ending ‘life-time’ employment), those with speciﬁc skills who remain employed – in fact the great majority – respond by building up savings. Savings result not from an interest rate incentive to substitute future for present consumption but from precautionary savings, and in response to actual recent cuts in state pension entitlements, life-cycle savings. Iversen’s (2005) analysis of the guarantor-insurance role of the welfare state for those with deep speciﬁc skills in a CME explains why this substantial proportion of the workforce should feel insecure as its welfare state beneﬁts start to be questioned. Many employees factor in the possibility of involuntary early retirement or part-time work from their mid-ﬁfties should economic conditions become difﬁcult – both schemes which depend on welfare state provision. By contrast to LMEs, such as the United Kingdom or the United States, where a workforce with more general skills could imagine at a similar age responding to

Reforms in Germany 85 economic difﬁculty by ﬁnding alternative employment, labour markets for older workers do not exist on any substantial scale in Germany. These fears are exacerbated by the consensus nature of political institutions in CMEs. As discussed above, this reﬂects an economic environment in which institutional change requires wide agreement if the environment is to continue to encourage investments in speciﬁc assets. To summarize, our claim is that the speciﬁcity of skills in Germany, aggravated by uncertainty about the future of the protective welfare state, ampliﬁes any adverse aggregate demand shock by further dampening household consumption growth.

The United Kingdom and Germany: comparative performance, reforms and policies in the 1990s and 2000s In this section, we investigate the hypotheses set out in Section ‘Reinstating a role for aggregate demand in analysing German economic performance’ by comparing the performance, aggregate demand behaviour and reform experience of the German and British economies. At ﬁrst sight, the contrasting fortunes of the British and German economies over the past 15 years appear to conform to the mainstream view. The vigour with which supply-side reforms have been undertaken in the United Kingdom since the 1980s is well known – as is the long period of robust growth, falling unemployment and low and stable inﬂation. Meanwhile, Germany has been dogged by increasing unemployment and sluggish growth and deemed to have a sclerotic and unreformed supply side.11 Table 5.1 provides comparative data for Germany and the United Kingdom, which shows the well-known deterioration in unemployment performance in Germany throughout the period and its improvement in the United Kingdom. Similar contrasts are revealed in comparative growth performance and in levels of public sector indebtedness. The United Kingdom has also achieved remarkable macroeconomic stability in the past decade. It is ranked ﬁrst amongst 20 OECD countries in having the lowest average absolute output gap and sixth for inﬂation stability. In terms of public sector stability as indicated by the lowest debt to GDP ratio, the United Kingdom ranked ﬁfth in 2005. Across each of these dimensions, Germany performed worse. However, the relatively weak performance of British productivity growth is reﬂected in the fact that GDP per hour worked continues to be higher in Germany than in the United Kingdom, although longer hours of work and a higher employment rate give the United Kingdom 10 per cent higher GDP per capita. Table 5.2 reports some commonly cited indicators of the institutional environment. The ﬁrst set refers to ‘economic freedom’ and the second to broader ‘framework conditions’ for economic performance. The two dimensions of the Fraser Institute Index of Economic Freedom that relate most closely to the concept of the free operation of market forces are ‘Legal structure and property rights’ and ‘Regulation of labour, credit and business’. The World Economic Forum’s (WEF) Growth Competitiveness Index is a broad index of framework conditions: ‘Competitiveness is deﬁned as that collection of factors, policies and

86 W. Carlin and D. Soskice Table 5.1 Economic performance: Germany and the UK

1980s 1990s 2000–2005

Unemployment rate (%)

Growth in real GDP (% p.a.)**

Level of public sector debt to GDP, (%)**

G

UK

G

UK

G

UK

5.8 7.0* 8.5

9.8 7.9 5.0

2.3 1.4 1.1

2.6 2.4 2.6

38.9 50.8 63.9

48.7 46.5 43.5

Level of GDP p.c. (US = 100), 2005

Level of GDP per hour Rank in OECD 20 worked (US = 100), 2005 smallest average absolute output gap; lowest variance in CPI inﬂation (1995–2005)

71.4

98.6

78.4

88.9

8; 8

1; 6

Sources: Growth and levels of GDP pc. At purchasing power parity are from Groningen Growth and Development Centre and the Conference Board, Total Economy Database, May 2006, www.ggdc.net. Other variables are from OECD Economic Outlook Database June 2006. Notes *Includes East Germany from 1993; **1980–90; 1991–99; 2000–05; West Germany for the ﬁrst period. Unemployment is the standardized measure of unemployment as reported by the OECD.

institutions which determine the level of productivity of a country and that, therefore, determine the level of prosperity that can be attained by an economy’ (2005: xiii). The WEF Business Competitiveness Index: ‘. . .speciﬁcally measures two areas that are critical to the microeconomic business environment in an economy: the sophistication of company operations and strategy, as well as the quality of the overarching national business environment in which they are operating’ (xvi). In the indices of economic freedom, the sharpest contrast between the United Kingdom and Germany is in the ‘regulation of credit, labour and business’ subindex. Here, the United Kingdom ranks close to the top across 20 OECD countries and has done so for the past decade – moving from ﬁfth to second place over the 1980s. Germany appears towards the bottom on this measure throughout the period. The countries are much closer together in their scores and ranking for the legal structure and property rights measure. Both countries rank in mid-table in the WEF’s Growth Competitiveness Index with similar scores on the technology sub-component. Germany rates higher in the public institutions dimension, which includes infrastructure and much lower on the macroeconomic environment. Perhaps more surprising is that Germany ranks higher than the United Kingdom on the business competitiveness indicator and each of its components.

Reforms in Germany 87 Table 5.2 Indicators of economic freedom and competitiveness, Germany and the UK Freedom in markets scores (a higher score represents more freedom) Legal structure and property rights

Germany UK

1980

1990

2003

Rank in OECD20, 2003

7.7 7.0

8.3 7.7

8.9 9.2

7 5

Regulation of credit, labour and business

Germany UK

1980

1990

2003

Rank in OECD20, 2003

5.2 6.0

5.3 6.8

5.7 7.4

19 3

Competitiveness scores, 2005 (rank in OECD20) Growth competitiveness index

Germany UK

12 10

Business competitiveness index

Germany UK

3 6

Sub-components Technology

Public Institutions

Macroeconomic environment

11 12

5 9

15 9

Sub-components Company operations and strategy

Quality of national business environment

2 6

4 6

Sources: World Economic Forum (2005), Fraser Institute 2005a. Data retrieved from www. freetheworld.com.

There is little dispute that British macroeconomic performance has been impressive over the past decade. As compared with Germany, UK economic strength is most apparent when measured by aggregate growth and unemployment and the success of policy when measured by macroeconomic stability. The institutional indicators – especially for economic freedom – conﬁrm that differences persist between the two economies. All this is in line with the prevailing characterization of these economies. However, German performance remains better in hourly productivity, and the rankings in the competitiveness measures differ most on the macroeconomic (to Germany’s detriment) and infrastructure (to the United Kingdom’s detriment) components rather than on technology or the quality of the business environment.

88

W. Carlin and D. Soskice

The analysis in Section ‘Reinstating a role for aggregate demand in analysing German economic performance’ indicates that in the standard closed economy model, a sustained fall in unemployment without inflationary consequences as was the case in the United Kingdom can only be the result of supply-side improvements either through reforms or through positive productivity shocks. Conversely, a sustained rise in unemployment without persistent deflation reflects supply-side deterioration. According to the standard model, supply-side reforms produced falling equilibrium unemployment in the United Kingdom, which allowed aggregate demand to expand so as to maintain buoyant growth. The expansion of aggregate demand could have been due to domestic household or business sector expenditure (or net exports), either responding directly or indirectly to the improvement in conditions on the supply side, as a consequence of the operation of the macroeconomic policy response functions. With lower equilibrium unemployment, the central bank will find that as unemployment falls, inflationary pressures do not rise, allowing it to refrain from tightening policy. Equally, the fiscal authorities will be able to undertake discretionary policies to help sustain aggregate demand consistent with lower unemployment. The relationship between supply and demand factors is summarized by Nickell et al. (2005: 22): it is important to recall that unemployment is always determined by aggregate demand. As a consequence we are effectively trying to understand the long-term shifts in both unemployment and aggregate demand (relative to potential output). We emphasise this because it is sometimes thought that the fact that unemployment is determined by aggregate demand factors is somehow inconsistent with the notion that unemployment is inﬂuenced by labour market institutions. This is wholly incorrect. In the case of Germany, the symmetric argument is that a persistent deterioration of the supply side was responsible for the trend increase in unemployment. Given this deterioration, macroeconomic policy had to remain tight in order to contain the upward pressure on inﬂation and to manage the shift of the economy to higher unemployment. If the shift in the equilibrium rate was not reﬂected in the measured output gap, such policy would appear to be pro-cyclical. Aggregate demand would be depressed and growth weak. In the conventional view as summarized by the above quote, aggregate demand can only play a passive role in accounting for longer-term changes in unemployment. Since only equilibrium unemployment is sustainable, ﬂuctuations in aggregate demand will either account for temporary deviations of the economy from the equilibrium or track the path of the economy to a new equilibrium following a supply-side shift. For example, if supply-side conditions have improved but the response of private sector aggregate demand remains too weak, the macroeconomic policy reaction functions will produce the required stabilization around the new equilibrium. Equally, if there are exogenous shocks

Reforms in Germany 89 to aggregate demand (e.g. from the rest of the world), this may require policy to dampen or stimulate activity. The open economy model set out in Section ‘Reinstating a role for aggregate demand in analysing German economic performance’ explains how shifts in aggregate demand can take on independent signiﬁcance in an open economy since there is a range of equilibrium unemployment rates – each associated with a different real exchange rate. The core of the argument is that a change in the real exchange rate implies a change in the real wage, which in turn changes conditions on the supply side. A positive aggregate demand shock will reduce unemployment and lead to a real appreciation, and a negative aggregate demand shock will raise unemployment and lead to a real depreciation. Since real wages rise as a consequence of a real appreciation, the lower unemployment associated with the higher level of aggregate demand is sustainable and, conversely, if there is a real depreciation, real wages fall and the higher unemployment associated with depressed aggregate demand will not produce falling inﬂation. Hence, even in the absence of any supply-side improvement, an economy can experience a sustained fall in unemployment following a sustained positive aggregate demand shock. The model predicts that such a fall in unemployment is associated with an appreciation of the real exchange rate (and a deterioration in the current account) and a rise with a real depreciation (and an improvement in the current account). As noted earlier, we assume there is no immediate feedback from the changes in the current account – the medium-run equilibrium can therefore persist for some time. Just as in the closed economy model, supply-side improvements will reduce equilibrium unemployment. If a fall in unemployment is due to a supply-side improvement, it is associated with a real depreciation: competitiveness improves either because of wage restraint associated with the supply-side improvement or because of enhanced productivity growth. The implications for the external balance are the opposite if the reduction in unemployment is mainly demand rather than supply driven (see Table 5.3). Prima facie evidence that supply-side developments do not fully account for the performance of the British and German economies over the past decade and a half comes from an inspection of Figure 5.4. The unemployment rate (standardized) is shown on the horizontal axis and the real exchange rate as measured by relative unit labour costs in manufacturing is on the vertical axis. A predominantly supply-side interpretation of UK performance would predict the fall in unemployment to be associated with a shift to the south west (the chart for Ireland provides an illustration of such a pattern). Instead, the economy moves along a path from south-east to north-west: this is consistent with a series of positive aggregate demand shocks. Consistent with some improvement on the supply side is the apparent shift to the left in the relationship between the real exchange rate and unemployment but that is far from the only factor involved as shown by the major real appreciation. Turning to Germany, a supply-side explanation for rising unemployment would predict that the rise in unemployment would be associated with a real

90 W. Carlin and D. Soskice Table 5.3 Comparison of predictions in the closed and open economy versions of the New Keynesian macro model

Can an exogenous shift in aggregate demand shift medium-run unemployment?

Can an exogenous shift in aggregate supply shift medium-run unemployment?

Closed economy model

Open economy model

No.

Yes.

Positive shift: unemployment falls and inﬂation rises. Aggregate demand contracts either endogenously or via policy reaction functions.

Positive shift: under ﬁxed exchange rates or inﬂation targeting, unemployment falls and inﬂation rises. Policy is tightened. There is a new equilibrium at lower unemployment, a higher real wage and an appreciated real exchange rate. The external balance deteriorates.

Yes.

Yes.

Positive shift: inﬂation falls. Aggregate demand expands either endogenously or via policy reaction functions.

Positive shift: inﬂation falls. There is a new equilibrium at lower unemployment, a lower real wage and a depreciated real exchange rate. The external balance improves.

appreciation – a move to the north-east – with a combination of wage pressure and poor productivity performance depressing net exports. However, the impression from Figure 5.4 is that the real exchange rate has moved within a narrow range. When combined with the evidence that net exports have been strong (not only relative to other components of aggregate demand in Germany but also in international comparison), this casts some doubt on the supply-side story as the dominant one. The alternative hypothesis as argued in Section ‘Reinstating a role for aggregate demand in analysing German economic performance’ is that with wage-setting led by tradable sector unions in Germany, wage growth has been oriented towards maintaining the real exchange rate and shifts in aggregate demand lie behind the horizontal path. In the rest of this section, we set out the contrasting behaviour of the components of aggregate demand in Germany and the United Kingdom, concentrating on the post-1990 period. We then turn to the supply side and evaluate some existing evidence about the extent to which supply-side changes are estimated to account for trends in unemployment in both countries. Aggregate demand and employment: Germany and the United Kingdom compared We begin by comparing the decomposition of GDP growth according to its components in the United Kingdom and Germany. Figure 5.5 shows for each

United Kingdom

Relative unit labour cost

120

05

100

00

91 80 95 60 4

6

8

10

Unemployment rate

Germany 120 Relative unit labour cost

05

00

100

0

81

80

60 4

6

8 Unemployment rate

10

Ireland

Relative unit labour cost

200

81

150

100

06 00 6

10 Unemployment rate

16

Figure 5.4 The evolution of the real exchange rate and unemployment in UK, Germany and Ireland, 1991–2005 (source: OECD Economic Outlook database 2006). Note: RULC = 100 in 2000; an increase in RULC is a real appreciation.

92 W. Carlin and D. Soskice UK 3.5 3 2.5 2 1.5 1 0.5 2004

2005

2004

2005

2003

2002

2001

2000

1999

1998

1997

1996

1995

1994

1

1993

0 0.5

Govt cons. and investment GDP

Household cons. and housing Net exports Private non-housing investment

Germany 3.5 3 2.5 2 1.5 1

Household cons. and housing Net exports Private non-housing investment

2003

2002

2001

2000

1999

1998

1997

1996

1995

1

1994

0 0.5

1993

0.5

Govt cons. and investment GDP

Figure 5.5 Contributions to the growth of real GDP (source: OECD National Accounts database, June 2006). Notes: Contributions to the growth of constant price GDP (3-year moving average). The contributions do not add up to GDP growth because of the omission of stock-building, a statistical discrepancy and the treatment of government investment in housing. Data for 2006 are OECD forecasts.

economy, the growth of real GDP and the contribution to growth of four key components: private sector consumption plus housing; private sector ﬁxed investment (excluding housing); government consumption and investment; and net exports. Each series is presented as the three-year moving average. The top panel of Figure 5.5 for the United Kingdom shows high and stable growth of GDP throughout the whole period from the early 1990s. There was a consistently high positive contribution to the growth of GDP from the household

Reforms in Germany 93 sector. The contribution of government expenditure was the mirror image of that of business sector investment: the sustained investment boom in the second half of the 1990s was mirrored by contracting public expenditure and the reverse characterized the 2000s. The real appreciation of the exchange rate from 1997 (see Figure 5.4) was followed by a deterioration of net exports, which remained as a negative inﬂuence on the growth of demand for the remainder of the period. As the lower panel indicates, German growth was much weaker than in the United Kingdom through the 1990s. The ﬁrst notable feature is the low and declining contribution of government expenditure to growth. This contributed less than 0.5 per cent per annum to real growth each year and became negative from the early 2000s. Second, the contribution of household expenditure fell sharply from 2000, itself becoming negative at about the same time as did government expenditure. Third, net exports made a positive contribution over most of the period and dominated as the source of positive demand growth after household expenditure collapsed in the early 2000s. The contribution of private investment ﬂuctuated within a narrow band from –0.5 to 1 per cent per annum. The contrast in aggregate demand behaviour between the two economies is stark. Domestic aggregate demand was strong throughout the period in the United Kingdom, with the government sector appearing to play a key role in stabilizing and sustaining it. Particularly in the most recent period, the simultaneous weakness of private and government components of domestic demand in Germany is striking. The upper panel of Figure 5.6 shows the evolution of the household savings ratio (as a percentage of disposable income) over the period in both countries. The pattern for Germany is a shallow U-shape, with trough at the turn of the century. By contrast, there was a dramatic fall in the UK savings rate beginning around 1997. An important structural difference between the two economies over this period that is linked to consumption and savings behaviour is the evolution of house prices. As the lower panel of Figure 5.6 indicates, house price trends in Germany and the United Kingdom have been opposite in the last 15 years. Since 1995, real house prices in the United Kingdom have risen by an annual rate of more than 8 per cent, whereas they have fallen by more than 2 per cent per annum in Germany.12 From the behaviour of government expenditure in the United Kingdom shown in Figure 5.5, it appears that discretionary ﬁscal policy was countercyclical. This impression is reinforced by the data in Figure 5.7, which compare the path of the discretionary ﬁscal stimulus (as measured by the change in the cyclically adjusted primary government balance) in Germany and the United Kingdom. By contrast, discretionary ﬁscal policy appears both more muted and less systematic in Germany, e.g. tightening in parallel with weakening household expenditure both in the mid-1990s and in the early 2000s. We can also trace the impact of the differential role played by aggregate demand in Germany and the United Kingdom by examining the changes in employment in different industries. Table 5.4 shows the percentage point contributions to the growth of the employment rate in Germany and the United Kingdom from 1995 to 2003. As is well known, the employment rate in the

Household savings ratio, %

14 13 12 11 10 9 8 7 6

Germany

2005

2004

2003

2002

2001

2000

1999

1998

1997

1996

1995

1994

1993

4

1992

5

UK

Real house price changes, % p.a. 12 10 8 6 4 2 0 2 4 6

1990–1995

1996–2000 Germany

2000–2005 UK

Figure 5.6 Household savings ratio and real house price changes, UK and Germany (source: The household savings ratio comes from OECD National Accounts database; the house price data is from OECD 2006; 79, p. 18).

Reforms in Germany 95 Change in cyclically adjusted primary budget balance, % 2.5 2.0 1.5 1.0

2005

2004

2003

2002

2001

2000

1999

1998

1997

1996

1995

1.0

1994

0.5

1993

0.0

1992

0.5

1.5 2.0 2.5 3.0

Germany

UK

Figure 5.7 Indicator of discretionary ﬁscal policy (source: OECD National Accounts database). Notes: The indicator of discretionary ﬁscal policy is the change in the cyclically adjusted primary (i.e. non-interest) budget balance.

Table 5.4 The industry composition of changes in the employment rate, 1995–2003 Industry

Germany

UK

1 2 3 4 5 6 7 8 9 10

Agriculture, forestry and ﬁshing Mining and energy Manufacturing Construction Distribution Transport and communications Finance, business services Public administration, health, education Other services Total

–0.4 –0.3 –1.3 –1.7 0.9 –0.1 3.1 1.2 0.7 2.0

–0.3 –0.2 –1.5 0.5 1.7 0.5 3.0 2.2 1.0 7.0

Source: Calculated from Groningen Growth and Development Centre, 60-Industry Database, October 2005, www.ggdc.net updated from O’Mahony and van Ark (2003). Note The table shows the absolute change in employment 1995–2003 divided by the 2003 population of working age.

United Kingdom grew much more than in Germany – a difference of 5 percentage points (row 10). As the ﬁrst three rows demonstrate, none of this difference arises from the tradable goods sector, where the decline in each country was very similar (agriculture and mining fell by more in Germany and manufacturing by less). Perhaps more surprising is that ﬁnance and business services (row 7) contribute nothing to explain the weaker employment performance in

96

W. Carlin and D. Soskice

Germany. Major differences between the two countries were in job creation in distribution (retail, wholesale, hotels and catering) and in public administration, health and education. In each case, about 1 percentage point of the difference in employment growth is accounted for. Above all, it is in construction that the employment patterns differed: falling by 1.7 percentage points in Germany and rising by 0.5 in the United Kingdom. The slow growth of household expenditure was not reﬂected in greater employment loss in the traded goods sector in Germany because of the strength of export demand. However, this was not the case in distribution: it is here that the differences in household spending behaviour between the United Kingdom and Germany documented above were reﬂected in job creation.13 Equally, the public sector expenditure growth in the post-2000 period in the United Kingdom contributed substantially to the employment differences. The differences in construction were due in part to the unwinding of the infrastructure boom associated with uniﬁcation in Germany but also to the absence in West Germany of a revival of residential investment. Accounting for changing unemployment: supply-side evidence Within the context of the mainstream model, i.e. where medium-run trends in unemployment are the outcome of supply-side shifts in equilibrium unemployment, a large number of empirical studies have been carried out into the relationship between changes in unemployment and in labour market institutions. More recently, this approach has been extended to include the role of product market regulation. Recent systematic surveys of the accumulated evidence are available in Baker et al. (2005),14 Blanchard (2005) and Bassanini and Duval (2006).15 Baker et al. emphasize the lack of robustness of the estimated effects of changes in labour market institutions on unemployment in OECD countries, Bassanini and Duval present a more positive view of what has been learnt and Blanchard’s assessment lies in between.16 We use estimates from two recent studies in this tradition to highlight the implications for the diagnosis of unemployment trends in the United Kingdom and Germany. The ﬁrst is Nickell et al. (2002, hereafter NNO) and the second is Bassanini and Duval (2006, BD). Both studies include the same 20 OECD countries. The NNO study uses a longer time series from 1960 to 1995, whilst BD use improved data for some of the labour market institutions variables and a period from 1982 to 2003. In both studies, the estimation techniques make use of the within country variation in institutions (along with some controls for shocks) to explain trends in unemployment. The labour market institution variables included are measures of the replacement rate, the tax wedge, union density, employment protection legislation and coordination in wage-setting. The studies vary in their treatment of the replacement rate and beneﬁt duration. NNO include a number of interaction terms between the institutional variables, use the change rather than the level of union density and also include the real

Reforms in Germany 97 interest rate. BD include the newly constructed OECD measure of product market regulation. In their baseline regression, BD use the output gap as a catchall for macroeconomic shocks and NNO use four separate shocks. In spite of these differences, the results are fairly similar (NNO Table 5, column 1; BD Table 1.2, column 2): employment protection legislation is not signiﬁcant in either study, unemployment beneﬁt variables are positive and signiﬁcant, coordination is negative and signiﬁcant and the tax wedge is positive and signiﬁcant. NNO ﬁnd a positive and signiﬁcant effect on unemployment of the change in union density, but the level is not signiﬁcant in BD. The index of product market regulation is positive and signiﬁcant in BD, as is the real interest rate in NNO. The results for our two countries from simulating the NNO speciﬁcation, which includes lagged unemployment and country-speciﬁc time trends, are quite instructive. The dynamic simulation tracks the path of actual unemployment reasonably well (top panel of Figure 5.8). The lower panel of Figure 5.8 compares the standard dynamic simulation for each country with the simulated path when institutions are held at their values in 1960. The graph for the United Kingdom shows that (based on the regression coefﬁcients) deleterious changes in institutions, from the 1960s through to the end of the 1970s, predict rising unemployment during this period. The narrowing of the gap between the standard and the ﬁxed institutions simulations thereafter reﬂects the effect of an improvement in the institutional variables. When the same experiment is repeated for Germany, the results are different. Although changing institutions predict a modest part of the rise in unemployment in the 1960s and 1970s, they predict nothing thereafter. Turning to the BD baseline regression, which focuses on the post-1983 period, we can decompose the predicted changes in unemployment for the United Kingdom and Germany by using the estimated coefﬁcients from the panel regression and applying them to the changes in the institutional variable for each country. Table 5.5 presents the results: the analysis for Germany is divided into preand post-uniﬁcation periods. The variables for which the coefﬁcients are not signiﬁcant are shown in italics. The mechanical application of the estimated coefﬁcients to the institutional changes in Germany suggests that unemployment should have fallen in both periods, with institutional improvements – especially product market deregulation – driving the reduction since 1993. The evolution of each of the supply-side variables for Germany and the United Kingdom is shown in Appendix 5.1. Parallel to the ﬁndings from the NNO exercise, the BD estimates are more successful in accounting for the fall in unemployment in the United Kingdom than for its rise in Germany. The big predicted contributions come from the reduction in the tax wedge and the decline in product market regulation. However, once the change in the output gap is taken into account, the equation predicts a substantially larger fall in unemployment in the United Kingdom than occurred (a fall of 9.3 points as compared with the fall of 5.5 points observed).

United Kingdom Actual unemployment rate

Unemployment

14 12

Standard dynamic simulation

10 8 6 4 2 0 1960 1965 1970 1975 1980 1985 1990 1995 Year Germany

Actual unemployment rate

Unemployment

14 12

Standard dynamic simulation

10 8 6 4 2 0 1960 1965 1970 1975 1980 1985 1990 1995 Year

United Kingdom

Unemployment

12 10

Standard dynamic simulation

8

All institutions fixed

6 4 2 1 1960 1965 1970 1975 1980 1985 1990 1995 Year

Germany

Unemployment

12

Standard dynamic simulation

10

All institutions fixed

8 6 4 2 0 1960 1965 1970 1975 1980 1985 1990 1995 Y

Figure 5.8 The role of changing institutions in explaining unemployment trends: simulations from Nickell et al. (2005) (source: Nickell et al. (2005) data set; speciﬁcations as in NNO).

–0.9 0.1 –3.8 0.0 –0.7 0.0 8.2

–0.6 –0.2 –8.6 –0.9 –2.4 0.0 –0.1

Change 1993–2003, Germany

–6.6 –12.5 –18.2 0.1 –3.4 0 6.1 Predicted change in U Predicted by policies, i.e. excluding output gap Actual change in U

Change 1982–2003, UK

–0.1 0.03 ns ns –0.4 0.0 –4.0 –1.5 –1.6 1.4

–0.5 –1.8

Germany, 1982–1990

–6.3 –5.5

–0.1 –0.1 ns ns –1.5 0.0 0.1 –9.3

Germany, 1993–2003

–0.8 –3.5 ns ns –2.1 0.0 –2.9

UK, 1982–2003

Contribution to unemployment change

0.1 0.3 –0.03 –0.3 0.6 –1.4 –0.5 –4.4

Bassanini Duval coefﬁcients

Note The predicted changes exclude the impact of changes in variables where the coefﬁcients were not signiﬁcant (ns). Columns may not add exactly due to rounding.

Source: Bassanini and Duval (2006).

Average replacement rate Tax wedge Union density Employment protection Product market regulation High corporatism Output gap

Change 1982–1990, Germany

Table 5.5 Accounting for the change in unemployment in Germany and the UK using the estimated coefﬁcients from Bassanini and Duval (2006).

100

W. Carlin and D. Soskice

Macropolicy and supply-side reforms: comparing the United Kingdom and Germany As we have seen, Germany has implemented supply-side reforms that should – according to the cross-country evidence – have produced lower unemployment. According to the standard model, these supply-side improvements would have enabled aggregate demand to evolve in line to support a higher growth rate and lower unemployment. In this section, we look at a number of the puzzles raised by this ﬁnding. First, we document the more limited activism of macroeconomic policy in Germany than in most other OECD economies. Second, we present some evidence that reform intensity has been relatively strong in Germany (as compared with other OECD economies). Third, we identify features of the German economy that may explain why the supply-side reform effort has had such a meagre pay-off. Bringing these ﬁndings together in the context of the discussion in Section ‘Reinstating a role for aggregate demand in analysing German economic performance’, we can propose a hypothesis to explain why domestic aggregate demand in Germany has been so weak and why macroeconomic policy has done so little to contribute to stabilization. Has less use been made of stabilization policy in Germany than elsewhere? In Section ‘Reinstating a role for aggregate demand in analysing German economic performance’, we argued that the presence of large wage-setters in the economy can inhibit the use of macroeconomic policy for stabilization. Unfortunately, there is no systematic empirical evidence testing this hypothesis. Using as very crude indicators of the use of discretionary policy, the standard deviation of the change in cyclically adjusted ﬁscal balance and the real interest rate, we ﬁnd that in a cross-country comparison, Germany has the second lowest variability on both measures in the 1990–2005 period across 20 OECD countries. In a cross-country comparison, the United Kingdom ranks as sixth most active in ﬁscal and twelfth in monetary policy.17 Recent work by the OECD conﬁrms that the use of counter-cyclical policy appears to vary systematically across groups of countries in a way that is consistent with the predictions of Section ‘Reinstating a role for aggregate demand in analysing German economic performance’. Figure 5.9 shows the contemporaneous correlation between the change in the cyclically adjusted primary balance and the output gap over the period 1981–2005 for the countries that now make up the euro zone, the Nordic countries and ‘other’ OECD countries. Finland is included with the Nordic group. The results indicate that ﬁscal policy has tended to be pro-cyclical in the euro zone countries but counter-cyclical in the two other groups. As suggested in Section ‘Reinstating a role for aggregate demand in analysing German economic performance’, one explanation for this may lie in the political economy of macroeconomic policy making. Discretionary stabilizing policy is easier for governments to use both in the Nordic countries with

0.45 0.35 0.25 0.15 0.05 0.05 0.15 0.25 0.35

Euroarea

Nordic

Other OECD

Cyclically adjusted primary balance

0.45 0.35 0.25 0.15 0.05 0.05 0.15 0.25 0.35

Euroarea

Nordic

Other OECD

Expenditure and revenue

Figure 5.9 Is discretionary ﬁscal policy counter-cyclical? (source: Data from OECD 2006, p. 32. Additional detail from Cotis (2006). Each bar shows the contemporaneous correlation between changes in the cyclically adjusted primary balance and the output gap over the period 1981–2005. Germany (1991–2005). The sign has been changed for primary expenditure in the panel with expenditure and revenue so that a positive correlation indicates counter-cyclicality. The euro area excludes Finland, which is included in the Nordic group (with Denmark, Norway, and Sweden). Other OECD includes US, UK, Japan, Canada, Australia, South Korea, New Zealand and Switzerland.

102

W. Carlin and D. Soskice

encompassing institutions and in the liberal economies that dominate the ‘other’ OECD group than in the economies with coalition governments and large wagesetters that are prevalent in the euro zone group. How intensive has the reform effort been in Germany? One way to standardize the assessment of the policy reform effort that has taken place is to use the indicators created by the OECD. Starting from the inﬂuential Jobs Report in 1994 followed by their assessment of progress in 1999 and the introduction in 2005 of annual reports of policy reforms in the series Economic Policy Reforms: Going for Growth, the OECD has developed an indicator-based approach to the evaluation of the reform effort across countries. Table 5.6 provides a summary of the OECD’s aggregate measure of labour market reform and includes all legislated reforms to the middle of 2004. This shows that Germany has been very active (only Denmark, the Netherlands and Finland rank higher) in the adoption of labour market reforms. The score shown in the table refers to the number of actions taken: the details of how the German reforms were scored are shown in Appendix 5.1. The United Kingdom has been less active, which reﬂects a smaller shortfall from an implicit benchmark for a ‘reformed’ labour market. Table 5.7 indicates that Germany has implemented reforms across all seven categories, whereas the United Kingdom has concentrated on reform in active labour market measures and tax and beneﬁt reform. German reforms have all been in the recommended direction. Appendix 5.2 presents a detailed explanation of how Germany’s reforms were scored in the OECD study. Can reform effort affect private sector aggregate demand? We have shown that aggregate household expenditure has behaved quite differently in Germany as compared with the United Kingdom over the past decade and especially since 2000. One hypothesis is that the nature of the reform programme debated and pursued in Germany has created uncertainty about job security, which has led to an increase in precautionary savings. One of the channels through which the labour market reforms, such as lower replacement ratios and shorter duration of unemployment beneﬁt, are supposed to have their employment enhancing supply-side effects is by increasing the cost of job loss. Table 5.6 The intensity of labour market reforms undertaken between 1994 and 2004.

Germany UK

Score

Rank (out of 30 OECD countries)

23.9 16.7

4 11

Source: Brandt et al. (2005). Table 8, p. 56. The score relates to the number of actions taken. See the Appendix for more details.

Reforms in Germany 103 Table 5.7 Labour market reform intensity from 1994 to 2004 by individual indicators. EPL

Unemp. beneﬁts

Germany 16.7 19.2 UK –10.0 11.5

Working time ﬂexibility

Early retirement

Wage formation and industrial relations

ALMP

Taxes and social security

16.7 0.0

25.0 25.0

9.1 –9.1

57.7 50.0

12.5 56.3

Source: Brandt et al. (2005). Table 8, p. 56. Measures introduced contrary to the recommendations of the Jobs Study are scored with a negative number. The score relates to the number of actions taken. See the Appendix for more details.

Reductions in employment protection are aimed at reducing the cost of hiring by making it easier to ﬁre workers. Whilst there has been considerable debate about whether reducing employment protection is likely to affect investment by workers and ﬁrms in job matches, less has been said about the potential impact on savings behaviour of such actual and prospective changes.18 As argued in Section ‘Reinstating a role for aggregate demand in analysing German economic performance’, the speciﬁc characteristics of the German labour market are likely to have played a role here. Vocational qualiﬁcations play a far greater role in the German labour market than elsewhere. Through the apprenticeship system, workers tend to acquire industry competences, which are supplemented by ﬁrm-speciﬁc skills. Job tenures are on average longer than in other countries. The consequence is that the ‘mid-career’ labour market is far less developed than in an economy like the United Kingdom. The prospect of labour market reforms that increase the cost of job loss is therefore likely to have a greater effect in boosting savings than, for example, in the United Kingdom. This effect is likely to be reinforced by the fact that in contrast to the Nordic countries where two-earner households are the norm, the German labour market and welfare state are characterized by provisions that encourage single earner households. Recent cross-country data for 20 OECD countries (upper panel of Figure 5.10) illustrates a positive – although not signiﬁcant – correlation between the female employment rate and fertility. The correlation in 1980 was weakly negative. The lower panel highlights that the Nordic countries have increased (or maintained very high) employment rates and increased fertility rates since 1980. Germany has both a low female employment rate and a low fertility. Variations in arrangements exist across the Nordic countries – e.g. in Finland, generous allowances are paid to women who have children and decide to remain at home while the children are very young. Their jobs are held open for them. In Sweden, families with young children have generous maternity and paternity leave, both parents tend to remain in employment and make use of high quality child care (e.g. Hoem, 2005). Both arrangements have been associated with a

Cross-country relation between fertility and the female employment rate, 2003 2.2 2 Total fertility rate, %

US NZ

Ir Fr

No Cl Swe

Ne Fl Au U

1.8 Es 1.6

Ca

1.4 It

G

S

Aut

P

Sw I

J

1.2 1 40

50

60

70

80

Female employment rate, %

Change in fertility rate and change in female employment rate, 1980–2003

Change in fertility rate, % point

Fl

10

0.4 0.2 Swe 0 0.2 0.4 0.6 0.8

US No

DK

US

Es Fr 10 Au G

20 NZ Sw I

Ne

30

40

J P S

1 1.2

Ir

1.4 1.6 Change in female employment rate, % point

Figure 5.10 Fertility and female employment rates: cross-country comparisons (source: OECD data). Note: The female employment rate is employment as a percentage of the population aged 15–64.

Reforms in Germany 105 revival of fertility rates along with continued high employment rates for women. Such policies presumably help to maintain the growth of household expenditure. For example, in the sharp post-1990 recession in Finland, families appear to have responded to the weakness of the labour market by shifting the timing of family formation: women received generous out-of-work beneﬁts whilst looking after young children at home.19 A second channel through which the reform effort in Germany may have affected domestic aggregate demand is through the impact of actual and prospective wage restraint in depressing consumption expenditure. As noted in the introduction, German exports have performed well over the period. Recent studies indicate that improved cost competitiveness and the concentration of German export markets in rapidly growing foreign markets have both played a role. There was also an ‘unexplained’ positive trend presumably reﬂecting unmeasured quality improvements.20 In contrast to small open economies where the aggregate demand response to a change in unit labour costs is dominated by its effect in raising net exports, the spread of wage restraint across the economy in Germany where tradables are a much smaller share of the economy will have a bigger depressive effect on consumption demand. More detailed analysis is required of the channels through which wage restraint affects demand – via net exports, consumption and investment. Some lessons for the design of economic reforms in Germany We have argued that the behaviour of aggregate demand can play a role in accounting for persistent changes in unemployment across the advanced economies. Whereas in the standard (closed economy) model, aggregate demand plays an accommodating role in the medium run – expanding or contracting to secure consistency with the supply conditions – the recent experience of the United Kingdom and Germany suggests that swings in aggregate demand have driven medium-run unemployment trends. Although unemployment at its current low rate may not be sustainable in the long run in the United Kingdom as the consequences of the apparent imbalances between the domestic and external sides of the economy force a rebuilding of domestic savings, there seems little doubt that a better functioning supply side has been achieved and with it the prospect of relatively satisfactory labour market performance over the medium run. Equally important was the gradual adoption through the 1990s of a macroeconomic policy framework that actively stabilizes the growth of aggregate demand. Although most attention has been paid to the role of the Bank of England, we have shown that ﬁscal policy has been the key to sustaining aggregate demand in the United Kingdom in the period since 2000. Concerns for policy makers in the United Kingdom are centred on the decline in household savings and the problems with pension provision, the relatively weak innovation performance of the economy and the deﬁcits in education and transport infrastructure. More generally, a substantial number of OECD countries appear to have evolved relatively successful combinations of policies and institutions to sustain

106

W. Carlin and D. Soskice

the growth of aggregate demand in line with a supply side that is consistent with high labour utilization and low unemployment. Thirteen of the twenty OECD economies have an estimated structural unemployment rate of less than 6 per cent in 2004–5: an Anglo-Saxon group (Australia, the United Kingdom, the United States, Ireland and New Zealand), a Nordic group (Denmark, Norway and Sweden) plus Austria, Japan, the Netherlands, Portugal and Switzerland.21 Too little attention has been paid to understanding the variation in supply-side reforms in the ‘successful’ economies, to understanding how aggregate demand has evolved and been managed in successful and unsuccessful ones and how the political system inﬂuences the ﬂexibility with which demand management measures can be used. Although the Nordic countries share features in common that differentiate them from the liberal Anglo-Saxon economies, most notably in their extensive welfare states and in the importance of large wage-setters in the economy, they differ amongst themselves in important ways. Figure 5.11 shows the labour market reforms implemented across seven dimensions for the Nordic countries. As can be seen, the ‘reform packages’ differ markedly across these countries (the scores for Germany are shown for comparison). The variation in reform packages implemented across countries (and even within the Nordic group) and the potential effect of announced reforms on savings decisions and therefore on macroeconomic activity suggest that the impact of a given change in labour market institutions on medium-run unemployment will differ across countries. It is therefore not surprising that the dominant empirical approach of estimating cross-country panel regressions for unemployment, which is based on the assumption of common coefﬁcients on proxies for labour and product market institutions, has proved unrobust. In addition to taking account of cross-country heterogeneity, the proper 8 7 6 5 4 3 2 1 0 1 2

Denmark

Finland

Norway

Labour taxes UB Early ret., validity, old age pensions Working time flexibility

Sweden

Germany

EPL ALMP IR and wage-setting

Figure 5.11 Labour market reforms undertaken between 1994 and 2003: Nordic countries and Germany (source: Annex 3 of Brandt et al. (2005). Each bar represents the number of reform measures implemented).

Reforms in Germany 107 Table 5.8 Model simulation: earned income tax credit and unemployment beneﬁt reform

EITC UB reform

Effect on employment rate (%)

Effect on training (% trained as apprentices)

Effect on quality of employee– employer match (relative to existing baseline)

Life-cycle effect on utility (relative to existing baseline) (%)

+1 points +4–5 points

–7 points –11 points

–4 –3.2

–1.4 +2.5

Source: Adda et al. (2006).

evaluation and hence design of the prospective impact of policy reforms should include their impact on much longer term decisions such as human capital accumulation. As an example, a recent study models the long-term life cycle effects of two different labour market reforms in a structural model implemented using detailed microeconomic data on German men (Adda et al., 2006). The model is one where jobs take time to locate and in which wages depend on the quality of the match between ﬁrm and worker, experience and tenure. The policy comparison is between the introduction of an earned income tax credit (EITC) as in the United States and a reform to unemployment insurance (replacing earningsrelated unemployment insurance with a ﬂat rate) as introduced in the United Kingdom. The exercise reveals that although both policies have their intended positive effect on employment, they have implications for training and job mobility that may not have been anticipated. Under the EITC, it is estimated that training would fall by 7 percentage points, whilst the impact on the employment rate is modest at about 1 percentage point. By contrast, the unemployment beneﬁt reform is predicted to reduce training by more but to also raise employment by more. The intuition is fairly straightforward. To take one example, under the unemployment beneﬁt reform, training rates are depressed for all types of workers by the disappearance of the anticipated increased beneﬁt when unemployed for those who are trained. This outweighs the effect of lower beneﬁts under the policy change in inducing non-apprentices to train. Employment is boosted because unemployment ‘pays less’. In both cases, the estimated quality of the worker-ﬁrm match declines following the implementation of the policy. Since the job match is worth less, workers search less hard.

Conclusions The aim of this chapter is to contribute to the debate about the conventional diagnosis of Germany’s poor economic performance, which focuses on supplyside weaknesses and the need for more vigorous and deeper reforms especially to make low-skill private service sector labour markets more ﬂexible. We can summarize the standard view as follows:

108 •

•

W. Carlin and D. Soskice Medium-run unemployment is determined entirely by supply-side forces. Hence, robust growth and falling unemployment in the United Kingdom is the outcome of supply-side reforms and weak growth and rising unemployment in Germany is the result of insufﬁcient reforms and continued deterioration of the supply side. Labour and product market reforms such as those recommended by the OECD (and undertaken in the United Kingdom) should be implemented with more determination in Germany.

We have argued that there are both theoretical and empirical grounds for pursuing this debate further. On the theoretical side, we showed that the claim that aggregate demand can play an important driving role in explaining persistent weak growth and rising unemployment does not rely on a naïve Keynesian macroeconomic model. Speciﬁcally, we used an extended version of the nowstandard New Keynesian macro model to show how: •

•

•

Shifts in aggregate demand in open economies can move the economy along a range of constant-inﬂation medium-term unemployment equilibria. This is reinforced in a common currency area where the nominal exchange rate cannot change. In this setting, both ‘exogenous’ and policy-related shifts in aggregate demand can contribute to the explanation of within-country trends and cross-country differences in unemployment performance. The evolution of the real exchange rate and the external balance can help to identify whether aggregate supply or demand shifts have been dominant in accounting for changes in unemployment.

Moreover, on the empirical side, we have shown: •

•

Germany has a relatively strong record in carrying through OECDrecommended reforms. There has however been little consequent evidence of growth of employment in low-skill service sectors and there has been an increase in poverty especially at younger age groups. By contrast, it is in high productivity sectors, including services, that the German economy has performed well, especially in exports, and these are largely sectors in which codetermination, vocational education at different levels and coordinated wage bargaining are important.

Taking the cases of the United Kingdom and Germany, we have shown that there is at least prima facie evidence to suggest that aggregate demand factors have played an important role in sustaining growth in the United Kingdom and weakening it in Germany over the medium run. The sharpest contrast is between the strength of the contributions to growth of household and government spending in the United Kingdom and its weakness in Germany, especially in the period since 2000. The real appreciation of the sterling exchange rate is consis-

Reforms in Germany 109 tent with aggregate demand having played a substantial role in the fall in unemployment in the United Kingdom after 1997. The strength of German export performance and the industry pattern of employment growth both point towards a predominantly demand rather than supply-side explanation of rising German. Our analysis points towards two key questions for the reform debate in Germany: 1 2

What does a reform package for Germany look like that will achieve improved macro performance? How can political support for such a package be found?

The complementarity between the speciﬁcities of the labour market, the training system and the welfare state indicates that only a reform package that takes these features into account is likely to have its intended employment boosting impact without depressing aggregate demand – e.g. via higher savings. It would also appear that there is scope for increasing ﬂexibility in the labour market without adversely affecting consumption and income distribution by removing penalties on second earners in the household and introducing reforms to increase the incentives for women to combine employment with child rearing. As part of a research agenda to contribute to a reform package, two speciﬁc issues emerge from this chapter. The ﬁrst question is how to make it easier for women to move between working at home, especially raising very young children, and market sector employment. Currently, women lose substantially by having children and then returning to work, since they do so at greatly reduced earnings and often conditions. The two well-known consequences are that young women with good career prospects are reluctant to have children, and women who have children and are engaged in home-working are reluctant to return to the market; these consequences are, as is also well known, compounded by the absence of readily available high quality child care. Where these problems do not apply as in the Nordic countries, it seems likely that the ability to move relatively costlessly between working at home and in the market acts as an additional mechanism to reinforce the standard insurance provided by the welfare state. This seems an important area of research in Germany, since it combines the possibility of increasing consumer expenditure and increasing the reproduction rate. But there are political economy problems: ﬁrst, attempting to develop guarantees in the private sector is difﬁcult since the interests of male-dominated works councils and sector-based unions are against them (damaging company proﬁtability) – this contrasts with confederal unionization in the Nordic countries. Second, if the public sector was developed as in Scandinavia and Finland, a major (political economy) problem would be to avoid using this as a means to mop up unemployment of low-skilled women with low educational attainments – for that could quickly bring such a system into disrepute. The second item in a research programme would be to examine the claim that ﬂexible labour market reforms designed to promote low income service sector

110

W. Carlin and D. Soskice

jobs have sharply dysfunctional effects in an environment with insufﬁcient aggregate demand. Such effects may include: 1 2

the growth of poverty and limited employment growth in these sectors; the knock-on effects of poverty on lower end educational and vocational training incentives and attainment.

We believe that it is through a greater understanding of these interactions between macroeconomic questions of insufﬁcient aggregate demand and microeconomic questions of labour markets and training that it may be possible to fashion a programme of reforms tailored to the speciﬁcities of the German political economy.

Appendix 5.1 Institution variables

2

2

1

1

05

00

20

95

20

90

19

19 80 19 85 19 90 19 95 20 00 20 05

3

Germany

19

80

United Kingdom

3

19 80 19 85 19 90 19 95 20 00 20 05

Employment protection legislation

Germany

19

19

20

20

19

19

19

05

20

00

20

95

30

90

30

85

40

80

40

85

United Kingdom 50

19

Union density

Germany 50

United Kingdom

1

.5

.5

0

0

19 8

85

20 30

15 20

Germany 05

40

00

25

20

50

95

Germany

20

30

19

05

20

00

95 20

1

90

1

19

2

90

2

85

3

19

3

85

4

19

80

4

19

19

05

20

00

95

20

19

90

19

85

80

19

19

Product market regulation

5

19

80

19

05

20

00

95

20

19

90

19

19

80

19

Average replacement rate 5

19 85 19 90 19 95 20 00 20 05

0

1

19 80 19 85 19 90 19 95 20 00 20 05

High corporation

Germany United Kingdom

United Kingdom

United Kingdom

112

W. Carlin and D. Soskice

20

19

19

20

85 19 90 19 95 20 00 20 05

30

80

30

80 19 85 19 90 19 95 20 00 20 05

United Kingdom 40

19

Tax wedge

Germany 40

Source: Bassanini and Duval database 2006.

Appendix 5.2 Scoring of labour market reforms introduced in Germany 1994–2004. Source: Annex 3 of Brandt et al. (2005). The following are taken directly from the above source. EPL 1994–9 ‘Regular contracts’ – company size limit for dismissal protection lifted from 5 to 10, but this was reversed, score: 0. ‘Temporary contracts’ – maximum duration and permissible renewal frequency increased with a resulting change in the EPL-indicator of 1.5, score: +3; total score for 1994–9 equals 3. 2000–4 ‘Regular contracts’ – size limit for dismissal protection lifted from 5 to 10, this reform is not reﬂected in the EPL indicator, score: +1. “Temporary contracts” – maximum age for restrictions on the renewability of ﬁxed-term contracts decreased and the maximum duration for contracts through temporary work agencies eased further with a corresponding decrease of the EPL indicator of 0.2, score: +1; total score for 2000–4 equals 2. Overall score for the 1994–2004 period equals 2.5; the overall score expressed as a percentage of the maximum score (=15, that is a total of 6 for a change in the EPL indicator bigger than 0.6 for both temporary and regular contracts plus 9 extra points) takes a value of 16.67 per cent.

Reforms in Germany 113 Unemployment beneﬁts 1994–9 Work availability requirements were tightened, score: +1; eligibility conditions tightened, score: +1; concerning the beneﬁt replacement rate, crediting of redundancy payments against unemployment beneﬁts was introduced and reversed in 1999, score: 0. 2000–4 Wage base for beneﬁt calculation extended resulting in an average increase of the replacement rate of 1.8 per cent, but in 2004, it was decided to reduce beneﬁts sharply beginning in 2005, to capture the latter reform measure a score of +1 is assigned; beneﬁt duration lowered: +1; work availability criteria tightened: +1. Overall score for the 1994–2004 period: 2.5, as a percentage of the maximum score (=13): 19.2 per cent. Work time ﬂexibility 1994–9 One point in the category ‘Flexibility of working time arrangements’ for the introduction of working time accounts. 2000–4 Employees obtain a right to unilaterally transform a full-time job into a part-time job, score: 0, because it is not clear that the possibility to transform full-time into part-time jobs without the employer’s consent eases constraints in a way that more job creation is to be expected. Overall score for the 1994–2004 period: 0.5; as a percentage of the overall score (=3): 16.7 per cent. Early retirement 1994–9 Entry ages to different retirement schemes gradually increased, involving a drop in the average implicit tax of 1.7 percentage points, score: 1; introduction and subsequent suppression of reforms to the sick pay and the disability pension system, score: 0.

114

W. Carlin and D. Soskice

2000–4 Early retirement involves discounts and unemployment beneﬁt pathway to early retirement is phased out progressively as the exemption of older workers from job search requirements expire, involving an average reduction in the implicit tax of 11.5 percentage point, score: +2. Overall score for the 1994–2004 period: 1.5; as a percentage of the overall score (=6): 25 per cent. Wage formation 1994–9 One point in ‘Wage determination’ for an increase in plant-level agreements over the 1990s; 1 point in “Use of opt-out clauses”. 2000–4 No action, score: 0. Overall score for the 1994–2004 period: 1; as a percentage of the maximum (=13): 9.1 per cent. Active Labour Market Policies 1994–9 Evaluation efforts stepped up, score: +1; ALMPs were targeted more on the long-term unemployed, the young and older workers, with some of these programmes being new, consequently 1 point was assigned in the category ‘Targeting of ALMPs’ together with a point in the category ‘Extent of ALMPs’; ALMPs were re-oriented towards encouraging job search while inefﬁcient job creation programmes were scaled down, for this measure a point was assigned in the category ‘Integration and restructuring of ALMP activities’ and a point in the category ‘Extent of ALMPs’ for enhancing the job search function of the PES (resulting in a total of 2 points in this category); one point was assigned in the category ‘Re-qualiﬁcation via ALMPs’ for abolishing re-qualiﬁcation via training programmes. 2000–4 Evaluation efforts stepped up once more, score: +1; for proﬁling, corresponding to an enhancement of the job search function of the PES, for individual action plans to become mandatory in 2005 and for efforts to introduce early activation, a score of 4 is assigned in the category ‘Extent of ALMPs’; a supplementary point in ‘Targeting of ALMPs’ is assigned for early activation of young jobseek-

Reforms in Germany 115 ers; in ‘Contestability of PES’ 1 point for introducing the possibility to outsource placement to private agencies is assigned; in ‘Integration of job counselling, ALMPs and various beneﬁt agencies’, 2 points are assigned for the centralisation of the responsibility for the long-term unemployment in one agency that will be introduced in 2005; a further point is assigned in the category ‘Re-qualiﬁcation via ALMPs’ for abolishing the possibility to re-qualify. Overall score for the 1994–2004 period: 7.5; as a percentage of the maximum score (=13): 57.7 per cent. Taxes on labour income 1994–9 ‘Overall taxes on labour income’ – labour tax decreases along with some increases and some decreases in social charges with a resulting increase in the tax wedge by 0.8 per cent, score: –1; ‘Taxes for low income’ – increases in tax allowances and in the lowest tax rate, social charges for ‘Minijobs’ introduced with a resulting decrease of the average of the three tax wedges of 1.1 per cent, score: +1. 2000–4 ‘Reduction on taxes in labour income’: tax rate decreases across all income ranges with a resulting decrease in the tax wedge of 0.8 per cent, score: +1; increase of the earnings ceiling for ‘Minijobs’ and rebates of social charges for income ranges just above that ceiling with a resulting decrease in the average of the three tax wedges for low income of 0.8 per cent, score: +1. Overall score for the 1994–2004 period: 1, the overall score expressed as a percentage of the maximum score (=8): 12.5 per cent which is the reform intensity indicator.

Notes 1 We are grateful for the excellent research assistance provided by James Cloyne and Etienne Hans and for useful discussions about these questions with Andrew Glyn, Bob Hancke and Matthew Harding. Andrea Bassanini, Romain Duval and Luca Nunziata have kindly provided data. 2 The poverty rate in Germany in 2001 was 8.9 per cent and it increased by 0.6 percentage points from 1994 to 2001, with the increase concentrated in the age group from 0 to 40 (Förster and Mira d’Ercole, 2005: Annex Table A7). 3 OECD Economic Outlook Annex Table 45. 4 See also Cotis (2006) for similar results using OECD Main Science and Technology Indicators Database, R&D database and Patent database. 5 Eventually, private sector or policy responses to external imbalances associated with different medium-run equilibria will tend to drive the economy to a long-run equilibrium, but these forces appear to be weak, which allows medium-run equilibria to persist.

116

W. Carlin and D. Soskice

6 For a more extensive presentation of some of this material, see Carlin and Soskice (2006). 7 For early treatments, see Rowthorn (1977); Carlin and Soskice (1990); Layard et al. (1991). 8 This assumes that the nominal exchange rate is ﬁxed at 1; since we are analysing Germany as a member of EMU we will maintain this assumption – though note this does not take account of variations in the value of the Euro itself. 9 For a simple exposition of how this works, see Carlin and Soskice (2005). 10 Since the slope of the AD curve depends on the combined effect of real wages on net exports, consumption and investment, a full analysis would also include the impact of wage restraint on investment. 11 For a concise presentation of the evidence of the effects of reforms for the United Kingdom, which includes comparative data for Germany, see Card and Freeman (2004). For an extensive analysis of performance and reforms in Germany, see Siebert (2005); for a view along similar lines from ‘the outside’, see Heckman (2002). 12 The OECD estimated the overvaluation of house prices in the United Kingdom in 2004 at 32.8 per cent (using the price to rent ratio, which was most recently at its 35year average in 2000) and undervaluation in Germany in 2004 at 25.8 per cent (also relative to 2000 – although the price to rent ratio has been declining there since the 1980s) (OECD Economic Outlook 78, 2005: 136). 13 For a detailed cross-country analysis of the role of demand in accounting for employment differences in services, see Glyn et al. (2004). 14 Updated in Howell et al. (2006). 15 Also reported in OECD Economic Outlook (2006). 16 For a trenchant critique of macro studies of this kind, see Heckman and Pagés (2003). 17 Calculated from OECD Economic Outlook Database, June 2006. 18 Some preliminary evidence that announced reforms to the pension system have measurable effects on savings (and labour supply) is provided in an analysis of German data (see Giavazzi and McMahon, 2006). 19 For more details including a comparison with Sweden, see Vikat (2004). 20 Allard et al. (2005). 21 OECD Economic Outlook 79 Appendix Table 22.

References Adda, J., Dustmann, C., Meghir, C. and Robin, J.-M. (2006). ‘Career progression and formal versus on the job training’, mimeo, UCL. Allard, C., Catalan, M., Everaert, L. and Sgherri, S. (2005). ‘Explaining differences in external sector performance among large Euro area countries’, IMF Working Paper. Allsopp, C. and Vines, D. (2005). ‘The macroeconomic role of ﬁscal policy’, Oxford Review of Economic Policy, Vol. 21, No. 4, 485–508. Baker, D., Glyn, A., Howell, D. and Schmitt, J. (2005). ‘Labor market institutions and unemployment: a critical assessment of the cross-country evidence’, in D.R. Howell, (ed.), Fighting Unemployment: The Limits of Free Market Orthodoxy, OUP: Oxford. Bassanini, A. and Duval, R. (2006). ‘Employment patterns in OECD countries: reassessing the role of policies and institutions’, OECD Working Paper 486. (Also summarized in OECD Economic Policy 2006, Chapter 7.) Blanchard, O. (2005). ‘European unemployment: the evolution of facts and ideas’, MIT Working Paper 05–24. Brandt, N., Burniaux, J.-M. and Duval, R. (2005). ‘Assessing the OECD Jobs strategy: past developments and reforms’, OECD Working Paper 429.

Reforms in Germany 117 Carlin, W. and Soskice, D. (1990). Macroeconomics and the Wage Bargain, OUP: Oxford. Carlin, W. and Soskice, D. (2005). ‘The 3-equation New Keynesian model – a graphical exposition’, Contributions to Macroeconomics, Vol. 5, No. 1, Article 13. www.bepress.com/bejm/contributions/vol5/iss1/art13. Carlin, W. and Soskice, D. (2006). Macroeconomics: Imperfections, Institutions and Policies, OUP: Oxford. Card, D. and Freeman, R. (2004). ‘What have two decades of British Economic Reform Delivered?’ in D. Card, R. Blundell and R.B. Freeman, eds, Seeking a Premier Economy: The Economic Effects of British Economic Reforms, 1980–2000, University of Chicago Press: Chicago, IL. Cotis, J.-P. (2006). ‘Economic growth and productivity’. Annual Conference, Government Economic Service, Nottingham, July 2006. Esping-Andersen, G. (1990). The Three Worlds of Welfare Capitalism, Polity Press: Cambridge, UK. Förster, M. and Mira d’Ercole, M. (2005). ‘Income distribution and poverty in OECD countries in the second half of the 1990s’, OECD Social, Employment and Migration Working Paper No. 22. Giavazzi, F. and McMahon, M. (2006). ‘Waiting for reforms that never come: saving and work in Germany’, mimeo, LSE. Glyn, A., Salverda, W., Moeller, J., Schmitt, J. and Sollogoub, M. (2004). ‘Employment differences in services: the role of wages, productivity an demand’, DEMPATEM Working Paper 12. (Forthcoming in W. Salverda, M. Gregory and R. Schettkat, (eds), Services and Employment: Explaining the US European Gap, Princeton University Press: Princeton, NJ.) Hall, P.A. and Soskice, D., (eds) (2001). Varieties of Capitalism: the Institutional Foundations of Comparative Advantage, Oxford University Press: Oxford. Hallerberg, M., Strauch, R. and von Hagen, J. (2001). ‘The use and effectiveness of budgetary rules and norms in the EU States’, Report for the Dutch Finance Ministry. Heckman, J. (2002). ‘Flexibility and job creation: lessons for Germany’, NBER Working Paper 9194. Heckman, J. and Pagés, C. (2003) ‘Law and employment: lessons from Latin America and the Caribbean’, NBER Working Paper 10129. www.nber.org/papers/w10129. Hoem, J.M. (2005). ‘Why does Sweden have such high fertility?’ Demographic Research, Vol. 13, Article 22, 559–572. www.demographic-research.org/Volumes/ Vol13/22/. Höpner, M. (2003). Wer beherrscht die Unternehmen? Shareholder Value, Managerherrschaft und Mitbestimmung in Deutschland, Frankfurt a.M.: Campus. Howell, D., Baker, D., Glyn, A. and Schmitt, J. (2006). ‘Are protective labor market institutions really at the root of unemployment? A critical perspective on the statistical evidence’, mimeo, New York. Iversen, T. (2005). Capitalism, Democracy and Welfare, Cambridge University Press: Cambridge, UK. Krugman, P. (2005). ‘Is ﬁscal policy poised for a comeback?’ Oxford Review of Economic Policy, Vol. 21, No. 4, 515–523. Layard, R., Nickell, S. and Jackman, R. (1991). Unemployment: Macroeconomic Performance and the Labour Market, OUP: Oxford. Lijphart, A. (1984). Democracies: Patterns of Majoritarian and Consensus Government in 21 Countries, Yale University Press: New Haven, CT.

118

W. Carlin and D. Soskice

Nickell, S., Nunziata, L. and Ochel, W. (2005). ‘Unemployment in the OECD since the 1960s. What do we know?’ Economic Journal, Vol. 115, January, 1–27. Rowthorn, R. (1977). ‘Conﬂict, inﬂation and money’, Cambridge Journal of Economics, Vol. 1, 215–239. Siebert, H. (2005). The German Economy: Beyond the Social Market, Princeton University Press: Princeton, NJ. Vikat, A. (2004). ‘Women’s labor force attachment and childbearing in Finland’, Demographic Research Special Collection, Vol. 3, No. 8, 175–211. Vogel, S. (2006). Japan Remodeled, Cornell University Press: Ithaca, NY. World Economic Forum (2005). The Global Competitiveness Report 2004-2005, Palgrave Macmillan: Houndmills, Basingstoke, Hampshire, UK.

6

Exportweltmeister – so what? Better goals for German foreign economic policy Adam S. Posen1

Introduction For more than 50 years, exports have been seen as the primary driver of German economic growth, and their promotion has been the nearly sole focus of German foreign economic policy. That this focus on exports has persisted through periods both of economic success, as during the 1950s and 1960s, and of economic decline, as during the last 25 years – and thus cannot be said to be clearly associated with one or the other – seems to have escaped notice. Instead, every year, the rankings of which countries sell the greatest volume of exports are compiled by German commentators as though they were as important in their own way as the rankings from the Fussballweltmeisterschaft, with Germany expected to be at the top of the list. Yet, unlike the pursuit of the World Cup, there is good reason for Germany to give up this contest. Germany’s pursuit of export competitiveness has been at best a deceptive distraction from the country’s underlying economic problems, if not a complete waste of effort that promotes distortions at home. Neither a country’s share of exports in GDP nor its relative rank in world export league tables has a signiﬁcant positive effect on its economic or productivity growth. What previously had been thought of as the connection from trade openness (not just exports) to growth has more recently been shown to be less powerful than was previously thought, once the combination of the effect of growth on trade, rather than vice versa, and of the impact of geography was taken into account. Moreover, many of the beneﬁts of openness can be ascribed to the presence of generally beneﬁcial liberal economic institutions that are also associated with the absence of trade protection.2 In any event, the remaining beneﬁcial effects of trade on wealth and growth are associated with openness, not with exports, net or total. In this regard, at least as troubling as the misguided export focus has been the associated near-mercantilist belief by some German ofﬁcials that it is justiﬁable to block further European economic integration (such as the European Union’s [EU] Services or Takeover Directives) so long as German (net) export totals are not threatened – or even that blocking such integration is justiﬁed to maintain the level of exports. If anything, however, the reverse is true for Germany and all developed economies: the economic beneﬁts of globalization arise out of

120

A.S. Posen

cross-border economic integration and the competitive pressure it puts on domestic companies through imports, expansion of variety, and capital mobility. Exports, in net or absolute terms, are far from vital to growth on their own merits. Similarly, it matters what an advanced country exports and on what basis, rather than exports simply being uniformly a good thing. A national economic strategy based on reducing real wages to make the current mix of a country’s exports more price competitive – as underlies the anemic current (2006–07) ‘export-led’ recovery in Germany through relative wage deﬂation vis-à-vis the rest of the eurozone – will not lead to sustainable growth any more than repeated nominal depreciations would. Export growth achieved through increases in productivity and creation of new products or markets would be far more beneﬁcial to German workers (and thus to consumption and investment) as well as far more lasting. Yet, the obsessive focus on exports in Germany has led to selfdeluding welcoming of the recent ‘export-led recovery’, despite an obvious lack of stimulus to domestic demand or its sustainability as a source of growth.3 Germany needs to reconceive its foreign economic policy to better serve the welfare of its citizens. This will require some radical rethinking of the relative importance of exports and of the distorting protection of German businesses that result from that misprioritization. The success of some German companies in exporting has in fact blinded German citizens and policymakers to the problems of the German corporate sector, thus allowing too much blame for German underperformance to be put on labor markets and the public sector. Simultaneously, Germany’s trade performance has hidden the relative absence for many incumbent German Mittelstand businesses of competition from new entrants (foreign but even more so domestic) or of pressures from capital markets for proﬁtability. The result has been low returns on capital, failure to consolidate and take advantage of economies of scale, minimal adoption of innovative and best practices despite high investment in R&D, and an insufﬁcient of reallocation of productive factors from older to newer sectors. Ultimately, this has produced a declining overall rate of productivity and per capita income growth in the German economy as compared to its peers over the last 25 years (not just to the US productivity surge of late).4 The main sources of this arrested development in the German corporate sector are domestic, lying in the failures of German corporate governance and of the country’s partly public-fragmented banking system. They reveal themselves, however, in the limited integration of German ownership, production, and capital markets with the global economy, in the persistence of the limited number of sectors in which German exporters are successful even as those sectors move down the global value chain, and the relative under-representation of German multinationals among the ranks of leading global companies and particularly in new technological and service sectors. Germany’s foreign economic policy should shift from reinforcing these patterns to aiding in removing them. They can be alleviated in part through a more aggressive pursuit of global economic integration rather than of Exportweltmeisterschaft as a goal for German foreign economic policy.

Exportweltmeister – so what? 121 Such a challenge to accepted German norms about the virtues of exports becomes all the more critical given today’s integration of China, India, the 10 new EU members, and other emerging markets into the global supply chain. That environmental shift simultaneously increases the competitive pressure on lower productivity businesses and diminishes much of the political support for continued international economic openness in the developed world.5 In fact, just as Germany has had to reassess its role in the EU and in world politics with regard to security issues, now that the Cold War is ended and uniﬁed, Germany has become a ‘normal country’ 60 years after World War II, so too Germany and the world can beneﬁt from Germany playing a leadership role in international economic affairs on behalf of greater economic integration.6 Relatively passive support for the global trading regime, with occasional pushes back against the demands of French (and now Polish) agricultural interests whenever multilateral trade negotiations reach impasses, which represents the minimum and recurring German role, will no longer sufﬁce to assure German and European economic well-being.

It is openness and perhaps imports, not exports, that matter for growth The best motivation to guide a country’s foreign economic policy is drawing attention to the development of a nation’s real per capita income. More than any other macroeconomic variable, this is the one that most closely approximates individual welfare.7 And over any time frame longer than a year or two, real per capita income growth is a direct function of productivity growth. With rare exceptions (i.e. the United States of the last decade), real wages tend to track productivity growth in democracies if not all economies. In a globalized world – that is a world economy where trade in goods and services is relatively free, ﬁnancial capital and industrial production moves across borders, and technology, best practices, and competitive pressures accompany and direct these ﬂows – the path to greatest productivity growth is to take advantage of the specialization and integration these conditions allow.8 So doing is much more about allowing in imports, adopting and utilizing the best practices and technology regardless of source, shifting production to the locales where the work force and wage level best suit it, and so on, than about export promotion or trade balances. Most of all, it is about putting competitive pressure on a country’s own businesses to attract capital, to adapt to changing conditions, and to have to earn market share at home and abroad.9 As such, ‘export competitiveness’ is no longer the most appropriate term as much as ‘encouraging competition at home from all sources, foreign and domestic’ but that will never be as politically popular or catchy a phrase. In today’s globalized world, there is no contradiction between sustainable strong growth and running a trade deﬁcit or closing lots of domestic businesses or allowing foreign ownership in many industries. If anything, such imbalances and growth are positively correlated since they can indicate that the country in

122 A.S. Posen question has more attractive investment opportunities than it can fund with internal savings and that its consumers (its citizens) are receiving maximum purchasing power and choice.10 Australia and the United States, both in the last 15 years and in the nineteenth century, are examples of this connection; Germany and Japan in the last two decades are examples of the reverse, where export ‘success’, declining productivity growth, and economic stagnation all co-existed for a long period. More formal econometric investigations of the links between exports, openness (by various measurements), and per capita income growth have spawned a large empirical literature. Dollar (1992), Dollar and Kraay (2004), and Frankel and Romer (1999) estimated that the beneﬁts to growth of economic policies that promoted trade were substantial, even when controlling for the inﬂuence of geography on trade levels and for causality running in the other direction (i.e. economic growth causing increases in the volume of trade).11 More recently, those results have come into question as less than robust and still overestimating the inﬂuence of trade policy per se on growth as opposed to the beneﬁts of more liberal and market-oriented, i.e. more open, economic policies more generally (Rodriguez and Rodrik 2001; Baldwin 2003; Hallak and Levinsohn 2004). The most recent studies employing state of the art econometrics (Lee et al. (2004) notable among them) ﬁnd that there is a positive relationship between openness and growth, but it is signiﬁcantly smaller than previously thought once reverse causality, geography, and general quality of institutions are controlled for. All of those results, though, it must be emphasized, concern openness, not exports. The evidence for the growth-promoting attributes of exports in and of themselves is uneven at best; the evidence on imports’ beneﬁts is stronger and more robust than that for exports’. There are a number of reasons why this makes sense: the competitive pressure from imports inducing efﬁciency gains in domestic ﬁrms, importation of technological innovations from abroad embodied in imports, and the increasing variety of products available both as inputs to production and as consumption goods, all of which are dynamic long-term gains, in addition to the static beneﬁts from lower prices associated with imports.12,13 Even setting aside the long-standing German misperceptions that exports drove the postwar recovery (rather than capital stock rebuilding and convergence) and that German companies must be highly productive if they are exporting (reasoning backward), there is little cross-national evidence that exports are in and of themselves beneﬁcial for growth over the long term.14 In an authoritative recent article, Sala-i-Martin et al. (2004) famously ran ‘a million regressions’ of generated sets of simulated country multi-year per capita GDP growth rates on various potential correlates of growth. These samples were generated from the distributions implied by the sample of actual country growth rates on which we have data in the postwar period.15 The result of the exercise was to produce a list of explanatory variables for growth in rank order of likelihood of robust signiﬁcant impact. And what one ﬁnds is that a measure of ‘Years open 1950–94’ comes in ranked fourteenth and, as discussed in several places in the literature, that measure of openness is highly correlated with both

Exportweltmeister – so what? 123 institutional quality and geography (which are themselves correlated with each other some argue), so it is not a clean or pure measure of openness. Real exchange rate distortion, which the authors measure with something that tracks undervaluation, does come in positively associated with growth, but in twentieth place, and the authors suggest that anything after seventeenth or eighteenth place in their rankings should not be taken seriously as a determinant of growth. More direct measures of exports speciﬁcally and variants thereof all come in much further down the list and can therefore basically be ignored according to this study as independent determinants of growth. These results are consistent with earlier comprehensive empirical work with cross-national datasets on the determinants of growth (e.g. Barro 1998; Bosworth and Collins 2003) that do not ﬁnd exports to be a signiﬁcant explanatory variable.16

German experience shows the costs of putting the export cart ahead of the productivity horse To the degree there is a strategic choice for the governments of advanced economies in today’s globalized world, it is analogous to that of many businesses: does one want to compete on price or on quality? In other words, does the economy specialize in high-value-added products and industries where wages are high as are technological requirements or does the economy concentrate on highvolume-commodiﬁed products and industries where cost savings (including on wages) are the key to success? The enormous and sustained increase of the global labor supply accompanying the integration of China, India, and the former communist block into the world economy only increases the starkness of this choice, by putting more price pressure on the low end, but does not fundamentally alter it.17 It overstates the case to present this as an one-or-the-other central decision to be taken – even the most advanced economies will have some low-wage commodity sectors co-existing with high-tech industries and even high-value-added products will be subject to cost pressures from time to time – but it is valid to emphasize that high exports are not a sufﬁcient statistic to indicate a successful economic policy if they are gotten through the lower-return path. No example better illustrates the costs to an economy of distraction by export competitiveness than Germany in recent years. The German corporate sector desperately needs competitive pressure on it and reform of its corporate governance but escapes those changes by insisting that large quantities of exports mean the fault lies elsewhere in the economy, like high taxes or wages. In fact, the very parts of the German economy that are most protected by overregulation, publicly subsidized ﬁnancing, and unaccountable corporate governance (albeit not many tariff barriers per se) – the much vaunted Mittelstand – use the export success of some of its ﬁrms to justify its protections. Yet, for all their exports, the resulting lack of consolidation or technical change in these sectors drives down productivity growth and returns to capital throughout the German economy. The large and truly efﬁcient German multinationals increasingly move their production abroad, and the service sector stagnates.18

124

A.S. Posen

Consequently, Germany’s successful export industries remain largely the same ones as they were 40 years ago (bulk chemicals and dyes, large electrical goods and appliances, machine tools, and autos and auto parts), while global technological progress and competition from emerging markets means that these sectors have moved down the value chain.19 The dysfunctions of the German corporate sector also mean that there are almost no German ﬁrms – and thus few German workers and investors – that have emerged in what are today’s growing high-technology and service sectors; Mann with Kirkegaard (2006), for example, ﬁnd one Germany company (SAP) among the top 25 software and IT services providers worldwide, and no German companies among the top 25 IT hardware producers. Such a focus on the currently exporting companies and the preservation of their current ownership structures also shows up in unexploited scale economies for German companies, because growth would require greater external ﬁnance and thus loss of managerial freedom from accountability – despite the common assumption that German multinationals are dominant both in Germany and in the EU, Germany actually has 25 percent fewer large companies in Europe than consistent with its share of the EU economy.20 By focusing on export success rather than productivity, Germany has brought about arrested development in its corporate sector. The costs of such export excuses substituting for constructive policy are not just long-term erosion of productivity and wage levels – dire though those are – the competitiveness focus also damages medium-term macroeconomic performance. Since adopting the euro, Germany has had an inﬂation rate well below the eurozone average, with the consequence that the country has suffered real wage deﬂation compared to its European trading partners. Some deemed this cause for celebration, not suffering: not only the German press, but The Economist magazine on its cover in August 2005, trumpeted the forecast Germany would have an export-led expansion and that would lead ﬁnally to a recovery in Germany it was claimed (Jansen (2005) gave a similar view from the European Commission). Given the low productivity of Germany during its lasting recession, however, this means that the change in German unit labor costs was both coming at some cost to domestic demand and could not be a source of sustained growth (Posen 2005d). And a year later, exports had risen, but, as recent Bundesbank and IMF studies make clear, that rise was overwhelmingly due to growth in Germany’s export markets, not to improvements in price competitiveness.21 If anything, the econometric evidence is that the impact of real exchange rates – and thus of changing ‘wage competitiveness’ – on German exports has been declining.22 More importantly, this export boomlet was not enough to overcome the drag on German consumption from several years of declining wages and purchasing power, nor was it sufﬁcient to induce a signiﬁcant increase in long pent-up German corporate investment.23 And once the export cycle had peaked along with foreign growth in mid-2006, Germany had little to show for relative wage deﬂation: German unemployment remained over 10.5 percent (4.4 million) in July 2006, GDP growth was universally forecast to be lower in 2007 than in

Exportweltmeister – so what? 125 2006 – and thus having peaked at the far from blistering pace of 2.0 percent real growth – and wage declines have substituted for innovation, entry into new sectors, or reform in the corporate sector. Yet, German policymakers, businesses, and the public were content to wait for this slow process of relative wage deﬂation to take hold over several years rather than to actively pursue policies that would have enhanced productivity growth in the same period.24 Real wage compression in Germany has proven no more a sustainable path to growth today than the nominal devaluations pursued by the United Kingdom and Italy intermittently from the 1970s to the mid-1990s.25 A dozen years ago, Krugman (1994) famously called competitiveness ‘a dangerous obsession’ in the US context. In fact, in every decade and country, a focus on export competitiveness distracts policymakers and the public from a more beneﬁcial emphasis on productivity. As a result, as seen in Germany, policy choices are made that tend to erode living standards, because if exports are the public criterion of economic success, policymakers can meet that goal quickly and easily only by selfdestructive means. Foremost, export promotion usually leads either to depreciating a country’s nominal exchange rate, thus eroding the purchasing power and the accumulated wealth of the citizenry, or to depressing wages in export sectors to drive down the real exchange rate, either directly or through relative deﬂation vis-à-vis trading partners, thus cutting real incomes and domestic demand – as seen in Germany today. It is an empirical question depending upon the industries, cyclical conditions, and behavior of consumers in a given country whether such strategies are even worth it in terms of net growth stimulus in the short run, and Germany seems to be one more example where the answer is no. But there is no evidence that such policies offer long-run beneﬁts, especially if the central bank can be expected to counteract inﬂationary pressures from rising import prices (as the Bundesbank and now the ECB would). If imports are not a sufﬁcient share of the economy to have a noticeable inﬂationary effect when depreciation occurs, or they are priced to the depreciating country’s market such that their sales generated do not rise much in response to depreciation, then net export growth is unlikely to be of great enough magnitude in response to depreciation to be a meaningful growth strategy in any event. Second, as seen in Germany, having exports as a policy goal leads easily to policies subsidizing or protecting exporting companies (mostly through domestic regulation rather than tariffs), thus distorting investment decisions and locking in old technologies and businesses at the expense of new entrants. In more technologically progressive and occasionally European dress, export promotion is associated with promoting national champions, thus increasing both wasteful public spending and the costs to domestic households and businesses of those champions’ products. The Galileo satellite program to mimic GPS is but one recent example. German resistance to the EU Takeover Directive, discussed further in the next section, is in part due to the rhetorical success of entrenched managers and directors of German companies rallying support for what would keep exporters in German hands and thus their products in the

126

A.S. Posen

national export totals – and thus either foils shareholders holding those managers and insiders accountable or gains from economies of scale due to crossborder mergers. In contrast, the United States for example has grown strongly since the early 1990s by generally encouraging a strong dollar, foreign competition for and investment in the US corporate sector, increasing openness to imports, and private investment in R&D (rather than picking speciﬁc winner technologies or champions) – all of which was only possible with a healthy disregard of export performance.26 Closer to the German home, Ireland, Denmark, the United Kingdom, and the Netherlands have all had similar priorities to those of the United States for foreign economic policy of late and enjoyed similar sustained improvements in growth rates. Note that it is these governments’ common commitments to pursuit of openness rather than of exports and to putting competitive pressure on domestic corporations rather than to implementing the so-called competitiveness policies – that is, it is these countries’ engagement with the global economy – which these nations share and Germany does not. It is not some neo-Dickensian absence of welfare state provision that is the common thread among these successful economies.27 On the other side, Italy has joined Germany in export-obsessed foreign economic policy: relying on exchange rate developments (albeit more nominal than real, with little change in wage restraint since the mid-1990s) and external sources of growth; being mercantilist with regard to foreign investment (i.e. accepting Italian companies buying foreign ones, but opposing foreign takeovers of Italian ones); and perhaps most importantly trying to defend by whatever means (including demands on EU trade policy) employment and investment in its traditional but increasingly lower-end export industries such as shoes, textiles, and ceramics – and Italy and Germany have been and remain together at the bottom of EU and OECD average GDP growth rates as a result.28 It should be clear to German policymakers why exports should be replaced with productivity and wage growth as the goal of foreign economic policy.

The EU is best used for micro-integration, not macro-rules Germany has been the motor of European integration for decades, since the European Coal and Steel Agreements were ﬁrst reached after the war. Economic integration has of course recurrently preceded further political integration in Europe throughout the postwar period, and Germany in particular has been willing to shoulder various economic burdens – contributing the most on net to the EU budget, giving up control of its monetary policy to launch the euro, sharing decision-making power in Brussels rather than insisting on voting rules and vetoes consistent with its relative size – to advance the cause of deeper and wider political integration.29 Taking European economic integration on its own terms, though, rather than as a means to a political end, the German contribution to the policy agenda has been more limited. Germany has been the (largely self-appointed) macroeconomic gatekeeper of

Exportweltmeister – so what? 127 membership in the EU. This was seen most obviously in the German insistence on the public debt and deﬁcit, inﬂation, and interest rate ‘Maastricht Criteria’ for admittance to the eurozone in the run-up to EMU, the addition of the Stability and Growth Pact to the Maastricht Treaty meant to maintain ﬁscal discipline after euro entry, and the recent continued application of the full ERM II waiting period and Maastricht Criteria to the ten new EU members before they will be allowed to adopt the euro. The structure of the European Central Bank, its independence and location in Frankfurt, also reﬂect this German control of the macroeconomic agenda, as did the earlier role of the Bundesbank in the 1980s and 1990s in de facto putting price stability ﬁrst as the main pursuit of the ERM and discouraging ‘bailout’ policies whether at the IMF or within Europe in the aftermath of the ERM crises of 1992 and 1995. In tactical terms, this approach had a lot of attractions – Germany had monetary credibility and stability to offer other European economies in bargaining, and leverage would be at a maximum over other countries’ policies just prior to entry in the EU, the ERM, or the eurozone. Substantively, however, this approach reﬂected an ideological penchant for rule-based macroeconomic policy that has been a repeated failure. These rules have either been repeatedly violated (and thus undermined as rules with the attendant political difﬁculties) for good reasons of economic stabilization or when adhered to have been applied too strictly leading to sub-optimal policies.30 Leaving aside their obvious lack of merits as macroeconomic policy frameworks (see the contributions by Solow and Mankiw to this volume), as the German agenda for EU economic policy, the German-encouraged macroeconomic rules fed and currently feed an environment in Brussels which emphasizes national identity and scapegoating, contributed to a public perception of hypocrisy and inefﬁcacy of European institutions, and worst of all concentrates attention on economic policy targets of which governments’ ability to deliver is far from adequate and whose beneﬁts even if delivered are limited.31 In contrast, the most economically beneﬁcial aspects of the EU’s economic agenda have been those emphasizing microeconomic integration of the member economies, the Single Market most signiﬁcant among those initiatives. The policies of the European Commission with regard to competition policy, standardization of regulations, encouraging mutual recognition of standards (and enforcing where those fail to apply), and generally integration of the European economies by breaking down barriers and national preferences have been a major force for liberalization and efﬁciency within Europe and in Germany speciﬁcally. Several studies (e.g. Rogers et al. 2001; Bradford and Lawrence 2004) using differing methodologies have established that price convergence at the producer and consumer level had taken place in the EU during the early 1990s, yielding substantial welfare beneﬁts. Notably, while during the 1980s and ﬁrst half of the 1990s, the pace of integration (on these measures) was faster within in the EU than for European economies with those economies outside of Europe, the relative pace of economic integration intra- and extra-EU reversed in the second half of the 1990s and in more recent years. That is consistent with

128

A.S. Posen

the view that the microeconomic integration efforts were more successful in promoting economic development within Europe than the monetary integration which came from the euro.32 While successive German governments were more or less somewhat favorably inclined toward the Commission’s efforts at microeconomic integration in Europe in past decades, and in fact the Single Market could not have become European policy without some German support, these efforts were actually led for the most part by ofﬁcials from the United Kingdom, Italy, and other European countries. The reluctance of German governments to actively champion such initiatives stemmed from a combination of factors, including some ideological doubts about the public interest in such liberal Anglo-Saxon-seeming policies, an (excessive) sense that Germany already had a relatively globally integrated and open economy given the country’s high levels of exports and imports, and pure pursuit of protection by self-interested lobbies (as in all countries) – as well as the aforementioned focus on macroeconomic rules as the German priority for EU economic policy. To the extent that the European Commission was empowered and shown to be delivering economic results through these efforts, and Germany viewed a stronger Brussels as a means to the overall German goal of deeper European integration – as was the case under the Jacques Delors Presidency, for example – German acquiescence in these initiatives was sufﬁcient to have them move forward, with their economic beneﬁts accruing to Germany along the way. Unfortunately, starting in the mid-1990s, the German approach to Brussels in general and to European economic policy shifted. As described in Posen (2003, 2005b), Germany began to pursue a more intergovernmentalist, less federalist approach to Europe. Of course, this took place against a general political background of economic reform fatigue in Germany and beyond, as well as skepticism about the further deepening of Europe, culminating in the lukewarm response to the new member states and the failed constitution of 2005, so it should not be seen as a considered consistent recalibration of policy. Nonetheless, the shift was real and had several deleterious effects. First, the new German approach directly weakened the ability of the European Commission to pursue a necessarily increasingly intrusive (to be effective) economic integration agenda. With the retreat of the Commission as a force for liberalization, no alternative source for change has emerged in its place.33 Even if one did, there is some question as to how far intergovernmental initiatives could work behind borders the way a Commission directive can without seemed overly politicized. Second, the Kohl and then the Schroeder governments, strongly emphasizing bilateral relations with France, began to engage increasingly in logrolling behavior whereby one country’s violation of the spirit (if not the code) of European economic integration in a given instance would be accepted by Germany in return for a commensurate allowance of similar behavior on the part of Germany. Examples included the mutual non-aggression agreement between France, Germany, and Italy not to impose ﬁnes on each other for violating the Stability and Growth Pact but not to revoke the Pact either; French government

Exportweltmeister – so what? 129 aid to French shipbuilders and steel companies being traded off against German bailouts of Philipp Holzmann and Landesbank Berlin; and most destructively the trade whereby Germany supported an extension of the CAP for France (even after the accession of Poland and other eastern agricultural members) in return for protecting Volkswagen and other poorly governed German businesses from passage of a strong EU Takeover Directive. Third, these German governments responded to persistent economic stagnation (and mounting unemployment) at home by turning generally suspicious of globalization, of which further EU economic integration policies were seen as a part. Given the near mercantilist aspects of the country’s long-standing export obsession, this was not a terriﬁc stretch, especially once European and global economic integration appeared to threaten German companies with low wage competition, new entrants to the home market and in previously non-traded industries, and even foreign ownership. Since most of the major economic initiatives that were coming up the European agenda (the aforementioned EU Takeover Directive; the EU Services Directive which would also involve movement of workers from the new member states into Germany; the EU Financial Services Action Plan, as well as the contemporaneous Basle II Capital Accord for banks, which together would lead to cutbacks in inefﬁcient lending by German public-sector banks) threatened to induce rapid change in long-standing domestic economic arrangements and uncompetitive markets, these were popular targets for German politicians to oppose. German politicians and public opinion are by no means alone in these concerns and pandering, but within EU decision-making, there has been no replacement for German advocacy of principles of European federalism and integration when Germany turned against these efforts. The problem for Germany is that these deeper integration policies, particularly in the areas of corporate governance and ﬁnancial markets, are precisely what Germany needs to raise its sustainable growth rate and to revitalize its corporate sector – it is the absence of competition in the German services market, corporate lending, and ownership which keep down German productivity growth.34 Germany’s intergovernmentalist, logrolling approach to EU economic policymaking in recent years, mounted in opposition to deeper economic integration, has in fact reinforced the worst tendencies of German domestic political economy. Meanwhile, the macroeconomic rules that Germany uses its political attention and capital to promote in Brussels have been increasingly ineffective, either as credible constraints or as sources of better economic performance. Were Germany to reverse its approach to economic policymaking in Brussels, and support a strong European Commission that imposed and implemented directives with liberalizing effects at the micro-level, it would be along with Italy perhaps the greatest beneﬁciary in Europe. Such a reversal would present opportunities for overcoming entrenched interest groups in the corporate sector within Germany, while shifting the responsibility and political accountability for breaking down their privileges away from elected German ofﬁcials; moreover,

130

A.S. Posen

such a move could be done in the name of deeper European integration, thus consistent with some of Germany’s core postwar political values. In 2001–02, the German Finance Ministry, under the leadership of Staatsekretar Caio KochWeser, demonstrated the potential of such a strategy by encouraging a ruling by the European Commission that Germany’s Landesbanken could not continue to have their cost of capital subsidized by state guarantees. Though this ruling would not be implemented until the start of 2005 – and the inclusion of that delay was accompanied, if not motivated, by much ostensible reluctance on the part of the Schroeder government for the sake of the domestic political audience – it has yielded meaningful increases in the cost of capital to Sparkassen and thus down the line to smaller non-ﬁnancial businesses in Germany, thus raising returns and investment efﬁciency.35 Unfortunately, this example proved not only the exception but that last known instance of such an approach by the German government. The opportunity for Germany to strengthen its own economic performance by revitalizing European economic integration remains, however. If anything, the German government’s bargaining position within the EU is increased today in three ways, as a result of its recent behavior toward and in Brussels: ﬁrst, given the obvious failures of the Stability and Growth Pact and the frustrations of the new members in having to wait through an ERM II period for euro entry, both still currently enforced largely by Germany, and were Germany to offer to reprioritize away from macro-rules to micro-integration, there would be any number of EU member countries eager to accommodate such a shift; second, given the leadership role Germany under the Merkel government played in creating the current EU budget compromise, including its ongoing role as largest net contributor to that budget, both the Commission and the other members are in a position to yield agenda-setting in EU economic policy to Germany; and third, given the pivotal role of Germany between the three emerging blocs within the EU on economic policy (the Anglo-Nordic Euro Outs and their sympathizers; the new member states; and the Franco-Belgian-Italian logrolling would-be core), Germany has the ability to play groups off of each other and deprive any bloc of an effective majority absent its support. In addition, following the failures of the constitutional referenda in France and the Netherlands and of the Lisbon Agenda to promote better economic policies, there is a near-complete absence of a Plan B or any other creative policies for the Commission to pursue. Thus, the already-tabled, and previously blocked by Germany, EU Services and Takeover Directives, with all their beneﬁts for the German economy, could easily become the main goals of EU economic policy following some revision to meet German demands.36

Germany can and should exercise global leadership on economic issues Beyond Germany and the EU’s approach to economic integration within Europe, there are several real and troubling economic challenges that only

Exportweltmeister – so what? 131 explicit multilateral action under German leadership can resolve.37 All of these issues concern guiding the EU toward more liberal stances externally and to increasing economic engagement of the EU with poorer countries beyond its borders. Ironically, as Germany in recent years has emphasized its relationships within Europe, as opposed to transatlantic or within multilateral organizations, it has shown less leadership of Europe in external affairs. For at least four decades, Germany had been the decisive swing vote between French and British views on trade and security issues and, more broadly, to encourage the EU to look responsibly beyond the mutual protection of narrowly deﬁned national interests. Moreover, given Germany’s high export share in GDP, even if geographically foreordained rather than directly beneﬁcial and policy-driven, German governments were usually motivated to keep the global economy open and stable. Due to the security relationship and the positive memories of postwar aid and reconstruction, as well as to the core role of Germany in efforts toward European unity, German–American relations were crucial in eliciting constructive behavior from US officials toward the European project and toward economic multilateralism more generally. At a minimum, German economic leadership until recently staved off the common American (and English, among others’) perception that continental European economies were in ongoing decline and that the EU’s international economic policies were solely about the feckless pursuit of state-sponsored advantages. For the past ten years, however, Germany has largely abdicated its economic leadership role in the EU, let other member countries (notably France) drive the EU’s approach to the Doha trade round, amplified rather than offset the image of continental decline and generally allowed a void to develop in management of global economic issues.38,39 A Germany that today, sixty years after Stunde Null, is a normal nation-state, ready to exercise the roles and responsibilities of a normal foreign policy, still can make far greater contributions with far less opposition (foreign and domestic) in the realm of economics and soft power than in extra-territorial security matters. Due to Germany’s sheer economic size, EU role, and underlying belief in the value of markets, it is nearly impossible to substitute for in economic policymaking at present (as demonstrated by the vacuum of recent years when it has yielded leadership), whereas on the security front, any number of countries can substitute for it in common efforts. The primary reason for Germany to exert greater leadership on global economic issues, however, is economic, not political or diplomatic: the kind of openness beneﬁcial to German and European economic growth discussed above is increasingly vulnerable to attack, particularly with the expansion of the global labor force and the breakdown of the Doha trade talks. Three key areas emphasized here present opportunities for German to take the initiative in agenda-setting for the multilateral economic system, all as part of (not as an alternative to) Germany’s re-assertion of leadership within the EU and transatlantic relations:

132 1 2 3

A.S. Posen buying out agricultural protection in the wealthy countries; coordinating with other advanced countries on relations with China; stemming the spread of resistance to foreign ownership and takeovers.

One could easily add to this list other issues for the longer term that could use German leadership, such as putting more economic meat on Germany’s ‘privileged partnership’ with Turkey while Turkish EU accession talks drag on or promoting adoption of cleaner energy technologies in developing countries and at home.40 The three issues I emphasize here, however, all are both ripe in two senses: one, of needing positive resolution among the leading nations’ governments to forestall harmful breakdowns in global commerce (and the governments involved are increasingly recognizing that risk of breakdown); and, two, beneﬁcial in that they would reinforce productivity-enhancing changes to Germany’s own domestic engagement with global economy. First, Germany should work to reformulate the EU position on agricultural protection and subsidies and use that to put pressure on the United States for a better offer of American farm subsidies reduction. As the collapse of the Doha round of multilateral trade negotiations in summer 2006 demonstrated, in the absence of further rich country liberalization of their agricultural markets, developing countries will walk away from the table. The rich world’s resistance to liberalization ultimately rests upon the ability of France and Poland to subsidize their farmers using the EU’s common agricultural policy rather than buying them out and then the United States and Japan using the EU’s programs as an excuse for not making their own cuts in aid to farmers. Only Germany has the bilateral relationships with the key players and the sizable EU budget net donor position to give it real inﬂuence over French and Polish policy, and every prior trade round has depended upon Germany at least stepping up in this manner.41 Germany could save the Doha round – and thus the future of German exports in a constructive manner – by reinterpreting the EU budget deal which it brokered in late 2005. Germany can keep the agreed amount of money for agriculture unchanged but use that money in different ways than export subsidies and trade-distorting agricultural aid.42 The model in both policy and rhetoric could be what Germany did with the Kohlpfennig to reduce its workforce of coal miners in decades past. So doing would remove a major source of transatlantic conﬂict and much of the excuse the United States has offered for refusing to move further on agriculture. It would also of course beneﬁt poor agricultural exporting countries and would help offset within Germany the regressive effects of a VAT-increase by reducing the cost of food. Most importantly for Germany’s own economic progress, besides keeping global markets open, it would break the cycle of Franco-German logrolling that has beset EU economic policymaking for the last several years. Second, Germany should try to build a common G7 (EU–the United States–Japan) front on economic relations with China. Inherently, the Chinese government has recurring opportunities to play off European, American, and Japanese governments and companies against each other, because it controls

Exportweltmeister – so what? 133 market access and the signing of long-term or large supplier contracts. This can involve simple bidding wars on terms, which of course can be justiﬁed as normal competition; this can, however, also involve mutually harmful erosion of intellectual property protections or of standards of conduct. Since Europe views China far less as a security threat than either the United States or Japan does, the incentive is strengthened for European (including German) companies and politicians to undercut Japanese–American reluctance to export certain technologies to China or to ignore calls to temporarily hold-up some commercial agreements to signal diplomatic displeasure. Unlike agriculture and Doha, this issue cannot be resolved once and for overall (similar pressures arose recurrently with trade with the Soviet Union and arise now in relations with Iran), but it can be managed to better mutual advantage in the west than at present. Germany can pick speciﬁc issues where coordination with the United States can be encouraged in Brussels and the G-7. Defense of intellectual property rights, where the advanced economies share a common interest, is one critical area where German leadership could broker a more united front. Opening up of standards in China or rather accession of China to minimum standards – in areas as varied as accounting to machine tool measurements – that all G-7 exporters and investors can meet with their products can be pursued by German-led G-7 agreement (and thus avoid escalating tit-for-tat attempts to shut out the other members’ products by getting China to adopt one supplier’s standard). Germany can also play a crucial role in getting the EU to move forward on consolidating its representation in the IMF, World Bank, and other international economic institutions (rather than leaving it in the various national governments’ seats), in part by expressing a willingness to give up its own. A cutback in the number of seats allotted to European nations would be necessary to keep the groups workable but would serve the ultimate German goal of strengthening the role of Europe – at the same time, it would create room for China to take a more responsible role as a stakeholder in those institutions and the system, while reducing the US skepticism about both the EU and the IFI (as well as the US dominance of the latter). All of these initiatives would reinforce a shift in German foreign economic policy away from the promotion of medium-tech exports to China (a present attractive boom for Germany) toward supporting open competition in high-value-added and new sectors (a path with longer-term beneﬁts). Finally, Germany should try to stem the spreading resistance to foreign ownership and takeovers in the advanced economies. Recently, the US Congress de facto blocked two foreign takeovers – CNOOC of Unocal in the oil sector; Dubai Ports of various port service contracts – on ostensible security grounds.43 Successive German governments, as noted above, have consistently opposed foreign takeovers in recent years, be they of Deutsche Bank (though they did allow an Italian bank to takeover HVB) or of smaller non-ﬁnancial ﬁrms by hedge fund ‘vultures’. Italy, France, Russia, Japan, and a host of emerging markets have followed suit, declaring everything from oil ﬁelds to yogurt producers off-limits to foreign ownership. This has even held true for intra-EU takeovers, as seen in the opposition to the EU takeover directive. Aging,

134

A.S. Posen

high-savings Germany cannot afford to lose the investment opportunities that FDI outﬂows would yield, and the moribund German services and high-tech sectors can only gain from FDI inﬂows, which infuse knowledge and best practices from abroad. The spiral of various countries’ ad hoc decisions against foreign bidders for companies should be stopped before precedents get set and more odious formal legislative barriers are erected around the world. Particularly if further expansion of trade ﬂows are at risk due to the collapse of the Doha negotiations, it is all the more important that global integration of production via investment proceeds. In practical terms, Germany could show leadership on this issue in four ways. First, by living up to the EU’s own stated rules for mergers and acquisitions and reversing the previous German blockage of the Takeover Directive’s implementation. Second, by working with the United States and NATO allies to forge a narrow and explicit deﬁnition of what constitutes a ‘national security exception’ to the presumption that cross-border mergers should be allowed. Third, by bringing accounting standards negotiations between the United States and EU to a close, whatever the speciﬁc outcome reached, thus reducing uncertainty for prospective investors and barriers to entry. Fourth, by taking advantage of German presidency of the G-7/8 in 2007 to forge a common statement by the member heads of state publicly repudiating the animal or insect imagery tarring foreign investors and asserting that cross-border investment ﬂows – including transfers of corporate ownership – are to be welcomed as a matter of course. All of this would feed directly into resolving some of the corporate governance problems of the German economy, in particular the proclivity for Mittelstand owner-managers to keep companies below efﬁcient scale in order to maintain control and the politicization of public sector banks’ corporate lending decisions to keep local businesses open without hard budget constraints.

Toward claiming the integrationweltmeisterschaft German foreign economic policy could do so much more for the public welfare than simply promote and publicize German export success. The export focus of German public debate as well as public policy reinforces the worst aspects of domestic German political economy, arrests development in the German privatesector, erodes long-term productivity growth, allows privileged and often inefﬁcient corporate incumbents to inaccurately shift the political blame for domestic economic stagnation to relative labor costs and globalization, undercuts European economic integration, and leaves a void in global economic leadership. The export focus is also simply unfounded as a legitimate goal of economic policy since it is openness, primarily imports and competitive pressure on domestic ﬁrms, and not exports, which is positively associated with productivity growth. In addition, the intergovernmental or statist and neo-mercantilist values consistent with such pursuit of the (uncontested) export world championship are contrary to the values on which the Federal Republic has based its foreign policy of multilateralism, constructive transatlanticism, and deepening European federal-

Exportweltmeister – so what? 135 ism. In short, export competitiveness does not work for Germany as a policy goal, and, with the progress of globalization, its harmful effects are likely to only get worse. Germany has a huge opportunity to leverage needed productivity-enhancing changes in its domestic economy through more enlightened foreign economic policies. Pursuit of greater international economic integration at the national, European, and multilateral levels would beneﬁt both Germany and – given the potential for German leadership to revitalize areas from European services to the Doha trade round – the world. This would require a reprioritization as well as a reconception of German foreign economic policy from the faulty course it has been on for 10 years (in the case of deserting the Commission and the EU’s economic integration agenda) or 50 years (in the case of mistakenly reasoning backward that if German companies export, the German economy must be both efﬁcient and subject to competition). Yet, for a Germany that now recognizes, it has severe ongoing economic problems which budget cuts and wage restraint have failed to solve, and that is now truly a normal country for foreign policy purposes and is searching for an arena in which to show international leadership, such a constructive reversal of past export obsessed policies would be aptly timed. To recapitulate, such a new German foreign economic policy would consist of the following measures: 1

2

3

4

5

6

no longer repeatedly waiting for exogenously driven export booms to stimulate growth, whether through relative wage deﬂation in the eurozone or exchange-rate depreciation outside of the eurozone or demand-raising productivity gains abroad; concentrating instead on productivity-enhancing policies that promote highvalue-added, high-wage, employment, and expansion into new services and sectors; recognizing that Germany’s highly concentrated export success is in a few sectors of declining value and that domestic non-tariff protections against new entrants (foreign and domestic) have led to arrested development in the German corporate sector; utilizing competitive pressures from abroad – on product markets, on investment returns, and on corporate ownership – to induce restructuring of German business; shifting German pursuits in Brussels from macroeconomic policy rules to microeconomic integration (and supporting a strong European Commission and the EU Takeover and Services Directives as the logical next steps in that pursuit); asserting greater leadership of Europe in transatlantic policy coordination and multilateral economic negotiations, thereby forestalling downward spirals in WTO trade negotiations, in bidding for Chinese market access, and in blocking foreign corporate takeovers with overly expansive national security excuses.

136

A.S. Posen

The underlying question for German economic policymakers henceforth should be ‘does this policy advance Germany’s integration with the world economy?’ and not ‘how are we doing on exports?’ While the latter may be easier to say and to measure, the former is more likely to produce sustainable and sustained growth in Germany and Europe. Rather than trying to be the export world champion, Germany should try to champion world economic integration.

Notes 1 This essay builds upon work from Posen (2006b), partially supported by a major grant from the German Marshall Fund of the United States. I am grateful for comments on this argument and related work from Martin Baily, Fred Bergsten, Moreno Bertoldi, Klaus Gunter Deutsch, Berend Diekmann, Ulf Gartzke, Monty Graham, Brad Jensen, Jacob Kirkegaard, Nick Lardy, Marcus Noland, Jean Pisani-Ferry, Waltraud Schalkle, Alfred Steinherr, Bruce Stokes, Harald Uhlig, and David Weinstein. The views expressed here and any errors of fact are solely my own and not necessarily those of the Friedrich Ebert Stiftung or of IIE. ©IIE, 2006. 2 Rodriguez and Rodrik (2000), Baldwin (2003), and Hallak and Levinsohn (2004) give critical overviews of the literature on trade openness and growth; these recent results are discussed further in the next section. 3 And this delusion is not limited to Germany. See the cover story in The Economist (2005), which was reportedly welcomed and passed out to the Berlin press corps by the then Chancellor Gerhard Schroeder. Posen (2005d) offers a rebuttal to that article forecasting that a wage-driven recovery would be no more than a temporary palliative at best. 4 Posen (2006b) describes this state of ‘Arrested Development’ in the German corporate sector and develops a political economy analysis for why it arose, how it persists, and the damage it does. 5 See Freeman (2003, 2004) and Sinn (2005a, 2005b) on the general change of global economic environment. Hans-Werner Sinn is one notable German economist to question the beneﬁts of German export success, but his ‘Bazaar Economy’ argument (2005b) is based on an extreme assumption of rapid factor (particularly labor) price equalization. Partly for that reason, Sinn emphasizes the false threat that outsourcing and further integration supposedly presents to the German economy, rather than pointing to the risks and costs from the current failure to integrate under the guise of export promotion. Bundesministerium für Wirtschaft und Arbeit (2004) gives a databased critique of the applicability of the Bazaar Economy idea to Germany. 6 Posen (2005c, 2006a) makes the case for such German international economic leadership. 7 Even accepting that priority, distribution (across people) and variability (over time) of income growth are also well recognized as legitimate concerns for economic policy. Variability of income growth has been successfully diminished since the 1970s in the advanced economies with improved monetary and ﬁscal policies; maldistribution remains an issue, particularly in the United States amongst the advanced economies, but with that notable exception generally growth in and equality of incomes can coexist. That does not make equal distribution automatic, but means there is no inherent unfairness of distribution from pursuing growth. 8 Among many references, see Bradford et al. (2005), Wolf (2004), and Mann with Kirkegaard (2006). This assumes that the basic preconditions for economic growth (property rights, rule of law, price stability, and health of the citizenry) are met, which they are in the OECD economies. See Sala-i-Martin et al. (2004) on how few and which factors can be said to predict per capita income growth rates.

Exportweltmeister – so what? 137 9 Recent research has emphasized this channel, whereby there is micro-evidence of trade inﬂuencing ﬁrm-level productivity and economies of scale in exporting. Helpman (2006) gives an opinionated survey. Bernard et al. (2005) demonstrates that ﬁrms change the products they produce in response to import competition, moving to products that are more capital intensive and face lower low-wage import competition, thus raising productivity; Bernard et al. (2006) presents the ﬁrst evidence for an industrialized country that falling trade costs, helpful to both imports and exports, spurs within plant productivity growth. Bernard and Jensen (1999) with US data, Eaton et al. (2005) with French data, Delgado et al. (2002) with Spanish data, and Baldwin and Gu (2003) with Canadian data offer results consistent with this view. 10 This is distinct from the fact that, over a business cycle, a country is more likely to run trade deﬁcits when it is booming than when it is in recession. This is not in contradiction to claims that ‘export-led’ growth is a path to success for some economies, since that tends to only work for smaller developing economies (though even the evidence for that tends to be weakening on reanalysis, see Baldwin (2003)). 11 Geography plays a critical role in determining a country’s level of engagement with trade – it is not an accident that Germany or the Benelux countries, with many bordering neighbors and limited space and natural resources, have much higher levels of imports and exports as a share of GDP on average than say Japan and the United States (largely) surrounded by water and covering lots of land. Generally, economists control for geography’s impact by use of ‘gravity models’ that tell analysts how much trade to expect for a given country as a function of the location and distances of potential trading partners. Policy and other variables are then invoked to explain how much trade exceeds or falls short of this gravity model baseline. 12 My colleagues Monty Graham, Brad Jensen, Nick Lardy, and Marc Noland provided helpful guidance in interpreting the most recent research on this topic. One cautions that disentangling the effects of exports and imports is non-trivial econometrically, and exports may be necessary to ﬁnance imports in some instances – though it should also be noted that is more a concern for less-developed countries than for Germany and other OECD economies. There also may be particular beneﬁts of exports for lowincome nations who need the expansion of scale in production in order to grow, again not an issue for today’s Germany. 13 Broda and Weinstein (2004) and Broda et al. (2006) have done seminal work on establishing the substantial economic welfare gains from increasing available product variety through import expansion. 14 In terms of those speciﬁcally German misperceptions, Temin (2002) and DeLong (1995) discuss overestimates of the impact of Germany’s ‘Koreaboom’ and of the role of exports in German performance in the 1950s more generally. Carlin et al. (1999) and Allard et al. (2005) independently both ﬁnd that Germany’s high volume of exports cannot be accounted for by normal trade predictors including productivity – and that in fact, Germany’s export performance becomes even more of an outlier (surprisingly high versus predicted) when its low productivity is taken into account. Posen (2006b) discusses this phenomenon in more detail and argues that the high level of German exports is largely the result of low capital returns in German corporations driving up German savings versus investment, and thus of a macroeconomic imbalance, rather than evidence of microeconomic (corporate) efﬁciency. 15 The idea was to get many more degrees of freedom for investigation than those allowed by the already thoroughly mined sample of actual postwar data and thus generate more robust conclusions. 16 The widely trumpeted so-called ‘competitiveness indices’ do not have any signiﬁcant predictive power for growth either. For a recent examination, see Ochel and Roehn (2006). 17 As discussed by Freeman (2004) and Sinn (2005a), the increase in global labor supply – mostly but not solely lower-skilled – make it all the more valuable to pursue the

138

18 19 20 21 22

23

24 25

26

27

28 29

A.S. Posen

high-wage and value-added strategy in the advanced economies or all the more difﬁcult and constantly threatened to pursue the low-cost and low-wage strategy. Posen (2006b) sets out this argument; see also the depiction of ‘Germany Stalled’ in Stokes (2006b). Siebert (2005) calculates that these sectors have consistently comprised more than 80 percent of German exports. Calculated from data on European top 100 companies and GDP in Veron (2006). Helpman (2006) and the references therein discuss the importance of scale economies to productivity given the costs of trade. Stahn (2006) and Allard et al. (2005). The decline in the elasticity of exports and growth with respect to real exchange rates (and thus wages) should be seen in part as reﬂecting global trends in that direction. When production integrates globally, supplier and marketing relationships become longer-term in keeping with direct investments, more price adjustments are internal to multinational ﬁrms’ ‘transfer pricing’ decisions, hedging and diversiﬁcation make relative price shifts less important, and for rich countries export production focuses on more capital intensive goods so relative labor costs are a smaller share of total costs. For those German exports which rely on brand and customization (high-end autos, specialized machine tools), these factors are even stronger than for most industrial country exports. See Carlin et al. (1999, 2002). Campa et al. (2005) offer an opposing set of estimates that show more limited declines in pass-through, but still those declines are still more pronounced in the industries consistent with this argument. See also Deutsche Bundesbank (2003, 2004). Posen and Gould (2006) show that German wage restraint vis-à-vis productivity growth did not decrease over the 1990s (immediate overpricing of eastern German labor in 1990–91 aside) and that wage settlements in the 2000s were not signiﬁcantly lower than typical, given the level of unemployment, monetary conditions, and slow growth. Thus, there was no evidence on this measure of structural change in the domestic labor market being induced by export promotion through relative deﬂation. Schumacher (2005a, 2005b) argues in contrast that at least some sectors of German business undertook signiﬁcant restructuring during this period. It remains to be seen how far and long this truly extends. Corsetti et al. (2000) gives a formal perspective on the limited welfare beneﬁts of competitive devaluations. Feldman (1994) made an early argument that relative wages were not the only component of German external competitiveness; see also Carlin et al. (1999) for a broader take. In this spirit, there is a distinction between acknowledging that, after several years of massive and growing current account deﬁcits, ﬁnancial markets are likely to require some adjustment in the US currency and net export position (which I do), and suggesting that either currency depreciation, limiting foreign borrowing and inward investment, or export promotion should be ongoing goals of US policy beyond the minimum amounts required to stabilize the level of foreign debt (which I do not support). After 27 years, Thatcher took ofﬁce, not even the United Kingdom, let alone any of the other countries mentioned, has anything like the absence of social protection, health insurance, or labor rights that the United States exceptionally maintains. See the contributions by Freeman and by Schettkat to this volume for more direct discussion of the limited role of social and labor market institutions in explaining German economic problems. Posen (2005b, 2006b) identiﬁes the largely unrecognized parallel sources of German and Italian decline. Eichengreen (2005) gives a useful political economy overview of this history. Of course, in earlier decades, the German willingness to yield ground was in part to restore West Germany’s political acceptance in the membership of the European community, but over time, this motivation became thankfully moot.

Exportweltmeister – so what? 139 30 Laubach and Posen (1997) demonstrate that the Bundesbank was far more ﬂexible in response to economic developments than it claimed it was when later seeking to constrain future eurozone monetary policy. Posen (2005a) contains contributions documenting the countercyclical logic of German and other European countries violating the Stability and Growth Pact, the sub-optimality of policies of the ECB had it adhered to its rules more strictly, and the proclivity of German-led policies to encourage rules-based behavior by the IMF. 31 The sometimes heard contention that these were the preconditions for economic growth and stability in Europe is severely misguided. Many countries have achieved low inﬂation without such rules, the beneﬁts of low interest rates are severely diminished in Europe due to structural rigidities, and the effects on public spending programs may well squander political capital more gainfully employed pursuing other reforms. In any event, they have been insufﬁcient to produce stable and substantial growth. 32 See Posen (2005a). Even in eurozone ﬁnancial markets, the degree of intra-eurozone ﬁnancial integration exceeded that of the integration of EU ﬁnancial sectors with those of economies outside of the eurozone since the launch of the euro, except for the overnight money market. 33 Ostensibly, the Lisbon Agenda was a vision for reform and economic liberalization, but since it concentrated on policy areas subject almost solely to national control, and targets rather than standards, it fell far short of what previous Commission initiatives achieved. Denmark, Sweden, and the United Kingdom, having had their sense of apartness reinforced by the other EU members’ adoption of the euro, also became more reticent to pursue an active internal EU economic agenda. 34 Posen (2006b). See also Deutsch et al. (2006) for an extensive discussion of the potential boons for the German economy where the EU Services Directive to be adopted and fully implemented. 35 See Brunner et al. (2004), Schumacher (2005a), and Posen (2003, 2006b) for more details. Though they largely agree on this interpretation, their views differ as to whether this will be a major, perhaps sufﬁcient, change to revitalize the German corporate sector or is merely a useful ﬁrst step in that direction. 36 Of course, too much parochial German revision of these Directives to minimize their liberalizing impact would defeat their economic purposes. A few German ofﬁcials will claim that the Takeover Directive was blocked under Schroeder for precisely this reason but that it was other countries that had made this version insufﬁciently liberalizing to be acceptable – this hardly seems credible given the array of domestic German opposition, from VW on down, to the present proposal, and the belief by many academic observers that it would constitute good progress even in the present version. 37 Posen (2006a) gave a preliminary version of this argument and some of the proposals here. 38 Stokes (2006a) gives a pointed overview of challenges facing current transatlantic economic relations. 39 This does not deny the Bush Administration’s responsibility in the 2000s for the harmful effect on German (and EU)–American relations more broadly because of its unilateralism and its narrow reliance on military instruments of foreign policy and the resulting US disregard for multilateral economic efforts. 40 Posen (2006c) makes proposals for German leadership on energy issues in a similar spirit. Posen (2005c) suggests a starting point for outreach to Turkey on the economic front. 41 Falke (2005), Meunier (2000), and Deutsch (1999) give interpretative histories with the latter premise. 42 Some will claim that the EU states committed to not re-opening the budget deal until at least 2009. But, as with the Stability and Growth Pact, the Services Directive, and

140

A.S. Posen

so on, all EU intergovernmental agreements are subject to renegotiation under the guise of reinterpretation if the major states decide to do so. There is no reason the current CAP agreement should be kept any more sacrosanct than these pacts. 43 Graham and Marchick (2006) analyzes security concerns and their inﬂuence on the US government process for evaluating foreign investment and comes out in favor of very narrow rules for blocking.

References Allard, C., Catalan, M., Everaert, L. and Sgherri S. (2005) ‘Explaining Differences in External Sector Performance Among Large Euro Area Countries’, IMF Country Report No. 05/401, November. Baldwin, R. (2003) ‘Openness and Growth: What’s the Empirical Relationship?’ NBER Working Paper 9578, March. Baldwin, John and Gu, W. (2003) ‘Export Market Participation and Productivity Performance in Canadian Manufacturing’, Canadian Journal of Economics, 36: 634–657. Barro, R. (1998) Determinants of Economic Growth: A Cross-Country Empirical Study, Cambridge: MIT Press. Bernard, A. and Jensen, J. B. (1999) ‘Exceptional Exporter Performance: Cause, Effect, or Both?’ Journal of International Economics, 47: 1–25. Bernard, A., Jensen, J. B. and Schott, P. (2005) ‘Importers, Exporters, and Multinationals: A Portrait of Firms in the US that Trade Goods’, NBER Working Paper 11404. Bernard, A., Jensen, J. B. and Schott, P. (2006) ‘Trade Costs, Firms, and Productivity’, Journal of Monetary Economics, 53: 917–937. Bosworth, B. and Collins, S. (2003) ‘The Empirics of Growth: An Update’, Brookings Papers on Economic Activity, 2. Bradford, Scott and Lawrence, Robert (2004) Has Globalization Gone Far Enough? Washington, DC: Institute for International Economics. Bradford, S., Grieco, P. and Hufbauer, G. C. (2005) ‘The Payoff to America from Global Integration’, in Fred Bergsten, C. (ed.), The United States and the World Economy: Foreign Economic Policies for the Next Decade, Washington, DC: Institute for International Economics, pp. 65–110. Broda, C. and Weinstein, D. (2004) ‘Globalization and the Gains from Variety’, NBER Working Paper 10314, February. Broda, Christian, Greenﬁeld, Joshua and Weinstein, David (2006) ‘From Groundnuts to Globalization: A Structural Estimate of Trade and Growth’, mimeo, Columbia University, New York, June. Brunner, A., Decressin, J., Hardy, D. and Kudela, B. (2004) ‘Germany’s Three-Pillar Banking System: Cross-Country Perspectives in Europe’, IMF Occasional Paper 233. Bundesministerium für Wirtschaft und Arbeit (2004) ‘Basarökonomie Deutschland?’ Wirtschaftsanalysen 4, October. Campa, J. M., Goldberg, L. and Gonzalez Minguez, J. (2005) ‘Exchange Rate PassThrough to Import Prices in the Euro Area’, Federal Reserve Bank of New York Staff Reports No. 219, September. Carlin, W., Glyn, A. and Van Reenen, J. (1999) ‘Export Market Performance of OECD Countries: An Empirical Examination of the Role of Cost Competitiveness’, Institute for Fiscal Studies Working Paper No. W99/21. Carlin, W., Glyn, A. and Manning, M. (2002) ‘Cost competitiveness and export perform-

Exportweltmeister – so what? 141 ance in the OECD economies: how does the UK compare?’ mimeo, UK Treasury, London, July. Corsetti, G., Pesenti, P., Roubini, N. and Tille, C. (2000) ‘Competitive Devaluations: Toward a Welfare-Based Approach’, Journal of International Economics, 51: 217–241. Delgado, M., Farinas, J. and Ruano, S. (2002) ‘Firm Productivity and Export Markets: A Non-Parametric Approach’, Journal of International Economics, 57: 397–422. DeLong, J. B. (1995) ‘Comment on Peter Temin, “The Koreaboom in West Germany: Fact or Fiction?” ’ mimeo, UC Berkeley, October Deutsch, K. G. (1999) The Politics of Freer Trade in Europe: Three-Level Games in the Common Commercial Policy of the EU 1985–1997, Münster: LIT Verlag. Deutsch, K. G., Frank, B. and Gornig, M. (2006) ‘The EU Services Directive: Nightmare or Opportunity? Implications for Transatlantic Business’, American Institute for Contemporary German Studies Policy Report No. 25. Deutsche Bundesbank (2003) ‘Germany’s Competitive Position and Foreign Trade Within the Euro Area’, Monthly Report, October, pp. 15–27. Deutsche Bundesbank (2004) ‘Purchasing Power Parity Theory as a Concept for Evaluating Price Competitiveness’, Monthly Report, June, pp. 29–42. Dollar, D. (1992) ‘Outward-Oriented Developing Economies Really Do Grow More Rapidly: Evidence from 95 LDCs, 1976–1985’, Economic Development and Cultural Change, 40(3): 523–544. Dollar, D. and Kraay, A. (2004) ‘Trade, Growth, and Poverty’, Economic Journal, 114(493): F22. Eichengreen, B. (2005) ‘The Political Economy of European Integration’, in Wittman, D. and Weingast, B. (eds), Oxford Handbook of Political Economy, Oxford: OUP. Falke, A. (2005) ‘German Trade Policy: An Oxymoron?’ in Kelly, D. and Grant, W. (eds), The Politics of International Trade in the 21st Century: Actors, Issues, and Regional Dynamics, London: Palgrave Macmillan. Feldman, R. (1994) ‘Measures of External Competitiveness for Germany’, IMF Working Paper No. 94/113, September. Frankel, J. and Romer, D. (1999) ‘Does Trade Cause Growth?’ American Economic Review, 89(3) (June): 379–399. Freeman, R. (2003) ‘Trade Wars: The Exaggerated Impact of Trade in Economic Debate’, NBER Working Paper 10000, September. Freeman, R. (2004) ‘Doubling the Global Workforce: The Challenge of Integrating China, India, and the former Soviet Bloc into the World Economy’, Lecture, IIE, Washington, DC, November 8. Graham, E. and Marchick, D. (2006) Foreign Direct Investment and US National Security, Washington, DC: Institute for International Economics. Hallak, J. C. and Levinsohn, J. (2004) ‘Fooling Ourselves: Evaluating the Globalization and Growth Debate’, NBER Working Paper 10244, January. Helpman, E. (2006) ‘Trade, FDI, and the Organization of Firms’, NBER Working Paper 12091, March. Jansen, H. (2005) ‘Domestic Gloom and Export Boom: A Look at German Competitiveness’, Ecﬁn Country Focus No. 2:6, Brussels: European Commission, April. Eaton, J., Kortum, S. and Kramarz, F. (2005) ‘An Anatomy of International Trade: Evidence from French Firms’, mimeo, University of Minnesota, October. Krugman, P. (1994) ‘Competitiveness: A Dangerous Obsession’, Foreign Affairs, 73(2): 28–44.

142

A.S. Posen

Laubach, T. and Posen, A. (1997) ‘Disciplined Discretion: Monetary Targeting in Germany and Switzerland’, Princeton Essays in International Finance No. 206, December. Lee, H. Y., Ricci, L. A. and Rigobon, R. (2004) ‘Once Again, Is Openness Good for Growth?’ NBER Working Paper 10749, September. Mann, C. with Kirkegaard, J. (2006) High-Tech and the Globalization of America, Washington, DC: Institute for International Economics. Meunier, S. (2000) ‘What Single Voice? European Institutions and EU-US Trade Negotiations’, International Organization, 54(1): 103–135. Ochel, W. and Roehn, O. (2006) ‘Ranking of Countries – The WEF, IMD, Fraser and Heritage Indices’, CESifo DICE Report 2/2006, pp. 48–60. Posen, A. (2003) ‘Frog in a Pot: Germany’s Descent into the Japan Syndrome’, The National Interest, Spring: 105–118. Posen, A. (ed.) (2005a) The Euro at Five: Ready for a Global Role? Washington, DC: Institute for International Economics. Posen, A. (2005b) ‘Auf der Kippe’, Internationale Politik, 1/2005, January. Posen, A. (2005c) ‘Mehr Mut, Deutschland’, Internationale Politik, 3/2005, March. Posen, A. (2005d) ‘Ein Aufschwung Reicht Nicht’, Financial Times Deutschland, August 29. Posen, A. (2006a) ‘The United States Needs German Economic Leadership’, IIE Policy Brief No. PB06–1, January. Posen, A. (2006b) Reform in a Rich Country: Germany, Washington, DC: IIE. Posen, A. (2006c) ‘Deutschland – ein Beispiel für den Rest der Welt?’ in Petermann, J. (ed.), Sichere Energie im 21. Jahrhundert, Hamburg: Hoffmann und Campe Verlag. Posen, A. and Gould, D. P. (2006) ‘Has EMU Had Any Impact on the Degree of Wage Restraint?’ in Cobham, D. (ed.), The Travails of the Eurozone, Palgrave (released as IIE Working Paper No. 06–06, Washington, DC, August). Rodriguez, F. and Rodrik, D. (2000) ‘Trade Policy and Economic Growth: A Skeptic’s Guide to the Cross-National Evidence’, NBER Macroeconomics Annual 2000, Cambridge, MA: MIT Press. Rodriguez, F. and Rodrik, D. (2001) ‘Trade Policy and Economic Growth: A Skeptic’s Guide to the Cross-National Evidence’, in Bernanke, B. and Rogoff, K. S. (eds), NBER Macroeconomics Annual 2000, Cambridge, MA: MIT Press for NBER. Rogers, J., Hufbauer, G. C. and Wada, E. (2001), ‘Price Level Convergence and Inﬂation in Europe’, Institute for International Economics Working Paper, No. 01–1, January. Sala-i-Martin, X., Doppelhofer, G. and Miller, R. (2004) ‘Determinants of Long-Term Growth: A Bayesian Averaging of Classical Estimates (BACE) Approach’, American Economic Review, 94(4): 813–835. Schumacher, D. (2005a) ‘Germany Proﬁts from Restructuring’, Goldman Sachs Global Economics Paper No. 122, February. Schumacher, D. (2005b) ‘German Manufacturing Will Survive EU Enlargement’, Goldman Sachs Global Economics Paper No. 135, December. Siebert, H. (2005) The German Economy, Princeton, NJ: Princeton University Press. Sinn, H.-W. (2005a) ‘The Dilemma of Globalisation: A German Perspective’, Economie internationale, 100: 111–120. Sinn, H.-W. (2005b) ‘Basar-Ökonomie Deutschland: Exportweltmeister oder Schlusslicht?’ ifo Schnelldienst, Sonderausgabe 6/2005, March. Stahn, K. (2006) ‘Has the Impact of Key Determinants of German Exports Changed? Results from Estimations of Germany’s Intra Euro-Area and Extra Euro-Area Exports’, Deutsche Bundesbank, Discussion Paper No. 07/2006.

Exportweltmeister – so what? 143 Stokes, B. (2006a) ‘The Stresses of Deep Integration: The Transatlantic Relationship’s New Economic and Political Challenges’, American Institute for Contemporary German Studies Policy Report No. 23. Stokes, B. (2006b) ‘Germany Stalled’, National Journal, July 15, pp. 36–42. Temin, P. (2002) ‘The Golden Age of European Growth Reconsidered’, European Review of Economic History, 6: 3–22. The Economist (2005) ‘The Surprising German Economy’, August 22. Veron, N. (2006) ‘Farewell National Champions’, Bruegel Policy Brief No. 2006/04, June. Wolf, M. (2004) Why Globalization Works, New Haven, CT: Yale University Press.

7

Wanted A new German Wirtschaftswunder Richard B. Freeman

Introduction The symptoms are unmistakable. A low employment to population rate. Limited hours worked of those employed. Weak growth of GDP. Modest increases in productivity. High unemployment of long duration. The post-World War II miracle economy, the centre piece of the Rhineland model, the locomotive of the European Union, Germany, is ill – the sick economy of Europe. Economic doctors hover around the patient. Dismal looking men with chalk dust on their sleeves, women carrying PCs talking about GSOEP, beady-eyed professors mumbling about invisible hands, ﬁnancial analysts looking for palms to grease. What can we do to cure Germany? Greater wage dispersion, say the doctors of orthodoxy. Weaker unions. Lower corporate taxes. Remove the welfare state bandages. We must be bold, force the patient from the bed. It worked for welfare mothers in America, they aren’t dead. What matters is how far we go. There is another shore, you know. Far from Brussels, where billionaires crow. More doctors push their way into the room, and the cacophony grows. Follow the ten commandments of the OECD Jobs Strategy, intones one wizened economist, carrying an ancient tablet. Less employment protection. Lower unemployment beneﬁts. Flexibility. Med-e-cines of the e-e-evening, Beautiful, beautiful laissez-faire med-e-cines! Structural adjustment pills, declares the man from IMF, left over from the Asian ﬂu. Korea recovered, so can you. Active labor market programs, says a woman with a shopping basket of organic produce. Parental leave. Flexicurity. Childcare. Will you, won’t you, will you, won’t you, join the Nordic folk dance? In a corner, a slithy salesman in a Russian ushanka beckons, Hartz, Agenda 2010, one euro a box – allow me to sell you a couple? A woman carrying a test tube chases him away. Listen to the lady with the science degree. Look what it did for the British bourgeoisie. One point three two babies per woman will never do, Germans will become an endangered species in some zoo. Sperm and eggs. Eggs and sperm. Embryonic soo-oop of the e-eevening, Who would not give two pennyworth for beautiful embryonic soup! What can Germany do to get itself back on the right track? This chapter tries to answer this question. It examines the problems that

A new German Wirtschaftswunder 145 afﬂict the German labor market and economy; reviews recent efforts to improve the labor market; directs attention at family policy and social insurance incentives to activate supply; and speculates about how to jump-start the demand side of the economy. I reach three conclusions: 1

2

3

From the 1990s to the mid-2000s, the German labor market became more ﬂexible and market driven. Weaker unions, less collective bargaining, greater wage and income inequality, the Hartz Commission reforms of the public employment and social insurance system, and other changes created a more market-driven ﬂexible labor system without reducing joblessness or sparking growth. Radical work incentives oriented around family policy – greater investment in day care, separate taxation of family earners, longer school days or after school programs – can activate female labor supply further. Providing bonuses to unemployed workers to ﬁnd jobs quickly can speed up jobﬁnding. But the big problem is to stimulate domestic demand in a global economy where ﬁrms can readily shift work to low wage countries and where governments cannot use deﬁcit spending to spur demand. I hypothesize that family policies that marketize household production and increase fertility can boost domestic consumption. Increased proﬁt-sharing and private pension fund investments directed inwardly might also help move the economy forward.

Diagnosing the illness Table 7.1 presents data on the economic ills that plague Germany and contrasts German outcomes with those in two nominally healthier economies, the United States and the United Kingdom. The ﬁrst category of ills is inadequate utilization of labor resources. This shows up in a low employment population rate (line 1) and limited hours worked among the employed (line 2), which cumulates to low work hours per adult (line 3). It also shows up in a high rate of unemployment (line 4) and a huge rate of long-run unemployment (line 5). In 2005, the proportion of Germans unemployed over a year exceeded the total unemployment rates in the United States and the United Kingdom. The second category relates to the growth of output and output per capita. GDP per capita grew more slowly in Germany than in the United States and the United Kingdom (line 6) and than in most other advanced countries. Population also grew slowly (line 7) so that aggregate GDP increased at less than half the growth of aggregate GDP in the United States (line 8). The slow growth of population and the slow growth of productivity each explain about half of the gap between German and US growth of GDP. As a result of its slow growth, Germany dropped in the GDP per capita tables from nine in 1991 to 13 out of 15 countries in 2005 for which the US Bureau of Labor Statistics provides

146

R.B. Freeman

tabulations. In 1991, West Germany was number two in GDP per capita among these countries; I estimate that it has fallen to ten in 2005. The third category relates to the growth of productivity. Most of the slow growth of GDP per capita in Germany is due to the slow growth of employment in Germany compared to the United States and the United Kingdom (line 10), but GDP per hour grew less rapidly in Germany than in the United States (line 12). Productivity in manufacturing, which is better measured than productivity in the entire economy, has also grown sluggishly relative to the United States, though it is comparable to the growth of productivity in the United Kingdom (line 13).

Table 7.1 The symptoms of being the sick economy of Europe, 2005 Germany Utilization of labor 1 Employment/adult population 2 Annual hours worked per employee 3 Workhours/adult population 4 Unemployment rate 5 Unemployment rate >1 year Economic Growth and GDP per capita ranking, 1991–2005 6 Growth of ln GDP per capita, ppp terms 7 Growth in population 8 Growth of GDP (=6 + 7) 9 Rank in GDP per capita among 15 advanced countries, 1991/2005 estimated West Germany Productivity Growth 10 Growth of ln employment 11 Growth of ln annual hour per employee 12 Growth of ln GDP per hour worked 13 Growth of ln output per hour, mfg

65.5 1435 940 11.3 6.1

0.178 0.030 0.208 9/13

US 71.5 1804 1290 5.1 0.6

UK 72.6 1672 1214 4.6 1.0

0.293 0.158 0.451 1/1

0.331 0.047 0.378 12/6

0.186 –0.031 0.303 0.658

0.084 –0.057 0.329 0.0398

2/10 0.012 –0.068 0.258 0.416

Source: Lines 1–5, OECD, 2006. Line 6, 9, U.S. Department of Labor Bureau of Labor Statistics Ofﬁce of Productivity and Technology, www.bls.gov/ﬂs/ June 16, 2006, Comparative Real Gross Domestic Product Per Capita and Per Employed Person Fifteen Countries 1960–2005; West German rank estimated by extrapolation of the difference between West Germany and Germany in 1998, BLS, Table 2. Line 7, Population, from www.oecd.org/dataoecd/62/38/35267227.pdf for 1991 with 2005 from country sources. Line 10, U.S. Department of Labor Bureau of Labor Statistics Ofﬁce of Productivity and Technology Comparative Civilian Labor Force Statistics, 10 Countries, 1960–2005 www.bls.gov/ﬂs/lfcompendium.pdf, Table 3. Line 11, OECD, 2005 and 2006 with hours 1990–2005. Line 12, tabulated from GDP per employee (U.S. Department of Labor Bureau of Labor Statistics Ofﬁce of Productivity and Technology, www.bls.gov/ﬂs/ June 16, 2006) and line 11. Line 13, ftp://ftp.bls.gov/pub/special.requests/ForeignLabor/prodsuppt01.txt.

A new German Wirtschaftswunder 147 Locus of employment gaps Table 7.2 gives the rates of employment for various demographic groups and the rate of full-time employment. Column 1 gives the rates for Germany in 2005. It shows signiﬁcant differences in rates of employment among demographic groups: higher rates for men than women, with low rates of full-time employment for women. The employment rate is much lower among workers aged 55–64 and those aged 15–24 than among prime age workers and is lower for less-educated workers than for those with tertiary education. Column 2 divides the German employment rates by the rates in the United States. It shows that in 2005, German employment rates were roughly 90 percent US employment rates for both men and women. The big difference in employment by gender is in the rate of full-time employment, which is around 92 percent for Germany and the United States for men, but which is much lower for women in Germany than in the United States. What is perhaps most striking in this table is that employment of women with children less than three years old is nearly the same in Germany and in the United States. In 1990, 41.4 percent of German mothers with children less than six years old were working, compared to 54.0 percent of American mothers with children less than six years old – a gap of 12.6 percentage points. By 2002, the rate of employment among German women with children less than six years old had risen to 57.1 percent compared to a US rate of 59.5 percent – a gap of just 2.4 percentage points. Still, there is a marked difference in the rate of full-time employment between German and US mothers, as there is between women in two countries overall. Seventy percent of employed US women with children less than six years old worked full-time,

Table 7.2 Employment population rates for selected groups, Germany and Germany relative to US, 2005.

Men Working full-time Women Working full-time Women with children <3 (2002) working full time(2002) 15–24 25–54 55–64 Less than upper secondary Upper secondary Tertiary

Germany

Germany/US

71.2 66.6 60.0 36.4 54.1 31.1 42.6 77.4 45.5 48.6 69.5 82.7

0.92 0.92 0.91 0.67 0.96 0.75 0.79 0.98 0.75 0.86 0.95 1.01

Source: OECD, 2006 and www.oecd.org/dataoecd/ 34/30/34542436.xls, SS41 and SS42. Notes There are modest differences between the Labour Force Statistics and Employment Outlook statistics.

148

R.B. Freeman

whereas 53.8 percent of employed German women with children less than six years old worked full-time. Even more striking, as children age, full-time employment falls in Germany while it rises modestly in the United States.1 This table also shows that employment rates differ between Germany and the United States largely among the youngest and oldest workers. The gap among the youngest reﬂects increased school-going in Germany and the modest employment of in-school Germans compared to US students, who combine school and work, which is necessary to fund tuition and living expenses at university. Employment of older male workers fell in Germany and the United States due to early retirement, but the stronger employment situation in the United States kept a larger proportion of older workers in the workforce. Finally, the differential employment rates for persons with the lowest level of education highlight another part of the German labor market problem, though comparisons of education groups with countries that differ so much. The drag of policy toward East Germany Why did the German economy, which had performed admirably in most of the postwar period, do so badly in the 1990s–2000s? In the 1980s, unemployment rose and economic growth slowed in West Germany, but few analysts would have labeled the economy ‘ill’. All advanced countries struggled in that decade. The US performance was sufﬁciently weak to impel Michel Albert to publish Capitalism against Capitalism (1991), which extolled the ‘Rhine model’ as superior to the US variant of capitalism. But in the 1990s, German problems exceeded those of most other advanced countries. Uniﬁcation of East Germany with West Germany created the difﬁcult problem of merging economies with different levels of productivity and modes of work, which Germany failed to solve. The Kohl government’s economic policies in reuniﬁcation were based on political rather than economic considerations. The government made a one-for-one conversion rate between the West German and East German marks, provided West German social beneﬁts for unemployed East Germans, and gave incentives for capital-intensive investments in the East, which created few jobs. For their part, German unions and employers negotiated West German pay in East Germany, in part to discourage ﬁrms from relocating there. These policies laid the groundwork for long-term economic difﬁculties in the East, which fed back onto the western part of the country and underlie current difﬁculties. With the collapse of communism, East Germany suffered an approximately 30 percent loss of GDP between 1989 and 1991, which exceeded the losses of output in most other transition economies. Employment fell by one million in 1990–19932 and never recovered. With wages out of line with productivity, East Germany failed to attract the investment that might have sparked a growth recovery and instead relied extensively on social transfers from West Germany for its income. East Germany received huge transfer payments ﬁnanced partially by a new ‘solidarity tax’ leveled on West German incomes. Annual transfers

A new German Wirtschaftswunder 149 from West to East Germany averaged 40–60 percent of East Germany’s GDP in the 1990s.3 In 2000, about 4 percent of West German GDP was spent ﬁnancing social support in the East, and upward of one-third of East German consumption came from transfers. The way in which West Germany treated East Germany reﬂects a common mode of behavior between wealthy and poor parts of the same polity. Analyzing the link between the United States and Puerto Rico, Enchautegui and I have labeled this pattern the “rich uncle” syndrome (Enchautegui and Freeman, 2006). The rich uncle (the wealthier area) has generous policies toward its struggling relative (the poorer area) that create unintended dependencies rather than development. US policies toward Puerto Rico resemble German policies toward East Germany in many ways. The United States funds social beneﬁts for Puerto Rico, which has contributed to Puerto Rican men having the lowest labor participation rate in the world just as federal German government funds the unemployed in East Germany, sustaining a rate of unemployment of 20 percent or so. US tax breaks to companies that invest in Puerto Rico produce high capital labor ratios with little employment, just as federal German government incentives produced high capital labor ratios in East Germany with little employment (Snower, 2006). Puerto Rico adopted the US federal minimum wage when the minimum was high enough to hurt employment on the Island, just as collective bargaining in Germany imposed high wages on East Germany. In both the United States and Germany, mobility from the poorer area to the wealthier area provides an exit option for those seeking work but unable to ﬁnd it in the poorer area. But there is one big difference between the rich uncle behavior of the United States toward Puerto Rico and that of West Germany toward East Germany. Puerto Rico is too small to affect the American economy, whereas East Germany is large enough to affect the German economy. German labor institutions On the basis of the better employment and unemployment record of the United States and the United Kingdom, many policy-makers and analysts believe that Anglo-American labor institutions are superior to German labor institutions. While the validity of this claim is debatable (Freeman, 2005), there is no gainsaying that institutions differ greatly between Germany and most other EU countries and the United States (and the United Kingdom). The most widely used measures of economic institutions are the indices of economic freedom produced by the Fraser Institute and Heritage Foundation. The Fraser Institute index of economic freedom scales economies from 1 to 10 based on metrics for ‘personal choice, voluntary exchange, freedom to compete, and protection of person and property’ (Fraser Institute, 2002: 5). Higher scores reﬂect greater dependence of outcomes on markets as opposed to institutions. The Heritage Foundation/Wall Street Journal (Heritage/WSJ) scores countries on a different list of variables. The conservative orientation of the organizations leads them to rate protection of property as a positive factor in economic

150

R.B. Freeman

freedom while viewing protection of labor as a negative factor that restricts business decisions. Lines 1 and 2 of Table 7.3 give the Fraser and Heritage/WSJ rankings of the economies of Germany, the United States, and the United Kingdom and the average of other advanced European OECD countries and Japan. Germany rates lower in market orientation than the United States and the United Kingdom but rates higher than the average of other advanced economies in this table. Disaggregating the Fraser index of economic freedom shows that the biggest differences between Germany and the United States and the United Kingdom are in labor market institutions. Line 3 shows the rankings of economies in the Fraser Institute sub-index for the labor market.4 Germany rates low on market orientation because it has historically relied more on collective bargaining and legal mandates than the United States and the United Kingdom and most other advanced economies.5 Some readers may be concerned that the conservative orientation of the Fraser and Heritage/WSJ foundations biases their scoring. Line 4 records the rankings of countries in the market orientation of their labor market from a different source. This is the 2004 Global Labor Survey (GLS) (Chor and Freeman, 2005) – an Internet-based survey that asked selected union leaders, labor law

Table 7.3 Rank of Germany, US, and UK by indices of reliance on markets vs institutions, 2003–2005. Germany

US

UK

19 18 101 23

3 12 10 6

6 7 19 13

World Economic Forum executives, 2003 (80 countries) 5 Wage ﬂexibility 6 Pay link to productivity 7 Hiring and ﬁring practices 8 Delegation of authority 9 Cooperative labor–management relations

79 44 79 12 20

3 2 3 6 21

6 6 11 7 18

GLS, labor practitioners, 2004 (33 countries) 10 Freedom of association/collective bargaining 11 Labor disputes 12 Employment regulations and working conditions 13 Employee beneﬁts

2 25 6 8

22 30 32 28

20 22 24 20

1 2 3 4

Economic freedom, Fraser, 2003 Economic freedom, Heritage, 2005 Labor market, Fraser, 2003 Labor market, GLS, 2005

Source: Heritage Foundation 2005, www.heritage.org/research/features/ index/ Notes Fraser Institute (2005) Tabulated from World Economic Forum, Tables 10.19 for ﬂexibility of wage determination, 10.21 for pay and productivity, 10.18 for hiring and ﬁring, 10.13 for delegation of authority; 10.20 for cooperation in labor-employer relations; Global Labor Survey, Freeman and Chor (2005), table 5, modules 3, 4, 5, 6

A new German Wirtschaftswunder 151 and industrial relations professors, government ofﬁcials, and company personnel executives to report on the actual situation of labor in their country. The GLS also shows that Germany relies more on institutions and less on the labor market than the United States and the United Kingdom and most other advanced countries.6 The next ﬁve lines of Table 7.3 examine speciﬁc institutional differences as reported by business executives on the global competitiveness reports of the World Economic Forum (2003). Line 5 shows that executives report that Germany relied more on wage-setting institutions than the United States and the United Kingdom. Line 6 shows that Germany falls below the United States and the United Kingdom in the leeway management has to relate pay to productivity. Line 7 shows that Germany’s employment protection laws give its ﬁrms less control over hiring and ﬁring practices than the United States and the United Kingdom. By contrast, lines 8 and 9 show no great difference between Germany and the other countries in the extent to which management delegates authority to employees and in the extent of cooperation in labor–management relations between Germany and other advanced countries. The GLS asked respondents to judge the favorableness of their country to labor interests as opposed to business interests in a variety of domains important to labor.7 Lines 10–13 give the rankings of the countries in indices of overall labor market performance, freedom of association/collective bargaining, labor disputes, regulations and working conditions, and employee beneﬁts. In each case, Germany is rated as more favorable to labor than the United States and the United Kingdom. In sum, both executives and labor practitioners show that Germany is more labor friendly than the United States and the United Kingdom and is among the most labor friendly of advanced nations. Do institutional differences explain Germany’s economic ills? Many economists believe that the primary reason why European economies, especially Germany, had a worse employment record than the United States in the 1990s/2000s is that they depend too much on labor market institutions to determine labor outcomes. Five of the recommendations in the 1994 Jobs Study of the OECD (1994a, 1994b) were about reform to lessen the role of institutions: increased ﬂexibility of working time; making wage and labor costs more ﬂexible; weakening employment security provisions; changing unemployment and related beneﬁt systems; and undertaking active labor market policies – training programs, job-ﬁnding assistance to workers, subsidies to employers to hire long-term unemployed or disabled workers, and special programs for youths leaving school. The other recommendations were for boilerplate good economic policies. Nearly a decade after the Jobs Study, the IMF published an article in its World Economic Outlook that predicted that if European countries deregulated their labor market and product markets in the direction of the US institutions, unemployment in Europe would fall massively:

152

R.B. Freeman labor reforms could produce output gains of about 5 percent and a fall in the unemployment rate of about 3 percentage points . . . those beneﬁts could be doubled by simultaneous efforts to increase competition in the product market.8 high unemployment is largely structural in nature – and thereby potentially affected by institutions – rather than cyclical (and therefore determined by the business cycle and macroeconomic policies). (Enact the reforms and) . . . unemployment could fall by about 6 percentage points9 (emphasis added)

Since in 2003, the United States had an unemployment rate of 6.0 percent while Germany had an unemployment rate of 9.1 percent, this analysis suggested that Germany would have had the same unemployment as the United States if it had US labor institutions and less than half the US unemployment rate if it had US product market institutions as well! At the same time, the study noted that ‘Institutions . . . hardly account for the growing trend observed in most European countries and the dramatic fall in U.S. unemployment in the 1990s’10 and stressed that Germany had broadly unchanged institutions while unemployment rose by about 6 percentage points. The claim that institutional change would resolve European unemployment problems did not take account of the longitudinal data. It rested entirely on cross-section comparisons of institutions and outcomes. In the German case, this is essentially asserting that the difference in institutions in Table 7.3 explain the differences in outcomes in Table 7.1. Studies that test this assertion with longitudinal data invariably come to the same conclusion as the IMF: that the data do not support the institutional story (Baker et al., 2005). Studies generally ﬁnd insigniﬁcant coefﬁcients with large standard errors on the institutional or policy variables. This allows analysts with strong priors to maintain their belief that making European institutions more like those in more market-driven economies, such as the United States, will resolve the problems of Europe. The orthodox analysis and prescriptions are based largely on priors rather than on evidence.11

Dr Orthodox’s medicine and German economic health ‘Medicines of the e-e-evening, Beautiful, beautiful lais-sez-faire medicines! Privatization pills. Deregulation shots. Flexible anti-wrinkle ointments. Tax cuts for capital. Wage cuts for workers. Mede-cines from Dr. Orthodox’s bag will cure your ills’ (chorus of economic doctors from Biergärten somewhere in Deutschsland). In the late 1990s to mid-2000s, Germany took much of the recommended medicine and moved its labor institutions closer to those of the United Kingdom and the United States. The Hartz Commission reforms of the public employment service and social insurance system improved work incentives in a manner mindful of the reforms of the UK public employment system with one-stop job shops (Seeleib-Kaiser and Fleckenstein, 2006). Falling union density and collective bargaining coverage gave greater weight to market wage determination. Real wages fell. The government reduced/announced plans to reduce taxes

A new German Wirtschaftswunder 153 on capital in various ways. And, proof that the medicine affected the price side of the market, the distribution of earnings and income widened. In this section, I review the policies that changed German labor institutions and their impact on employment and growth. The Hartz reforms ‘[T]he most ambitious German reform project in social insurance policy since World War II’ (Kemmerling and Bruttel, 2005: 1). In the early 2000s, I visited the Berlin ofﬁce of the German Public Employment Service to see if the Service’s innovative posting of the resumes of unemployed persons on the Internet had improved job-ﬁnding. To make the discussion concrete, I asked to see the employment service record of an unemployed man about 50 years old. Regulations did not allow outsiders to see the employment service ﬁles, but since I could not read German, the young computer-savvy professional who was hosting me went into the computer ﬁles and pulled up the record of an unemployed 50- to 55-year-old bachelor’s graduate in economics. I asked how many employers had downloaded the resume from the Internet and interviewed him. None. How many interviews had the employment service arranged for him? None. Why not? Employers wanted young workers, so it would be a waste to send this man for an interview. In a few years, the man would qualify for the pension system so the best thing was to forget him. Forget him? I was horriﬁed. A German aged 50 would live to 80 or so, which meant that the man would live half his adult life with nothing productive to do. He must feel awful. Not so, the analyst replied, most of the older unemployed are happy with their state. They refuse to interview for jobs outside Berlin and often tell employers they interview only to keep their beneﬁts. That they did not want to work I found unbelievable. Studies of life satisfaction invariably ﬁnd that the least happy people are the unemployed. I proposed that we visit the unemployed man and ask him how he felt about being on the dole. The analyst shook his head. It was one thing to break regulations by looking at a record on the computer. It was another matter entirely to barge into an unemployed man’s home and ask him whether he truly wanted to work. In 2002, the federal government discovered that the employment service had doctored their placement rates and set up the Hartz Commission to reform the system. The Commission’s recommendations were extensive: privatization of some employment activities through Personnel Service Agencies (PSAs) who earned bonuses by ﬁnding temporary jobs for the unemployed; income support so some unemployed persons could become self-employed; the creation of miniand midi-jobs with tax breaks to encourage secondary earners to accept the jobs and to induce ﬁrms to hire them; lower unemployment beneﬁts and greater employment service pressure to search for jobs and accept employment when offered. To top it off, the government commissioned 20 research institutes and 100 or so researchers to give independent academic assessment of the policies. Direct job-creation for economists, though not, I would bet, for the unemployed

154

R.B. Freeman

Berliner economics graduate whose situation introduced me to the failure of the German employment ofﬁce. The reforms were supposed to cure the ills of the German labor market. Chancellor Schroeder and Peter Hartz promised that employment would grow by two million. ‘Getting two million people back to work – that’s something we support’ – declared the head of IG Metall when he endorsed the Hartz Commission program. Hartz told the country that the PSAs would by themselves move upward of 250,000 unemployed persons into work. At this writing, the studies of the Hartz reforms are coming in. They indicate that after some chaos due to changes in computer systems and responsibilities, the administrative reforms have improved the operation of the employment agency. But the centerpiece of the reforms, the PSAs, was an utter failure. By March 2005, they had found jobs for 23,000 persons. Most unemployed persons who used the services report negatively on their experiences. By contrast, there was a huge take up of mini- and midi-jobs. In 2005, several million persons held such jobs. If most of these persons had previously been unemployed, the reforms would have greatly reduced unemployment. But relatively, few of the mini/midijobholders had been unemployed. Many persons working part-time shifted to the new midi- and mini-jobs. Firms restructured their jobs to take advantage of the tax breaks. The provision of self-employment grants (Ich-AG) had a substantial take-up by 2005, but it is doubtful that many of these businesses will survive long. The death rate of new ﬁrms is high. The major change in the unemployment beneﬁts system – reducing unemployment beneﬁts after one year and placing greater administrative pressures and sanctions on the unemployed to seek and accept work – occurred later than the other reforms. By increasing the incentives for the unemployed to get jobs, the tightening of the unemployment beneﬁts should speed the transition from unemployment to employment. Still, their effect, if any, and that of the other reforms have been at most modest. In 2002, when the government set up the Hartz Commission, there were 4.1 million unemployed and 39.1 million employed in Germany. In 2005, there were 4.9 million unemployed and 38.8 million employed. Active labor market programs and increased incentives to choose work over the dole may be necessary for the German economy to return to health, but they are not sufﬁcient. They do not create jobs nor spur demand. Declining unionism and collective bargaining coverage Without legal actions to weaken unions per the United Kingdom or virulent antiunion campaign by employers per the United States, Germany has rapidly deunionized its workforce. Union density fell from 38 percent in 1993 to 19 percent in 2003 in East Germany and from 27 percent in 1993 to 21 percent in 2003 in West Germany (Fitzenberger et al. 2006). The result is a drop in unionization economywide from 32 percent to 22 percent – an annual decline of one percentage point per year that matches the decline in union density in the United Kingdom under Thatcherism and exceeds the annual decline in density in the

A new German Wirtschaftswunder 155 United States. Scattered estimates of collective bargaining coverage show that the proportion of workers whose wages are set by union–employer bargaining has also fallen though the proportion remains substantial. In West Germany, collective bargaining set the wages of 69 percent of workers in 1996, whereas in 2003, it set the pay of 62 percent of workers. In East Germany, the drop in the percentage of workers whose wages were set by collective bargaining fell from 56 percent (1996) to 43 percent (2003). In metalworking, approximately 75 percent of workers covered by collective agreements in the late 1980s dropped to about 60 percent in West Germany and 21 percent in East Germany in 2003 (Eichhorst and Kaiser, 2006: 14). Why did unionization and collective bargaining fall without active government or employer opposition to unionism? As best I can tell, the problem lies with short-sighted union policies that led some workers and ﬁrms to prefer nonunion solutions to union policies and that placed unions in non-winnable battles. Shortly after reuniﬁcation, the German unions signed up millions of East German workers for whom they gained West German wages, in part to protect employment in the West. The unions hoped that subsidies to employers would create jobs in the East. But with high wages, employment contracted in the East, and as East German jobs went, so too did the members in that part of the country. In West Germany, union responses to the economic problems of ﬁrms led to strikes that failed to accomplish their goals. The German Workplace Representation and Participation Survey conducted in 2003 by the Max Planck Institute showed that the attitudes of German workers toward their unions deteriorated in the late 1990s to early 2000s, with increasing numbers reporting less favorable views of unions (Schedlitzki, 2003). German unions, once the leading progressive union force in Europe, have lost their way in ﬁnding solutions for Germany’s labor problems. Declining real wages/rising dispersion From the 1980s through the mid-1990s, Germany had rising real wages and stable or falling dispersion of wages, which contrasts to falling real wages and rising inequality in the United States. When the Clinton Administration took ofﬁce in 1990, it looked at Germany as a model for improving wage and training outcomes for lower paid workers in the United States. But in the 1990s, Germany’s wage performance deteriorated. Responding to the sluggish economy and high joblessness, unions and employers bargained for modest wage changes. Between 1991 and 2004, the US Bureau of Labor Statistics shows higher real hourly compensation increases for US manufacturing workers (23.5 percent) than for German manufacturing workers (22.2 percent). The OECD’s real compensation per employee in the business sector series shows German real wages falling from in 2004 and 2005 after barely changing in 2001–2003. Since unions and ﬁrms were signing modest wage agreements over this period, there is no great surprise in these patterns. What is surprising is that the distribution of wages and income widened considerably as well.

156

R.B. Freeman

Table 7.4 Measures of dispersion of earnings and income in Germany Study and data

Early/late

Measure

1 Gernandt and Pfeffer (2006), SOEP 2 Kohn (2006), IABS

1984/2004

90/10 ratio

2.59

3.01

0.42

1992/2000

3 Steiner (2000), SOEP

1990/1997

80/20 ratio, m 80/20 ratio, f 90/10 ratio, m 90/10 ratio, f

1.71 1.91 2.39 2.38

1.87 2.06 2.52 2.27

0.16 0.15 0.13 –0.11

90/10 90/10 family incomes 90/50 gross income 99/50 99.999/50

2.88

3.04

0.16

3.90

4.58

0.60

4.18 13.4 606.1

4.29 13.6 825.4

0.11 0.20 219.3

4 OECD (2006), LF + SOEP 5 Burkhauser and Rovba (2006) 6 Bach et al. (2006)

1984/1998 1991/2001 1992/1998

Early

Late

Change

Source: Gernandt and Pfeiffer (2006), Table 2. Kohn, Steiner, OECD, cited in Schettkat, Burkhauser and Rovba (2006), Table 1. Bach, Corneo, Steiner (2006), Table 3.

Schettkat (2006) and Kohn (2006) have documented the increased dispersion in earnings in various series and measures. Burkhauser and Rovba report that the Gini coefﬁcient for all income in Germany rose from 1990 to 2000 to attain roughly the same level as in the United Kingdom – which had greater inequality than Germany since at least the Thatcher period (Burkhauser and Rovba, 2006: Table 1). The biggest increase in inequality is among the young. Using tax data, Bach et al. (2006) have found that during 1992–1998, the share of income going to the upper 10 percent increased somewhat, but that the biggest change in income was toward the upper 0.001 percent (Table 12), who make most of their money through ownership of capital than through work. Table 7.4 summarizes the growing evidence of a substantial shift in the income distribution toward the well to do. Other policies Germany has tried some of the other medicines of the economic doctor’s medical bag. It has lowered the top rate of taxes and plans to lower corporate income taxes in the hope that companies will invest in the country. It has lowered the inheritance tax for family-owned ﬁrms who maintain the business for ten years. It has liberalized some product market regulations, such as retail store hours, and required less reporting from small- and medium-sized ﬁrms. It has reduced modestly the social insurance charges on employment. It has increased the ﬂexibility of working time. These policies, together with the weakening of unions and collective bargaining, Hartz reforms of the employment service and tightening of the unemployment beneﬁt system, and rising inequality, have made the German economy more friendly to capital and less friendly to labor. But they have not cured the patient. Employment levels remain low and

A new German Wirtschaftswunder 157 economic growth sluggish. It is possible that bigger doses of the orthodox medicine will eventually cure the patient, but then again, they might just continue shifting income from the lower paid to the highest paid and from workers to owners. At the minimum, the negligible success of extant economic prescriptions suggests the value of exploring new policies and alternative medicines.

Activating supply through . . . . . . marketization In a series of papers, Ronald Schettkat and I have documented that Germans, particularly women, spend more time on traditional household production – food preparation, childcare, elderly care, cleaning houses – than do Americans (Freeman and Schettkat, 2000, 2002, 2005). Using time budget data, we found that German women worked as many hours in total as American women but did their work in the household rather than in the market. For example, German women spent more time cooking than American women while Americans bought more meals from restaurants. German women with children less than four years old spent 18.4 hours taking care of them while American women with children less than four years old spent 10.8 hours taking care of them. The flip side is that 56 percent of American women with children less than three were working full-time in the job market compared to 34 percent of German women with young children (Table 7.2). On the European Labor Market Survey 2002, 62.5 percent of German women working part-time said that they were doing so because they looked after children or an adult compared to 28.7 percent of women working part-time in the EU-12 (Jaumotte, 2003: Table 7). To increase the employment and work hours of German women and to allow them to reconcile jobs and family without permanently endangering careers or discouraging childbearing requires moving family policy from the periphery of policy discourse to center stage. There are family policies that can contribute to the goal of increased employment and that can do so without lowering the rate of fertility. The ﬁrst such policy is the development of day care centers, which are generally associated with greater labor participation and hours worked by women. One reason US mothers work more than German mothers is that the United States has a larger proportion of children less than three years old in day care (54 percent) than does Germany (19 percent). Looking at regions within Germany, van Ham and Büchel (2004) report higher employment for mothers working in regions with a high density of childcare provision, and other factors held ﬁxed. Wrohlich (2006) ﬁnds that increasing the number of childcare slots has greater effects on the demand for childcare and maternal employment than a reduction of parents’ fees to existing slots. Another policy that could increase the supply of female workers is to move from joint taxation of family members to separate taxation. This would reduce

158

R.B. Freeman

the marginal tax rate on the person with lower earnings in a family with children, usually the female. Cross-country comparison indicate that single earner taxation substantially increases female labor participation (Jaumotte, 2003). Microsimulation models for Germany predict that a shift to separate taxation would increase the labor supply of wives sizably in West Germany while inducing a significant number of husbands to drop out of the labor force (Steiner and Wrohlich, 2004). By contrast, family tax-splitting reforms would benefit higher income families a lot while increasing the hours worked by women only modestly (Steiner, 2006). Thus, analyses of taxation and labor supply seem to come down in favor of complete separation of the pay checks. A third policy that would increase female employment would be to lengthen the school day and provide school lunches so that women with children could more readily hold full-time jobs. Many school districts in the United States have after-school programs funded by parents to keep the children in a play or learning environment after the normal school day. Scandinavian countries have national programs that accomplish the same result. I am unaware of studies of the impact of the school day on female labor participation and hours worked, but I would expect the effects to be sizeable. All of the family policy reforms – greater provision of childcare facilities, lower taxes on secondary earners, and lengthened school days – should impact hours worked as much or more than labor participation. Still, for these supplyside activities to bear great fruition, ﬁrms would have to restructure career ladders and promotion and training to help women maintain careers during and after the period of childbearing and rearing. . . . and child policy ‘The question is not whether women will work. They will work. The question is whether they will have kids’ (Ursula von der Leyen, federal minister for Family Affairs, Senior Citizens, Women, and Youth, 2006; Landler, 2006). Historically, one of the major economic costs of a child was the time spent in bearing and rearing the child. This time lowered female educational attainment and employment in the market sector. But the increased participation of women with young children working has grown in Germany and most other advanced countries, which suggests that policies that increase incentives for women to work in the market need not reduce childbearing. Indeed, there has been a remarkable shift in the cross-country relation between fertility and economic and labor market activities in advanced OECD countries. In the past, countries with high fertility had proportionately fewer women in higher education and employment. In the 1990s, these patterns shifted, so that by the 2000s, the cross-country relation between fertility and the education and employment of women had been completely reversed! In 2003, higher fertility countries have greater female enrollments in college, higher employment rates, and higher GDP per capita (d’Addio and d’Ercole, 2005: Figures 8, 10, 12, 13). Figure 7.1 displays the new cross-

A new German Wirtschaftswunder 159 1980

3.4 IRL 3.0 Total fertility rate

KOR 2.6 2.2

PRT

GRC ESP

1.8

ITA

BEL

FRA

AUT DEU

NLD

1.4

AUS

NZL

GBR

USA JPN

CHE

SWE FIN

1.0 20

30

40

50

60

Employment rates of women

2000

2.6 MEX

Total fertility rate

2.2

ISL USA NZL IRL NOR FRA AUS NLD DNK FIN GBR CAN KOR CHE HUN DEU SWE AUT POL JPN SVK ITA ESP CZE

1.8

1.4

1.0 20

40

60

80

Employment rates of women

Figure 7.1 The reversal between female employment and fertility (source: d’Addio, A. C. and M. M. d’Ercole (2005), Figure 13).

section pattern in terms of the relation between the total fertility rate and the employment rate of women that is most relevant to this chapter. On the basis of these relations, there is no reason to believe that there is any contradiction between the desire to increase fertility and the desire to increase female employment.

160

R.B. Freeman

. . . and unemployment bonuses The Hartz Commission scheme of establishing public service agencies to move the unemployed into temporary help jobs failed. As the OECD noted in its 2006 report on Germany, the program ‘proved to be costly without improving placement outcomes in general . . . (and) should be terminated’ (OECD, 2006: 12). But this does not mean that only negative incentives will work in moving the unemployed to ﬁnd jobs more quickly than they have done in Germany. In the 1980s, Illinois experimented with ways to encourage the unemployed to ﬁnd employment before they exhausted their unemployment beneﬁts. One experiment offered bonuses to unemployed workers to ﬁnd work quickly. Another experiment paid the same amount to the ﬁrms to hire unemployed workers. The Coase Theorem predicts that the property right to the bonus ought not to affect the employment outcome. But the payments to workers led to greater employment and bonus receipt than did the payments to ﬁrms (Donahue, 1989). On the basis of this result, the employment service should consider experimenting with different ways to pay lump sum amounts to unemployed persons who ﬁnd jobs quickly. For instance, the service might offer a half year’s unemployment beneﬁt for ﬁnding a new job within three months. Ideally, the savings on unemployment beneﬁts from the reduced time on the system would ﬁnance the lump sum payments.12 But such a scheme would have to be planned carefully so that workers and ﬁrms do not exploit the system: you lay me off, I become unemployed, and then you hire me back quickly and we split the lump sum. One way to control such behavior might be to limit the number of times a worker could access the lump sum, which amounts to an experience rating for individuals that resembles individual unemployment accounts.13 Another way to limit incentives for cheating on the lump sum scheme would be to tie it to experience rating of ﬁrms, which the German parliament rejected in consideration of the Hartz reforms. The lesson I draw from the Hartz PSA failure is not that positive incentives for job-ﬁnding must fail but that those incentives should go to the workers rather than to some intermediary or even to ﬁrms.

Activating demand through . . . . . . marketization Policies that encourage marketization of household activities have the potential not only to supply of time offered to the market by educated and skilled women but also to increase the demand for labor. They do this because marketization of household production creates demand for market substitutes for the goods and services previously produced at household work. This has knock-on effects on the demand for labor, in some cases particularly for lower paid workers. As more highly educated women work full-time, the demand will grow for teachers, childcare workers, cooks, and other restaurant workers. Finally, while policies that encourage marketization of household activities are likely to beneﬁt highly

A new German Wirtschaftswunder 161 educated and skilled women more than others, the changes will have knock-on effects on the demand for lower paid workers. As more highly educated women work full-time, the demand for workers to provide market substitutes for the reduced household production will increase. This means greater demand for more teachers and childcare workers, for cooks and other restaurant workers, for household cleaners, and the like. The United States has approximately twice as many persons employed per capita in the restaurant sector as Europe because of the marketization of household services. . . . and child policy Policies that encourage childbearing are also likely to stimulate demand. Families with children spend more than families without children. They buy a bigger ﬂat or house, new clothes, toys, milk, baby food, as well as wine and beer. Families with children spend 10–20 percent more than those without children (d’Addio and d’Ercole, 2005: 48, Box 3). In countries like the United States, where the family pays for higher education, spending rises sharply when young persons go to university. From a micro-perspective, these costs deter individuals from having children, but from a macro-perspective, the costs can stimulate consumption spending and demand in an economy with underutilized resources. If part of Germany’s economic problem is inadequate consumption spending, which seems a reasonable deduction, then family policies that marketize household production and increase fertility offer a way to spark additional private consumption. Perhaps I have been overly struck by the reversal in the relation between fertility and female employment and education and income per capita among advanced countries, but at the minimum, these new pattern direct attention to population and family policies that can affect demand as well as supply. The growth stimulus that Keynesians saw coming from deﬁcit spending and that conservative economists hoped would come from tax cuts to the wealthy may instead be found in family policies that induce more women to have children and to work. . . . and negative income taxes The most successful income redistribution program in the United States has been the earned income tax credit, which gives low wage workers a positive incentive to work in the form of a negative income tax as opposed to being on welfare and which reduces their supply price to employers. This stimulates demand as well as supply. The success of the mini- and midi-job experiments shows that policies that raise the rewards to work by lowering taxation can succeed in Germany. Gradual increases in the hours worked for which people would be covered by these tax breaks – turning the mini- and midi-jobs into fulltime low paid work – can complement schemes that use the stick of reduced unemployment compensation to provide an incentive for employment.

162 R.B. Freeman . . . and increase worker shares of proﬁts The collapse of the Soviet Empire, China’s move to market capitalism, and India’s market reforms and entry into the global trading system have altered the global economy in favor of capital. The multinational ﬁrms on the technological frontier of modern production can now recruit highly skilled low wage workers outside of the advanced capitalist economies. They can move jobs and modern technology to the low wage countries, which was not possible when the countries were in their own autarkic economic space. As a result, corporate proﬁts are high and likely to remain so, albeit subject to business cycle and other shocks. There are three ways to try to deal with this new economic world. The ﬁrst is to try to induce capital to invest in high wage countries through tax breaks. Such schemes failed in the United States and Germany to induce ﬁrms to create job-creating investments in Puerto Rico and East Germany. Corporate tax changes such as the 2000 German changes may have some positive effect on investment in the domestic economy as opposed to overseas (Becker et al., 2006) and perhaps even on employment but aggregate economic forces invariably overwhelm their impact. The second is through reductions in wages in advanced countries. Studies of multinational employment in the 1990s found little evidence that cross-country wage differentials reduce home country employment.14 There is a scale effect and substitution effect from growth into new markets, and in some circumstances, the growth from new markets may dominate substitution in the demand for domestic labor. But Muendler and Becker (2006) estimate fairly sizable elasticities of substitution between German labor and labor in other countries in German multinationals. Their estimates are consistent with substantial job loss in Germany due to globalization creating new opportunities for firms to hire workers elsewhere.15 Wage reductions need not lead to reductions in living standards if the government uses negative income taxes or reduces taxes for normal workers to compensate for lower pay, but such a strategy would require increased taxes on the wealthy, which in Germany means capital owners. There is, however, a third possible way to protect the well-being of workers in a era when capital is in the ascendancy. This is to give labor a cut of the proﬁts accruing to capital as owners of shares in ﬁrms or as part of proﬁt-sharing schemes. Some countries – the United States, Australia, the Netherlands, Sweden – have large private pension systems that invest in equity and give workers a stake (and risk) in how ﬁrms do. Many ﬁrms in the OECD have introduced proﬁt-sharing as part of pay, which has similar effect with respect to the workers’ own ﬁrm. With its limited stock market capitalization and extensive insider-owned ﬁrms, it would seem easier for Germany to push for proﬁt-sharing than for greater workers’ ownership of shares, which would require a radical shift in the nature of the German mode of capitalism. Still, it might be worthwhile for Germany to give tax breaks to small- and medium-sized ﬁrms that set up employee ownership schemes.16 This would complement or replace the recent

A new German Wirtschaftswunder 163 policy of eliminating inheritance taxes for small ﬁrms that remain operating for a decade afterward.

Conclusion Could these new approaches to Germany’s economic problems cure the patient? I am not sure. The point of this chapter is not to form a new Bieregarden chorus of economic doctors who claim that they know the solution to Germany’s problems but rather to pose different places to look for solutions. Perhaps policies that encourage marketization of household production and increase fertility and female full-time work and that link worker incomes to returns to capital would not spark a new German Wirtschaftswunder. But perhaps, they would. I realize that the medical profession would never give pills or operate on a sick patient on the basis of the arguments in this chapter, but they also would never allow a physician to follow the solutions of the orthodox economic doctors on the basis of the record of those solutions. Korea and other Asian countries recovered from their 1990s ﬁnancial problems in spite of the medicines the IMF tried to force down their throats and the East Asian miracle occurred with policies that made the World Bank ill. Today, China sits at the bottom of indices of market freedom and corruption and grows and grows and grows. While economists cannot run medical trials on animals to see if novel medicines work, we can simulate policies through artiﬁcial agent modeling and at least determine the parameters and patterns under which they might work. If Germany was to experiment with some novel policies and if the sick economy were suddenly to stir from the bed, it would give the dismal doctors unexpected fright and then great delight. Beautiful Soup, so rich and green, Waiting in a hot tureen! Instead of arguing about what Germany might do, we would argue over whose medicine, after all, got it right.

Notes 1 2 3 4

www.oecd.org/dataoecd/34/30/34542436.xls, SS41 and SS42. www.destatis.de/presse/englisch/pm2000/p2810031.htm. World Bank, Transition the First Ten Years, Washington, DC, 2003, pp. 36–39. Until 2001, the Fraser index dealt only cursorily with labor. In its 2001 report, however, the Institute presented a more comprehensive freedom index for 58 countries that included six indicators of labor institutions. By 2003, it extended these measures to 103 countries. The rankings of countries by labor market status gives high scores to countries with little or no labor protection, such as Uganda, the United Arab Emirates, Zambia, and Haiti (the top four countries in the Fraser labor market freedom sub-index for 2003), while giving low scores to countries that have welldeveloped legal systems of protecting workers, such as Germany. This scoring produces a negative relation between ‘economic freedom’ in the labor market and GDP per capita (Freeman, 2002). 5 Because the Heritage/WSJ measure of economic freedom does not treat the labor market with the same attention, we do not report its sub-index in Table 7.3. 6 That the rankings of countries by labor practices are similar in the GLS and Fraser measures indicates that the orientation of analysts did not noticeably affect their reporting of country labor practices.

164

R.B. Freeman

7 Speciﬁcally, I subtracted the score from 34, so that the country ranked 1 (most labor friendly) obtained a rank of 33, and the country ranked 33 was ranked 1, and so on. 8 International Monetary Fund (2003: Chapter 4, 129). 9 International Monetary Fund (2003: Chapter 4, 131). 10 International Monetary Fund (2003: Figure 4.4, 134). 11 I say ‘largely’ because there is microeconomic evidence regarding the impact of unemployment compensation systems on unemployment that shows that greater duration of unemployment beneﬁts produce greater duration of jobless spells. There is also evidence that many active labor market programs do little to reduce joblessness. 12 This is similar in spirit but differs from Dennis Snower’s proposal to give vouchers to workers that go directly to the employers. The Illinois experiments suggest that this will not work. 13 See Snower for one such scheme. Feldstein and Altman (1998) discuss such policies in the United States. 14 See Muendler and Becker (2006) for a summary of the literature. 15 They examine the effect of changes in wages in Germany and in other countries on employment and show sizable effects, with own elasticities of employment for ﬁrms with German operations on the order of –0.57 (Table 10) and for large cross-country effects due to both changes in employment and changes in the location decisions of plants. They do not assess a counterfactual in which China, India, and most important for Germany, the Soviet Empire, remains outside the global economy. 16 The United States gives tax advantages to ﬁrms that do this through ESOPs. The United Kingdom has recently eliminated some of the tax breaks it gave to employee beneﬁt trusts that had the same effect (Equity Incentives, Inc).

References Albert, M. (1991) Capitalisme contre Capitalisme, Seuil; English translation (1992) Capitalism against Capitalism, Whurr Publishers. Bach, S., Corneo, G. and Steiner, V. (2006) ‘Top Incomes and Top Taxes in Germany’, CESifo Working Paper 1641. Baker, D., Glyn, A., Howell, D. and Schmitt, J. (2005) ‘Labor Market Institutions and Unemployment: A Critical Assessment of the Cross-Country Evidence’, in Howell, D. (ed.), Fighting Unemployment, Oxford University Press, Oxford and New York, 72–118. Baker, D., Glyn, A., Howell, D. R. and Schmitt, J. (2006) ‘Are Protective Labor Market Institutions at the Root of Unemployment? A Critical Review of the Evidence’, mimeo, New York. Becker, J., Fuest, C. and Hemmelgarn, T. (2006) ‘Corporate Tax Reform and Foreign Direct Investment – Evidence from Firm-Level Data’, CESifo Working Paper 1722. Burkhauser, R. and Rovba, L. (2006) ‘Income Inequality in the 1990s: Comparing the United States, Great Britain, and Germany’, DIW Discussion Paper 576. Chor, D. and Freeman, R. B. (2005) ‘The 2004 Global Labor Survey: Workplace Institutions and Practices Around the World’, NBER Working Papers 11598. d’Addio, A. C. and d’Ercole, M. M. (2005) ‘Trends and Determinants of Fertility Rates: The Role of Policies’, OECD Social Employment and Migration Working Papers 27, OECD Publishing. Donahue, J. J. III (1989) ‘Diverting the Coasian River: Incentive Schemes to Reduce Unemployment Spells’, Yale Law Journal, 99: 549–609. Eichhorst, W. and Kaiser, L. (2006) ‘The German Labor Market: Still Adjusting Badly?’ IZA Discussion Paper 2215.

A new German Wirtschaftswunder 165 Equity Incentives, Employee Beneﬁt Trusts (2006) www.equityincentives.co.uk/documents/pdfs/Employee_beneﬁt_trusts.pdf. Enchautegui, M. E. and Freeman, R. B. (2006) ‘Why Don’t More Puerto Rican Men Work? The Rich Uncle (Sam) Hypothesis’ in Collins S. M., Bosworth, B. P. and SotoClass, M. A. (eds) The Economy of Puerto Rico: Restoring Growth, Brookings Institution Press: 152–188. Feldstein, M. and Altman, D. (1998) ‘Unemployment Insurance Savings Accounts’, NBER Working Paper 6860. Fitzenberger, B., Kohn, K. and Wang, Q. (2006) ‘The Erosion of Union Membership in Germany: Determinants, Densities, Decompositions’, IZA Discussion Paper 2193. Fraser Institute (2002) ‘The Areas and Components of the EFW Index’. www. freetheworld.com/2007/2007Dataset.xls. Fraser Institute (2005a) ‘Economic Freedom of the World Report: 2005 Annual Report’, Vancouver. Fraser Institute (2005b) ‘Index of Economic Freedom’. www.heritage.org/research/ features/index/. Freeman, R. B. (2002) ‘Varieties of Labor Market Institutions and Economic Performance’, IRRA Session Labor Market Institutions and Economic Outcomes, January 4. Freeman, R. B. (2005) ‘Labour Market Institutions Without Blinders: The Debate over Flexibility and Labour Market Performance’, International Economic Journal RIEJ, 19 (2): 129–145, Juni and NBER Working Paper 11, 286, April. Freeman, R. B. and Schettkat, R. (2000) ‘The Role of Wage and Skill Differences in US–German Employment Differences’, NBER Working Paper 7474. Freeman, R. B. and Schettkat, R. (2002) ‘Marketization of Production and the US–Europe Employment Gap’, NBER Working Paper W8797. Freeman, R. B. and Schettkat, R. (2005) ‘Marketization of Household Production and the EU–US Gap in Work’, Economic Policy, 20 (41): 6–50 (45). Gernandt, J. and Pfeifer, F. (2006) ‘Rising Wage Inequality in Germany’, ZEW Discussion Paper 06–019. International Monetary Fund (2003) ‘Unemployment and Labor Market Institutions: Why Reforms Pay Off’, in World Economic Outlook (April), Washington, DC: IMF. Jaumotte, F. (2003) ‘Female Labour Force Participation: Past Trends and Main Determinants in OECD Countries’, OECD Economics Department Working Paper 376. Kohn, K. (2006) ‘Rising Wage Dispersion, After All! The German Wage Structure at the Turn of the Century’, ZEW Discussion Paper 06–031. Kemmerling, A. and Bruttel, O. (2005) ‘New Politics in German Labor Market Policy?’ WZB Discussion Paper. Landler, M. (2006) ‘Quoth the Raven: I Bake Cookies, Too’, New York Times, April 23. Muendler, M.-A. and Becker, S. O. (2006) ‘Margins of Multinational Labor Substitution’, CESifo Working Paper 1713. www.equityincentives.co.uk/documents/pdfs/ EBTs_Stash.pdf. OECD (1994a) OECD Jobs Study, Evidence and Explanations, Part I: Labor Market Trends and Underlying Forces of Change, Paris: OECD Publishing. OECD (1994b) OECD Jobs Study, Evidence and Explanations, Part II: The Adjustment Potential of the Labor Market, Paris: OECD Publishing. OECD (2006) Economic Surveys 2006 – Germany, Paris: OECD Publishing. Schedlitzki, D. (2003) Master’s Thesis, Oxford University. Schettkat, R. (2006) Lohnspreizung: Mythen und Fakten Eine Literaturübersicht zu Ausmaß und ökonomischen Wirkungen von Lohnungleichheit, Hans-Böckler-Stiftung.

166

R.B. Freeman

Seeleib-Kaiser, M. and Fleckenstein, T. (2006) ‘Lesson Learnt from Britain? The Case of German Labour Market Reforms’, PSA Annual Conference, Oxford. www.psa.ac.uk/2006/pps/Seeleib-Kaiser.pdf#search=%22SeeleibKaiser% 2C%20 Martin%20and%20Timo%20Fleckenstein%20%E2%80%9CLesson%20Learnt %20from%20Britain%3F%20%22. Snower, D. (2006) ‘Reducing Unemployment through Fundamental Labour Market Reform’. www.agf.org.uk/pubs/pdfs/1510web.pdf. Snower, D. J. and Merkl, C. (2006) ‘The Caring Hand that Cripples: The East German Labor Market After Reuniﬁcation’, IZA Discussion Paper 2066. Steiner, V. (2006) ‘Introducing Family Tax Splitting in Germany: How would it affect the income distribution and work incentives?’ IZA Discussion Paper 2245. Steiner, V. and Hölzle, T. (2000) ‘The Development of Wages in Germany in the 1990s – Descriptions and Explanations’ in Hauser, R. and Becker, I. (eds) The Personal Distribution of Income in an International Perspective, Berlin: Springer: 7–30. Steiner, V. and Wrohlich, K. (2004) ‘Household Taxation, Income Splitting and Labor Supply Incentives. A Microsimulation Study for Germany’, CESifo Economic Studies, 50: 541–568. Van Ham, M. and Büchel, F. (2004) ‘Unwilling or Unable? Spatial, Institutional and Socio-Economic Restrictions on Females’ Labor Market Access’, IZA Discussion Paper 1034. ftp://ftp.iza.org/dps/dp1034.pdf. World Economic Forum (2003) ‘The Global Competitiveness Report 2002–2003’. Wrohlich, K. (2006) ‘Labor Supply and Child Care Choices in a Rationed Child Care Market’, IZA Discussion Paper 2053.

Index

accounting standards 134 Adda, J. 107 aggregate demand: and economic performance 75–85; effects of reform 102–5; and long-term unemployment 88–9; open economies 76–9; shortage of 22–4, 73–5; swings in 105–7; UK/German comparison 85–6, 90–6 aggregate output 22–3 aggregate unemployment 76–9 agricultural protection 132 Albert, Michel 148 Allsopp, C. 80 analyses 7–16 Angeloni, I. 41 Anglo-Saxon countries 106 apprenticeship system 74 Australia 36, 38, 39, 40, 122 Austria 36, 38, 39, 40 automatic stabilizers 61–4, 65 Bach, S. 156 Baker, D. 15, 96, 152 Baldwin, R. 122 Baltensperger, E. 56 Bank of England 105 banking sector: fragmentation of 120; importance of 44–7 Barro, R. 123 Basle II Capital Accord 129 Bassanini, A. 14–15, 96–8, 99 Baumol, W.J. 6 Becker, J. 162 Belgium 36, 38, 39, 40 Benedetti, M. 66 birth rates 16 Blanchard, O. 16, 96 Blinder, A.S. 7, 10 Boeri, T. 68

Bosworth, B. 123 Bradford, S. 127 Brandt, N. 102, 103, 106, 112 Bretton Woods system 55 Büchel, F. 157 budget deﬁcits 62–6; EMU limit 58 Bundesbank 69–70: loss of independence 7–8; output estimation method 28; performance 25; studies 124; success of 55–7 Burkhauser, R. 14, 156 Burniaux, J.-M. 103, 112 business cycle theory 29 business regulation 86–7 Canada 36, 38, 39, 40 capacity, unused 11 Capitalism against Capitalism (Albert) 148 capital mobility 67 Carlin, Wendy 9, 11, 14–15, 17, 72–115 child policy 157–9, 161 China 11, 132–3 Chor, D. 150–1 Christiano, L. 29 Clarida, R. 29, 42, 56 closed economies 76–7, 79–80, 88–90 coalition governments 81 Coase Theorem 160 collective bargaining 154–5 Collins, S. 123 Common Agricultural Policy (CAP) 132 competition: effects of EMU 67; from imports 122–3 competitiveness, Germany 123–6 consensus political systems 83 consumption: contribution to growth 92–3; effects of monetary policy 41; falling 124–5; spending 161

168

Index

coordinated market economies (CMEs) 84–5 Corneo, G. 156 corporate sector, problems in 120, 123–4, 125, 129 Costa Storti, Cláudia 8, 9, 29–51 Cotis, J.-P. 101 counter-cyclical ﬁscal policy 80–1, 100–2 country size, and monetary policy effects 44–7 ‘creative accounting’ 26, 59 credit regulation 86–7 cyclically adjusted budgets 26, 62–6 d’Addio, A.C. 158, 159, 161 d’Ercole, M.M. 158, 159, 161 de Grauwe, Paul 8, 9, 29–51 debt ceiling, EMU 58–9 demand-side policies 68–9, 160–3 demographic groups, employment rates 147–8 Denmark 36, 38, 39, 40, 126 dependencies, creation of 149 descriptive statistics, effects of monetary policies 31–4 deutschmark (DM), as reference currency 57 discretionary ﬁscal policy 80–1, 88, 100–2 Doha Trade Round 131, 132, 134, 135 Dollar, D. 122 domestic demand 10–12, 24–8 Donahue, J.J. III 160 Dustmann, C. 107 Duval, R. 14–15, 67, 68, 96–8, 99, 103, 112 earned income tax credits (EITC) 107, 161 East Germany: drag of policy toward 148–9; social transfers 16, 26–7 econometric analysis: output effects 34–7; price effects 37–41 economic freedom indices 85–7, 149–51 economic growth: components of 5, 12–14; overview 2–3; requirements for 121–3 economic ills, diagnosis of 145–52 economic integration 126–30 economic issues, German leadership 130–4 economic performance: Germany 75–85; UK/German comparison 85–107 economic policy: European approach to 128–30; and reuniﬁcation 148–9 economic policy debate 1–5; international analysis of 5–7

economic reforms 2–3, 61, 102–5 education: differences in 83–4; and employment 147–8; females 158, 161; investment in 13–14; standards in 72 Eichel, Hans 3 Eichenbaum, M. 29 Eichhorst, W. 155 Elmeskov, J. 67, 68 employee ownership schemes 162–3 employment: changes by industry 93, 95–6; growth in 2–3; UK/German comparison 90–6; see also unemployment employment gaps 147–8 employment incentives 160 employment protection 102–3, 112 employment rates 145–6, 147 Enchautegui, M.E. 149 entry criteria, EMU 58 equilibrium employment 76–9, 88–9 Esping-Andersen, G. 83 Estonia 58 Eurobarometer 60 European Central Bank (ECB) 17, 25, 58, 69–70; as continuation of Bundesbank 60–1; economic philosophy of 7–9; and monetary policy 79–80 European Commission 127–8, 129–30 European Innovation Scoreboard 2005 72–3 European Labor Market Survey 2002 157 European Monetary Union (EMU): asymmetric demand shocks 80; design of 58–9; and ﬁscal policy 61–9; and monetary policy 55–61; overview 54–5 European Union: economic challenges 130–4; economic policy 128–30; German approach to 128–9; German bargaining position in 130; microeconomic integration 126–30; Services Directive 129, 130, 135; Takeover Directive 125–6, 129, 130, 133–4, 135 Eurozone, effects of monetary policy 41–3, 49 Evans, C. 29 exchange rate distortion 123 Exchange Rate Mechanism (ERM II) 127, 130 exchange rate regimes 44–7 exchange rates 77–8, 82; changes in 89–90 expansionary policies 3–4, 59–60; in ECB model 8 expectation impact, restrictive policy 65

Index 169 export-led growth strategies 11, 125–6 exports 121–3; competitiveness 123–6; dependence 24; focus on 119–21 family policy 16, 157–8 Federal Reserve 8–9, 41–2 female labour force participation 15–16, 74–5, 103–5, 109, 147–8, 157–9, 160–1 fertility 15–16, 103–5, 158–9, 161 Finland 36, 38, 39, 40, 105 ﬁscal policy: contractionary 25–8, 62–4, 65; expansionary 59–60, 62–4; German experience 62–4; and institutional blockage 80–1; and monetary union 64–6; principles 61–2; pro-cyclical 9–10, 55 Fischer, S. 42 Fitzenberger, B. 154 Fleckenstein, T. 152 foreign demand 10–12 foreign economic policy: EU 126–30; German experience 123–6; German global leadership 130–4; openness 121–3; overview 119–21; policy proposals 134–6 foreign ownership/takeovers, resistance to 133–4 France 21, 36, 38, 39, 40, 128–9, 131, 132 Frankel, J. 122 Fraser Institute Index of Economic Freedom 85–7, 149–50 Freeman, Richard B. 5, 14, 15–16, 18, 144–63 Friedman, Milton 4 Fuest, C. 162 G7/G8 72, 132–3, 134 Gali, J. 29, 42, 56 Galileo satellite program 125 German Workplace Representation and Participation Survey 2003 155 Gernandt, J. 156 Gertler, M. 29, 42, 56 Giavazzi, F. 57, 66 Giovannini 57 Global Labor Survey 2004 150–1 global labour supply 123 globalization: economics beneﬁts of 119–20, 121; suspicion of 129 Glyn, A. 15, 96, 152 government consumption 92–3, 94 Greece 36, 38, 39, 40 Groningen Growth and Development Centre 95

gross domestic product (GDP) 66–7, 90–3, 122–3, 124–5, 148–9 Growth Competitiveness Index, WEF 85–6 Gylfason, T. 42 Hall, P.A. 84 Hallak, J.C. 122 Hallerberg, M. 81 Hartz Commission 145, 152, 153–4, 160 Hemmelgarn, T. 162 Heritage Foundation 149–50 Hoem, J.M. 103 Höpner, M. 73 house price trends 93, 94 household expenditure/savings 92–3, 94, 96, 102–5 household production, marketization of 160–1 Howell, D. 15, 96, 152 Hufbauer, G.C. 127 imperfect markets 76–9 imports 121–3 India 11 industrial relations, differences in 83–4 industries, employment changes by 93, 95–6 inﬂation: Eurozone 58, 59; low/stable 8, 27; rate of 44–7, 49; and real wages 78–9 inﬂation–output tradeoff 68 inﬂation-targeting monetary policy 79–80 innovation 28 institutional blockage and ﬁscal policy 80–1 institutional differences 151–2 institutional environment indicators 85–7 institutional reform, Germany 14–15 institutional setup, EMU 58 institutional variables, UK/Germany 110–12 intellectual property rights 133 interest rates: effects of increases 31–2; and inﬂation-targeting monetary policy 79–80; real short-term 25 international analysis, German economic policy 5–7 international economic institutions 133 International Monetary Fund (IMF) 124, 133, 151–2 investment: contribution to growth 92–3; effects of monetary policy 41; highwage countries 162; private 28; as share of GDP 12–14

170

Index

investment-based recovery 23 Ireland 36, 38, 39, 40, 126 ‘island model’ 44 Issing, O. 8, 65 Italy 36, 38, 39, 40, 126, 128–9 Iversen, T. 42, 84 Jansen, H. 124 Japan 36, 38, 39, 40, 80, 122, 132–3 Jappelli, T. 66 Jaumotte, F. 157, 158 job security 102–3 Kaiser, L. 155 Kashyap, A. 41 Keynesian economics 3, 7, 161 Kirkegaard 124 Knell, M. 30 Koch-Weser, Caio 130 Kohl government 128–9, 148 Kohn, K. 154, 156 Kraay, A. 122 Krugman, P. 11, 80, 125 labour force participation 15–16 labour institutions, Germany 149–51 labour market participation 152–60 labour markets: demand 160–3; deregulation of 151–2; ﬂexibility in 9, 74, 81; global 123; heterogeneity in 83–5; reform of 1–2, 27–8, 96–9, 100–7, 112–15; regulation 86–7; supply 157–60 labour mobility 67 Landesbanken 130 Landler, M. 158 large economies, wage restraint 10–11, 81–3 Lawrence, R. 127 Layard, R. 11 Lee, H.Y. 122 Levinsohn, J. 122 liberal market economies (LMEs) 84–5 Lijphart, A. 83 long-term unemployment 153–4 Lucas, R. 44 Luxembourg 36, 38, 39, 40 Lyndbeck, A. 42 M3 growth 60 Maastricht criteria 10, 26 Maastricht Treaty 58, 73, 74, 127 McKinsey Global Institute 28 macroeconomic gatekeeping 126–7 macroeconomic performance 54

macroeconomic policy: broadening discussion of 20–8; misunderstanding of 6–7; UK/German comparison 87–9, 100–5 macroeconomic variables, role of 43–7 majoritarian political systems 83 Mankiw, G. 7, 127 Mann, C. 124 market orientation 150–1 marketization 15–16, 157–8, 160–1 Marshall, A. 4 Meghir, C. 107 Merkel government 130 microeconomic integration 126–30 mid-career labour markets 84, 103 Mojon, B. 41 monetary conditions, Germany/Eurozone 59 monetary dominance 57 monetary policy: activist 41–2; comparison of effects in US and Eurozone 41–3; contractionary 25–6, 28; and ECB 60–1, 79–80; effectiveness of 7–9; effects on output and prices 30–41; Germany 55–61; role of macroeconomic variables 43–7 monetary targeting 55–7, 60 money, long-run neutrality of 49 Muendler, M.-A. 162 multinational companies 123–4 Musgrave, R. 28 NATO 134 natural rate of unemployment (NAIRU) 4, 15, 17 neo-Keynesian economics 29, 73–4, 75–80, 90 Netherlands 7, 36, 38, 39, 40, 126 Nickell, S. 14–15, 88, 96–8 Nijkamp, P. 40 Nordic countries 100–2, 103–5, 106 Nunziata, L. 14, 88, 96–8 Ochel, W. 14, 88, 96–8 OECD: data 13–14, 100–8, 146, 147, 156, 160; Economic Outlook Database 4, 13, 86, 91; Growth Project 3–14; Jobs Report 102; Jobs Strategy 144; Jobs Study 14, 151; macroeconomic policy in member states 100–5; National Accounts database 92, 94–5 Okun’s law 4–5, 12, 22–8 open economies 76–9, 89–90 output effects of monetary policy:

Index 171 conclusions 47–51; data 30–1; descriptive statistics 31–4; econometric analysis 34–7; US/Eurozone comparison 41–3, 49 output gap 62–4, 66–7 output growth 145–6 Pagano, M. 66 per capita GDP growth 122–3 per capita income growth 121–2 per capita output growth 145–6 Personnel Services Agencies (PSAs), Germany 153–4 Pfeffer, F. 156 Phelps, Edmund 4 Phillips curve 44 Poland 132 political system structure 83 Poot, J. 30 Portugal 36, 38, 39, 40 Posen, Adam S. 10, 11–12, 119–36 ‘potential output’ 4–5, 21–3, 28 poverty 74 precautionary savings 83–5, 102 price rigidities 42 price-setting 76–9 price stability 7–8; public support for 55–6 prices, effects of monetary policy: data 30–1; descriptive statistics 31–4; econometric analysis 37–41 principle–agent theory 81 private sector: aggregate demand 102–5; consumption/investment 92–3; reactions to ﬁscal policy 61 privatization, employment activities 153–4 product markets: deregulation of 151–2; reform of 28 productivity growth: comparisons 146; Germany 123–6 proﬁt sharing 162–3 public debt 62–6 Public Employment Service, Germany 153 public opinion polls 54, 55–6, 59, 60–1 Puerto Rico 16, 149 re-exports 24 real wages: decline in 89, 155–6; and equilibrium unemployment 76–9, 81–3 recession 22–3 replacement rate 96–9 retirement age, Germany 113–14 reuniﬁcation 21, 26–7, 57, 148–9 Ricardian equivalence principle 61 Ricci, L.A. 122

‘rich uncle’ syndrome 149 Richter, R. 55 Rigobon, R. 122 Robin, J.-M. 107 Rodriguez, F. 122 Rodrik, D. 122 Rogers, J. 127 Romer, D. 122 Rose, A. 30 Rovba, L. 14, 156 Sala-i-Martin, X. 122 savings 83–5, 93, 94 Schedlitzki, D. 155 Schettkat, Ronald 1–18, 157 Schmidt, Helmut 8 Schmidt, M. 13 Schmitt, J. 15, 96, 152 Schroeder government 128–9, 130, 154 Seeleib-Kaiser, M. 152 service sector labour markets 72, 74 shocks 23, 73–4, 75–6, 77–8, 79–83, 88–9 Siebert, H. 1 single earner households 103 Single Market 127, 128 single party governments 81 Sinn, H.W. 1 skills 72, 74–5, 83–5 small economies, wage restraint 10–11, 81–3 Smets, F. 29 Snower, D. 149 social insurance policy 153–4 social transfers, East Germany 16, 26–7, 148–9 ‘solidarity tax’ 148–9 Solow, Robert M. 3, 4, 5, 6, 9, 10, 12, 17, 20–8, 127 Soskice, David 9, 11, 14–15, 17, 42, 72–115 Spain 36, 38, 39, 40 Stability and Growth Pact 2, 10, 26, 55, 58–9, 61, 64–5, 68, 70, 74, 127, 128–9, 130 stabilization policy 100–2 stabilizing effects of ﬁscal policy 61–6 Steinbrück, Peer 3 Steiner, V. 156, 158 Stiglitz, J. 7 Stix, H. 30 Strauch, R. 81 strong-currency countries 57 structural reforms 14–15, 66–7; EMU incentives for 67–8

172

Index

structural rigidities 1–2, 4, 9, 15, 41–2 structural vector autoregression (VAR) models 7, 31, 35, 37, 40 subsidies 125–6, 132, 155 supply-side policies: demand management support for reforms 68–9; and EMU 67–8; labour 157–60; and unemployment 96–9 supply-side reforms, UK/German comparison 85–107 Sweden 36, 38, 39, 40 Tabellini, G. 42, 68 taxation 16, 27–8, 96–9, 107, 115, 156, 157–8, 161, 162–3 Taylor, J. 42 Terlizzese, D. 41 Thatcherism 14, 154–5 trade imbalances 121–2 trade openness 44–7, 121–3; beneﬁts of 119–20, 121 trade protection 125–6 training, differences in 83–4 Turkey 132 UK: aggregate demand/employment 90–6, 105, 108–9; economic performance 85–107; employment 145–6; in EU 131; foreign economic policy 126; GDP 145–6; labour institutions 149–51; labour markets 83–5; macroeconomic policy/supply-side reforms 100–5; productivity growth 146; regression results 36, 38, 39, 40; unemployment 14–15, 96–9; unionism 154–5 unemployment: and aggregate demand 73–85; funding of 149; historical overview 21; impact of structural reforms 14–15, 96–9; and Okun gap 23–8; open economies 76–9; UK/German comparison 96–9 unemployment beneﬁts 113, 154, 160 unemployment bonuses 160 unions: bargaining power 76–9, 80–3; density 14, 154–5, 96–9 US: agricultural trade policy 132; economic performance 148; employment 145–6, 147–8; engagement with EU 131–2, 133; foreign economic policy 122, 126; GDP 145–6; labour force participation 14–15; labour

institutions 149–51, 152; labour markets 7–8, 83–5, 157; marketization 161; monetary policy effects 41–3, 49; productivity growth 146; regression results 36, 38, 39, 40; tax credits 161; unionism 154–5; wage distribution 155 US Bureau of Labor Statistics 145–6, 155 US Congress 133 US–German relations 131–2, 133–4 value-added export sectors 72–3 value-added products 123–4 Van Ham, M. 157 vector autoregression (VAR) models 30, 31, 34–7, 40, 45, 48 Vines, D. 80 vocational qualiﬁcations 103 Vogel, S. 84 von der Leyen, Ursula 158 von Hagen, J. 81 Wada, E. 127 wage bargaining 76–9, 80–3 wage competitiveness 124–5 wage dispersion 155–6 wage ﬂuctuation 24–5 wage formation 114 wage reductions, advanced countries 162 wage restraint: Germany 10–12; impact of 105; small and large economies 81–3 wage rigidities 42 wage-setting, coordination in 96–9 Wald tests 42–3 Walrasian equilibrium 4 Walsh, C. 29 Wang, Q. 154 weak-currency countries 57 welfare states: guarantor–insurance role 84–5; reform 83–5 Woodford, M. 42 work time ﬂexibility 113 working hours 21 working mothers 147–8 World Economic Forum (WEF) 85–7, 150–1 World Economic Outlook, IMF 151–2 World Trade Organization (WTO) 131, 132, 135 Wouters, R. 29 Wrohlich, K. 157, 158 Wyplosz, Charles 7–8, 9, 10, 54–70