Fiscal Policy in the European Union

Fiscal Policy in the European Union

This page intentionally left blank

Fiscal Policy in the European Union Edited by

Jesús Ferreiro Giuseppe Fontana and Felipe Serrano

Selection and editorial matter © Jesús Ferreiro, Giuseppe Fontana and Felipe Serrano 2008 Chapters © individual contributors 2008 All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission. No portion of this publication may be reproduced, copied or transmitted save with written permission or in accordance with the provisions of the Copyright, Designs and Patents Act 1988, or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, Saffron House, 6-10 Kirby Street, London EC1N 8TS. Any person who does any unauthorized act in relation to this publication may be liable to criminal prosecution and civil claims for damages. The authors have asserted their rights to be identified as the authors of this work in accordance with the Copyright, Designs and Patents Act 1988. First published 2008 by PALGRAVE MACMILLAN Palgrave Macmillan in the UK is an imprint of Macmillan Publishers Limited, registered in England, company number 785998, of Houndmills, Basingstoke, Hampshire RG21 6XS. Palgrave Macmillan in the US is a division of St Martin’s Press LLC, 175 Fifth Avenue, New York, NY 10010. Palgrave Macmillan is the global academic imprint of the above companies and has companies and representatives throughout the world. Palgrave® and Macmillan® are registered trademarks in the United States, the United Kingdom, Europe and other countries. ISBN-13: 978 0 230 20399 0 ISBN-10: 0 230 20399 X

hardback hardback

This book is printed on paper suitable for recycling and made from fully managed and sustained forest sources. Logging, pulping and manufacturing processes are expected to conform to the environmental regulations of the country of origin A catalogue record for this book is available from the British Library. Library of Congress Cataloging-in-Publication Data Fiscal policy in the European Union / edited by Jesus Ferreiro, Giuseppe Fontana and Felipe Serrano. p. cm. Includes bibliographical references and index. ISBN 978 0 230 20399 0 (alk. paper) 1. Fiscal policy European Union Countries. I. Ferreiro, Jesus. II. Fontana, Giuseppe, 1968- III. Serrano Pérez, Felipe. HJ1000.F564 2008 2008029926 339.5’2094 dc22 10 9 8 7 6 5 4 3 2 1 17 16 15 14 13 12 11 10 09 08 Printed and bound in Great Britain by CPI Antony Rowe, Chippenham and Eastbourne

Contents List of Figures

vii

List of Tables

ix

Notes on the Contributors

xi

Introduction Jesús Ferreiro, Giuseppe Fontana and Felipe Serrano 1

2

3

4

5

6

7

Fiscal Federalism in the European Union: How Far Are We? Rui Henrique Alves and Oscar Afonso

1

6

Is There Compatibility between the Stability and Growth Pact and Automatic Fiscal Stabilizers? Sabrina Rostaing-Paris

25

Stabilization Capacity and Fiscal Policy in the EMU Jorge Uxó González and M. Jesús Arroyo Fernández

53

Fiscal Adjustment and Composition of Public Expenditures in the EMU Jesús Ferreiro, M. Teresa García-del-Valle and Carmen Gómez

84

Public Capital and Economic Growth: a Spurious Empirical Link? Gwenaëlle Poilon

109

Has France Become a Deficit Running Country? Jérôme Creel, Guy Gilbert and Thierry Madiès

129

How to Deal with Economic Divergences in the EMU? Catherine Mathieu and Henri Sterdyniak

157

v

vi

Contents

8

A Scheme to Coordinate Monetary and Fiscal Policies in the Euro Area Carlo Panico and Marta Vázquez Suárez

Index

184 209

List of Figures 2.1(a)–(k)

Primary cyclical balance and output gap in European countries from 1992 to 2006 2.2 Primary cyclical balance vs. output gap 2.3(a)–(b) Comparison between the size of automatic stabilizers and the reference value with respect of the sign of the output gap in Germany and France since 1992 3.1 Inflation dispersion in EMU 3.2 Average inflation rate, 1999–2006 3.3 GDP rate of growth (average, 1999–2006) 3.4 Differences between GDP and trend (rates of growth 1999–2006) 3.5 Differences between GDP and trend (rates of growth, sum 1999–2006) 3.6 Different cyclical position and single monetary policy in EMU: Spain and Germany 3.7 Real interest rates 3.8 Real exchange rates ULCE (1999 = 100) 3.9 A restrictive demand shock when the competitiveness effect is weak (country 1) 3.10 A restrictive demand shock when the competitiveness effect is strong (country 1) 4.1 Evolution of public expenditure and taxation between the periods 1990–8 and 1999–2005 (% GDP) 4.2 Size of public social expenditures before EMU (1990–8) and after EMU (1999–2005) (% GDP) 4.3 Evolution of productive and unproductive public expenditures according to the functional classification (% GDP) 6.1 Government budget balance, % of GDP 6.2 Net and gross public debt, France, % of GDP, and nominal GDP growth 6.3 Various fiscal surpluses, France, % of GDP

vii

36 37

45 57 57 58 59 59 63 64 65 71 74

91 96

104 130 132 135

viii List of Figures

6.4 6.5

Cyclically adjusted government primary balance, % of potential output Estimated non-tax revenues and acquisitions and disposals of non-financial assets

136 144

List of Tables 1.1 2.1 2.2 2.3

Financial perspectives for the period 2007–2013 Summary table of compliance conditions Automatic fiscal stabilizers, 1992–2006 Periods of compatibility between automatic fiscal stabilizers and a balanced or surplus budget in EMU countries since 1992 2.4 Periods of compatibility between automatic fiscal stabilizers and a budget deficit lower than 3 per cent of GDP 2.5 Periods of compatibility between automatic fiscal stabilizers and a balanced structural budget combined with a budget deficit limit in EMU countries since 1992 3.1 Alternative fiscal policy rules 3.2 Stabilizing and destabilizing mechanism and model parameters 3.3 Stability conditions with different fiscal policy rules 3.4 Value of model parameters used in the simulations 4.1 Evolution of fiscal deficit and public expenditure (% GDP) 4.2(a)–(c) Evolution of the size of public expenditures (in % GDP) between the periods 1990–8 and 1999–2005 by kind of expenditure (economic classification) 4.3 Public expenditures (in % GDP) by kind of expenditure (functional classification) 4.4 Evolution of public expenditure between 1990–8 and 1999–2005 4.5(a)–(b) Evolution of productive and unproductive public expenditure (% GDP) 5.1 Government net capital as a percentage of GDP 5.2 Government net capital as a percentage of private non-residential net capital 5.3 Studies based on a Cobb-Douglas production function in level 5.4 Studies with first differences

ix

15 41 42

43

46

47 69 72 73 77 91

93–5 98 102 103 110 110 113 114

x

List of Tables

5.5 5.6 5.7 5.8 5.9 5.10 6.1 6.2

6.3

6.4 7.1 7.2 7.3 7.4 7.5 7.6 7.7 7.8 7.9 7.10

Augmented Dickey-Fuller stationarity test Johansen test results Estimation results using first-differences for France, Germany and Italy Johansen test results OLS results with individual effects Final results with GMM Discretionary fiscal policy over the electoral cycle in France (1978–2005) Fiscal deficits and the political cycle: contributions to the cumulative change in general government surplus (in percentage points of GDP) over the last three decades Fiscal deficits and the business cycle: contributions to the cumulative change in general government surplus (in percentage points of GDP) over the last three decades Contributions to the rise of general government net financial liabilities (in percentage points of GDP) GDP growth rates and GDP per head Unemployment and employment rates Inflation and real interest rates Taylor rules and effective central bank rates External positions and savings ratios Fiscal policies Public finance sustainability in 2006 Employment protection indicators Economic performance disparities in the EU-15 Disequilibrium index

117 118 118 121 122 124 138

140

143 144 159 161 162 164 166 167 168 170 170 171

Notes on the Contributors Oscar Afonso is Assistant Professor in Economics in the Faculty of Economics at the University of Porto. He has a PhD degree from the University of Porto. His research interests are in the areas of macroeconomics and international trade. In particular, he is now working on the subject of wage inequality and economic growth. He is the author of a number of papers on those subjects in refereed journals (Applied Economics, Economic Modelling, Intereconomics, International Economic Journal, Journal of International Trade and Economic Development) and of chapters in edited books. Rui Henrique Alves is Invited Lecturer in the Faculty of Economics at the University of Porto, Portugal, where he teaches European Economics, International Economics and Macroeconomics. For some years, he has also been adviser of the Portuguese Exchange in the area of training and education. He holds an MSc degree in Economics and is currently waiting for the approval of his PhD thesis in European Economics. His main areas of research include political and economic organization of the European Union, fiscal federalism, fiscal policies and macroeconomic effects. He has published a book and several articles concerning these areas. M. Jesús Arroyo Fernández has a PhD in Economics from the Universidad de Alcalá (Madrid, Spain). She is Associate Professor of Economics (with tenure) at the Universidad CEU San Pablo, Madrid, Spain. She teaches Macroeconomic Policies and Applied Economics. Her main investigation lines are monetary and fiscal policies in EMU, activist fiscal policy rules and new normative macroeconomics. She has published a number of articles in specialized journals and she is author of some contributions to books on economic policy. Jérôme Creel has been a Deputy Director at Observatoire Français des Conjonctures Economiques (OFCE, Sciences Po) in Paris since 2003, and a Professor of Economics at ESCP-EAP European School of Management since 2007. He has been the Academic Dean of the European Studies Programme at Sciences Po since 2006, and was a Junior and then Senior Economist at OFCE, starting in 1998. He holds a PhD from the University of Paris-Dauphine in Economics (summa cum laude). His works, xi

xii

Notes on the Contributors

published in leading journals in the field of macroeconomics, have dealt with monetary and fiscal policies in the EMU, including coordination issues, the economics of EU enlargement and institutional economics, notably related to the Constitutional Treaty and delegation issues. Jesús Ferreiro is Associate Professor in Economics at the University of the Basque Country, in Bilbao, Spain. He has a PhD from the University of the Basque Country. His research interests are in the areas of macroeconomic policy, labour market and international financial flows. He has published a number of articles on those topics in edited books and in refereed journals such as Economic and Industrial Democracy, Economie Appliquée, Ekonomia, European Planning Studies, International Review of Applied Economics, and the Journal of Post-Keynesian Economics. Giuseppe Fontana is Professor of Monetary Economics at the University of Leeds (UK) and Associate Professor at the Università del Sannio (Italy). He has recently been awarded the 2008 L. S. Shackle Prize, St Edmunds’ College, Cambridge (UK). He is a Life Member Fellow at Clare Hall (University of Cambridge, UK), and a Visiting Research Professor at the Centre for Full Employment and Price Stability (University of Missouri Kansas City, USA), and the Cambridge Centre for Economic and Public Policy (University of Cambridge, UK). He has recently published Money, Time, and Uncertainty with Routledge. Teresa García-del-Valle is Associate Professor of Statistics at the University of the Basque Country, in Bilbao, Spain. She has a PhD from the University of the Basque Country. Her research interests are in the area of applied statistics. She has published a number of articles on this topic in books, edited books and in refereed journals. She has been Invited Professor in several Latin American universities, such as Universidad Autónoma Juan Misael Caracho (Tarija, Bolivia), Universidad Nacional Autónoma de México (México, D.F.) and Universidad del Salvador (El Salvador). Guy J. Gilbert is Full Professor of Economics at the Ecole Normale Supérieure at Cachan (Department of Social Sciences). He is a member of the Centre d’Economie de la Sorbonne-Paris I University, and a member of Paris School of Economics. His major field of research is public economics and the economics of public finance. He has published several books and numerous articles in academic journals, relating to taxation, fiscal decentralization and fiscal federalism.

Notes on the Contributors

xiii

Carmen Gómez is Lecturer in Economics at the University of the Basque Country, in Bilbao, Spain. She has a PhD from the University of the Basque Country. Her research interests are in the areas of macroeconomic policy, labour market and international financial flows. She has published a number of articles on those topics in edited books and in refereed journals such as Comercio Exterior, Economic and Industrial Democracy, Economie Appliquée, and Problemas del Desarrollo. Thierry Madiès is Full Professor of Economics at the University of Fribourg (Switzerland) and a member of the Council of Economic Analysis of the French Prime Minister. His field of research concerns public finance and international economics. Catherine Mathieu is a Senior Economist, Analysis and Forecasting Department, OFCE (Observatoire Français des Conjonctures Economiques), Paris. Her current research areas include: European macroeconomic analyses, UK economy, and macroeconomic forecasting. She is Chairwoman of the ‘Medium term prospects and structural changes’ working group, AIECE (Association of European Conjuncture Institutes) and a member of the EUROFRAME group of research institutes. Recent publications include: ‘A European Framework Designed for Stability or Growth?’, co-authored with Henri Sterdyniak, European Economic Policies: Alternatives to Orthodox Analysis and Policy Concepts, MetropolisVerlag (2006); ‘Comment expliquer les disparités économiques dans l’UEM?’, co-authored with Henri Sterdyniak, Revue de l’OFCE, 102 (2007); and ‘Le modèle social européen et l’Europe sociale’, co-authored with Henri Sterdyniak, Revue de l’OFCE, 104 (2008). Carlo Panico is Professor of Economics at the University of Naples ‘Federico II’ (Italy). He has articles published in the Cambridge Journal of Economics, Contributions to Political Economy, Economic Journal, Metroeconomica, and the European Journal of the History of Economic Thought and has published monographs on the theory of growth and distribution and on the theory of money. His research interests are the influence of monetary phenomena on growth and distribution and the history of economic analysis (among other authors, Marx, Marshall, Keynes, Myrdal, Harrod, Sraffa and Kaldor), the development of ‘British monetary orthodoxy’ in the nineteenth century, the formation of the European Monetary Union, the relationship between central bank independence and democracy, the debate on the Marshallian supply curves, the coordination of policies within the European Union, the influence on regional development of

xiv

Notes on the Contributors

the institutional framework and of the different forms of planning government intervention. Gwenaëlle Poilon is a PhD student in the programme ‘Economic Governance’ directed by Jean-Paul Fitoussi in the Institut d’Etudes Politiques de Paris; her research centre is the Observatoire Français des Conjonctures Economiques (OFCE). Her dissertation deals with the Stability and Growth Pact; more precisely, on the Golden Rule of Public Finances that has been proposed as a reform for the actual Stability and Growth Pact. Her research interests are European economic policies and education and health topics. Sabrina Rostaing-Paris is a Lecturer at the University of Bordeaux 4. She has a Masters degree in Macroeconomics from the University of AixMarseille 2, France. Her areas of interest are macroeconomics, public finances, fiscal policy and the economy of the European Union. Felipe Serrano is Professor in Economics at the University of the Basque Country, in Bilbao, Spain He is the head of the Department of Applied Economics V at the University of the Basque Country. His research interests are in the areas of social security, the welfare state, labour market, innovation and economic policy. He is the author of a number of articles on those topics in edited books and in refereed journals such as Economies et Sociétés, Ekonomia, European Planning Studies, International Review of Applied Economics and the Journal of Post-Keynesian Economics Henri Sterdyniak is Head of the Economics of Globalization Department at OFCE (Observatoire Français des Conjonctures Economiques), Paris, and Professor of Economics at the University of Paris-Dauphine. He is a member of the Steering Committee of AFSE (Association Française de Science Economique) and of EUROFRAME (European network of economic institutes). He has published many articles on macroeconomics, economic policy, monetary and international economics, European economy, fiscal and social issues. Recent publications include: ‘A European Framework Designed for Stability or Growth?’, co-authored with Catherine Mathieu, European Economic Policies: Alternatives to Orthodox Analysis and Policy Concepts, Metropolis-Verlag (2006); ‘Comment expliquer les disparités économiques dans l’UEM?’, co-authored with Catherine Mathieu, Revue de l’OFCE, 102 (2007); and ‘Le modèle social européen et l’Europe sociale’, co-authored with Catherine Mathieu, Revue de l’OFCE, 104 (2008).

Notes on the Contributors

xv

Jorge Uxó González has a PhD in Economics from Universidad Complutense de Madrid. He is Associate Professor of Economics at the Universidad CEU San Pablo, Madrid, Spain, where he has been Dean of the School of Business and Economics and Chief of the Department of Economics. He teaches Macroeconomic Policies and Applied Economics. His main investigation lines are monetary and fiscal policies in EMU, activist fiscal policy rules, new consensus on macroeconomics and new normative macroeconomics. He has published a number of articles in specialized journals such as Applied Economics, Hacienda Pública Española, and Pensamiento Iberoamericano, and he is author of several contributions to books on economic policy. Marta Vázquez Suárez has a PhD in Economics from Santiago de Compostela University, Spain. After her graduation in Economics Sciences in 2002, she received a scholarship to do a postgraduate course, combining it with teaching economics and with her investigation about foreign investment, trade and growth in Spain for the Santiago University. Afterwards she oriented the research towards the developing economies, combining the investigation work carried out at universities in Mexico, Spain and Italy.

This page intentionally left blank

Introduction Jesús Ferreiro, Giuseppe Fontana and Felipe Serrano

In contrast to what happened during the golden age of Keynesian policies in the 1950s and 1960s, today fiscal policy has lost relevance, both in the short term as an instrument to stabilize the business cycle, with monetary policy gaining relevance, and in the long term with supply-side policies playing a predominant role. Two reasons explain the downgrading of fiscal policy. The first reason is a practical one, namely, the need to reduce the high levels of public deficit and debt, fiscal imbalances generated by the mix of increases in public spending and cuts in taxation and a tight monetary policy leading to high real interest rates. The second reason is related to the dominance of neoclassical theories, which maintain that both fiscal imbalances and the large size of public revenues and expenditures have a negative impact on economic activity and growth in the short and long term. The institutional framework and the rules of fiscal policy operating in the European Union (EU) are built upon this theoretical basis. The Maastricht Treaty has imposed the reduction of fiscal imbalances as the main objective of national fiscal policies. The fear of a relaxation of fiscal discipline has led to the approval of the Stability and Growth Pact (SGP). The SGP has lowered the discretional leeway of fiscal policy in the short term, and in this way has, de facto, subordinated fiscal policy to monetary policy. However, this strategy has not escaped problems and criticisms. Thus, as a consequence of the massive violation of the criteria of excessive deficits, the Stability and Growth Pact was reformed in 2005, in order to give, at least in the short run, more flexibility to national fiscal policies. More recently, the worldwide turmoil in the financial markets has generated support for a high level of fiscal activism, in order to alleviate the negative impact on the economy. 1

2

Fiscal Policy in the European Union

Pari passu with these problems, the criticisms of the current framework of fiscal policy in the European Union are growing. The criticisms take different forms: the lack of formal coordination between a single European monetary policy, whose only objective is price stability, and several national fiscal policies, which are the only available tools to smooth economic fluctuations; the lack of a federal budget that could contribute to solving the asymmetric shocks in the European Union, thus alleviating the coordination problems arising from an increasing number of national fiscal policies (nowadays, 27); and, lastly, the dissatisfaction with the theoretical models underpinning the use of fiscal policies. The various chapters in this book discuss these issues and provide a critical perspective on the current role of fiscal policy in the European Union. Chapter 1, by Rui Henrique Alves and Oscar Afonso, argues that the current framework of fiscal policies in the European Union is far from what would be an ‘optimal’ fiscal policy from the perspective of the theory of fiscal federalism and the subsequent distribution of competences among the different levels of government. The result is that fiscal policies in the EU are inefficient in terms of their capacity for macroeconomic stabilization due, mainly, to three reasons: the problems with the definition and implementation of fiscal rules (like the SGP), the insufficient degree of coordination of national fiscal policies, and, lastly, the small size of the Community budget. Only by solving these problems will the effectiveness of fiscal policies in the EU be improved. However, these changes require a significant reform in the general institutional and political framework of the EU, a reform that deepens not only the economic but also the political integration of the EU. In Chapter 2, Sabrina Rostaing-Paris deals with the internal inconsistency of fiscal policies in EMU countries. On the one hand, active discretionary fiscal policies are rejected, leaving the working of automatic built-in stabilizers as the only tools to smooth economic fluctuations. On the other hand, the Maastricht Treaty and the Stability and Growth Pact involve the existence of a rule-based fiscal policy that limits the size of fiscal deficits. These objectives could be contradictory, thus leading to a potential trade-off between the economic and fiscal stabilizations. The empirical analysis made by Rostaing-Paris leads her to conclude that such a trade-off does exist in the European Monetary Union, and, therefore, if fiscal policy is to play a role in smoothing business cycles, fiscal rules (i.e. SGP) must be reformed, with the new fiscal rules setting objectives not in terms of the size of fiscal deficits but in terms of cyclically adjusted budget balances.

Introduction

3

Chapter 3, by Jorge Uxó González and M. Jesús Arroyo Fernández, argues that the existence of fiscal rules, such as those imposed by the Stability and Growth Pact, limits the effectiveness of fiscal policy to stabilize economic fluctuations. In as much as the EMU does not suffer asymmetric shocks this limitation would not be a serious problem. However, Uxó and Arroyo show the permanence of significant divergences among EMU countries in terms of the inflation rates and the position in the business cycle. In this situation, a single monetary policy can be destabilizing. Uxó and Arroyo support a more active fiscal policy, not only in terms of the working of the built-in stabilizers but also of a more active discretionary policy, as the more effective tool to correct the impact of asymmetric shocks in EMU. Chapter 4, by Jesús Ferreiro, Teresa García-del-Valle and Carmen Gómez, focuses on the analysis of the composition of fiscal adjustments and of public expenditures. The models of the expansionary fiscal consolidation argue that fiscal adjustments based on cuts in public expenditures are not only more lasting than those based on tax increases but also they can contribute to accelerate the economic growth in the long run. However, the public policy endogenous growth models argue that it is the composition of public expenditures (and revenues) that really matters for economic growth. Thus, for fiscal adjustments to be growth-enhancing, they should involve a recomposition of public spending towards productive expenditures. Ferreiro, García-del-Valle and Gómez argue that although EMU countries have based their fiscal adjustments on public expenditures, the quality of public expenditures has not improved since in most countries productive expenditures have declined. Chapter 5, by Gwenaëlle Poilon, analyses the impact of public capital on long-run economic growth. This is a relevant point in the analysis of the role to be played by fiscal policy in the EU. Thus, the Lisbon Agenda strongly supports public investments, mainly in infrastructures and new technologies, as a driving force for economic growth, and many economists and politicians argue in favour of a golden rule for public finances. The chapter reviews the empirical research on the impact of public capital spending on economic growth, focusing on the methodological issues and the treatment of econometric problems surrounding public capital estimations. For Poilon the empirical analysis of the impact of public capital and investment suffers from serious mis-specifications leading to biased results. To solve these problems, mainly that of endogeneity, Poilon makes use of dynamic panel estimation techniques. With that technique she concludes that public capital does have a productive role. The size of the estimated coefficient is also more realistic than

4

Fiscal Policy in the European Union

those obtained in the literature and the results turn out to be robust. This empirical result confirms that the link between public capital and growth is not only theoretical but also empirical, justifying the priority given in the Lisbon Agenda to public investment. In Chapter 6, Jérôme Creel, Guy J. Gilbert and Thierry Madiès present an overview of public deficits and debts in France over the last three decades with a special focus on the problem of sustainability of public finances. Their contribution identifies the causes of deficits and debt accumulation, concluding that fiscal imbalances in France are not only explained by economic reasons (the gap between real interest rates and the rate of economic growth, and the low economic growth registered in the 1990s) but also for political reasons, that is because of the implementation of pro-cyclical policies related to the electoral cycle. Finally, the authors deal with changes made to improve short-term fiscal imbalances and show that recent structural measures have been undertaken in order to enhance the transparency of the budgetary process to meet public deficit and public spending targets and, consequently, to increase the efficiency of fiscal policy. Catherine Mathieu and Henri Sterdyniak argue in Chapter 7 that since the launch of the euro, persistent and even rising disparities among member states have made it difficult to implement short-term or structural common economic policies. The chapter gives an overview of euro-area disparities in terms of growth and inflation and imbalances, mainly unemployment and current accounts. For the authors four elements explain these disparities: the benefits of the single currency for catching-up countries, the weaknesses of the euro area economic policy framework; the implementation of non-cooperative domestic policies which have induced excessive competition and insufficient coordination and hurt mainly the larger economies; and the crisis of the European continental model in a global world. The existence of binding fiscal rules was based on the assumption that EMU would lead to economic convergence among member states that would reduce the need for active fiscal policies. Four strategies are discussed: increasing market flexibility; moving towards the knowledge society of the Lisbon Agenda; renationalizing economic policies; and introducing a more growth-oriented policy framework. In any case, the role of national fiscal policies should be enhanced, setting objectives in terms of achieving high rates of economic growth under the restriction of maintaining public expenditures and taxation at levels compatible with the current European social model. The final chapter, by Carlo Panico and Marta Vázquez Suárez, analyses from a post-Keynesian perspective the problems of coordination

Introduction

5

between monetary and fiscal policies in the euro area. For Panico and Vázquez the coordination of macroeconomic policies is necessary to deal with both cyclical problems and structural problems. The authors argue that despite the need for a true cooperation between monetary and fiscal policies, the current institutional framework of the euro area does not favour it. On one side, the European Central Bank defends its independence in order to counteract the supposed deficit-bias of national fiscal policies. On the other side, the national political authorities resist giving up their autonomy in smoothing the business cycle via the management of fiscal policies, since the only objective of the single monetary policy is price stability. This lack of coordination has negative effects both in terms of the economic performance of EU and EMU economies and in the legitimacy of the construction of the European Union. The proposal of Panico and Vázquez is to reform the current organization of policy coordination in the euro area, whose central element is the transformation of the Eurogroup into a Euro Area Fiscal Agency, an institution able to combine political and technical abilities and endowed with formal decision authority on matters regarding policy coordination. The chapters in this book were first presented at the special session on fiscal policy at the 4th International Conference ‘Developments in Economic Theory and Policy’ (Bilbao, 6–7 July 2007), an annual conference organized by the Department of Applied Economics V of the University of the Basque Country, Spain, in collaboration with the Cambridge Centre for Economic and Public Policy (CCEPP), Department of Land Economics of the University of Cambridge, UK. We thank the participants and contributors to the conference, and the University of the Basque Country that hosted the event. Financial support from the University of the Basque Country, International Papers in Political Economy, the Basque government and the Círculo de Empresarios Vascos is gratefully acknowledged. Finally, special thanks must go to Amanda Hamilton and Alec Dubber at Palgrave Macmillan, who first suggested this volume, and for their encouragement and support throughout.

1 Fiscal Federalism in the European Union: How Far Are We? Rui Henrique Alves and Oscar Afonso

1 Introduction Almost 60 years have gone by since Robert Schuman’s famous speech (Schuman, 1963), in which he described the ‘European federation’ as the ultimate goal of the integration process that was beginning at that time. However, the results of this process are clearly different for the two perspectives that comprise it: economics and politics. In economic terms, advances have been made at a good pace, and the European Union (EU) is now in the most advanced state of economic integration possible, with the single currency, free movement of goods, services, capital and people and several common policies. However, these advances have not been matched in the political field, where the Union has, in fact, failed to establish a strong entity and to develop common actions with a similar impact. This divergence is a fundamental trait of the present EU situation, which might be characterized as a crossroads. Along with the abovementioned advances, there are four aspects in which the results fall well short of what would be desirable, or four fundamental deficits in: (i) ‘competitiveness and economic growth’, even if the situation seems to be changing recently; (ii) ‘political weight’, with the EU often behaving as a ‘dwarf’ on the international political scene; (iii) ‘legitimacy and participation’, in view of the poor scrutiny of some European institutions, ambiguity in the sharing of competences between member states and the EU and indifference of citizens facing the process of integration; and (iv) ‘capacity for decision and action’, faced with unsuccessful institutional reforms and the permanence of a poor-sized community budget. These ‘deficits’, particularly the last three, show that the goal of creating an expanded space of European solidarity, which would give rise to 6

Fiscal Federalism in the European Union

7

a political entity and a true European citizenship, remains well out of the EU’s reach. Fifty years after the Rome Treaty, the weight of national interests is often still predominant, which means that these ‘deficits’ will be difficult to overcome unless there is a significant change in the Union’s politico-institutional organization model. That is to say, unless there is a rebalance in the two sides of the integration process, with a deepening of the political field. In a previous study (Alves, 2007) we argued that this rebalance would involve the EU becoming an organization capable of dealing efficiently with the need for unity in clearly supranational fields, without endangering European diversity, and that the most suitable model for doing so would be that of a ‘Federation of Member States’, with highly decentralized competences. Such an evolution would have to be accompanied, at the level of economic organization, by a growing approximation to the ‘rules’ of the theory of fiscal federalism. This is the context in which this chapter is presented, as its main topic is the comparison between the present process of defining and implementing fiscal policies in Europe and the one that should result from the application of those ‘rules’, thus with a focus on the function of macroeconomic stabilization. If we add to this comparison some elements related to the functions of supply of public goods and services and redistribution and to the exercise of democratic legitimacy, it becomes possible to analyse the gap between the method of economic organization in the European Union and that which should correspond to a ‘Federation of Nation-States’. That is the overall aim of this chapter. It begins with a brief look at the literature on fiscal federalism, establishing its fundamental traits and possible implications for the European case (section 2). The chapter continues by presenting the essential elements of the European Union’s economic organization, with particular emphasis on the discussions concerning fiscal discipline and the size of the community budget (section 3). The path is then open to analyse the essential question: how far is this economic organization from that resulting from the theory of fiscal federalism (section 4)? It concludes with some final remarks and implications for the future (section 5).

2 The theory of fiscal federalism: what major lessons for the EU? As Oates (1999) points out, the use of the term ‘federalism’ in economics is somewhat different to its normal use in political science. In the latter,

8

Fiscal Policy in the European Union

it refers to a political system with a constitution that guarantees a set of principles and proceeds to the sharing of competences between the various levels of power. In the case of economics, all public sectors are relatively ‘federalized’, given that all of them provide public goods and services and have some autonomy of decision: it deals, then, essentially with the questions that involve the vertical structuring of the public sector. In this context, the fundamental aim is to find the most suitable way of sharing responsibilities and of using instruments through the various levels of ‘government’, so as to optimize their performance. As there are clearly no rules or rigid formulas that determine a situation of ‘fiscal optimum’, which is highlighted by the diversity of fiscal structures in the various federations (similarly, in fact, to the case of political and institutional structures), the literature in the framework of so-called ‘fiscal federalism’ has attempted to find some guidelines for the vertical structuring of government. The essential purpose of this literature is, therefore, the suitable sharing of competences among the various levels of government (and not, as it may sometimes seem, fiscal decentralization for itself alone), or, as Oates (1998) says, identifying the institutional design that will best allow the public sector to respond to the variety of the demand aimed at it. Traditionally, the theory of fiscal federalism is concerned with three essential aspects: (i) the sharing of functions between the different levels of government (particularly at four levels (Spahn, 1994): supply of public goods and services; redistribution of income; macroeconomic stabilization; and taxation); (ii) the identification of welfare gains resulting from fiscal decentralization; and (iii) the use of the instruments of fiscal policy (particularly issues associated with taxation and inter-governmental transfers). Through time, the field of fiscal federalism has been broadened as other topics emerged, including, among others, questions related to inter-jurisdictional competition and ‘environmental federalism’ (Enrich, 1996; Oates and Schwabb, 1996), ‘market-preserving federalism’ (Weingast, 1995; McKinnon, 1997b) or decentralization in developing economies or those in transition (Bahl and Linn, 1992; Shah, 1994). Briefly analysing these fundamental topics from the theory of ‘fiscal federalism’, it is possible to find some elements that should be determinant in structuring the competences of the various levels of power within the EU context, particularly taking into account a possible evolution towards a federalist model. The main conclusions of the theory of ‘fiscal federalism’ seem largely compatible with the idea of the evolution of the EU towards a broadly

Fiscal Federalism in the European Union

9

decentralized federal model (the ‘Federation of Nation-States’) and for the need, in this context, to create a ‘European economic government’ that would be responsible for competences assigned, in this field, to the central level of power. Similarly, they seem to sustain the idea that it is possible to obtain significant welfare gains through the creation of such a strongly decentralized federal fiscal system, provided that it is suitably designed and taking into account aims of equity and efficiency. In this design of the Federation’s vertical structure at the fiscal and budgetary level, the aspects below would be decisive for the success of the model, taking into account the main elements of the available literature. In the first place, the principle of subsidiarity should be applied in a clear and transparent way, in what concerns the supply of public goods and services. In this area, centralization should only occur for a small number of policies: those that have clear supranational nature (such as defence, security and monetary policy, among other fields). This would avoid too much ‘central’ intervention (Oates, 1972), something that may have happened in the European case within the present institutional model. In this field, the (still) marked diversity in demands and national preferences, plus the (still) low mobility of families should not put the gains of decentralization at risk (Spahn, 1994): they will probably even highlight them, since they will not cause exaggerated distortions (Flatters et al., 1974). In fact, even if there were spillover effects, the advantage of cooperation among the different levels of government that are closest to citizens could easily outweigh the centralized solution. Still within the context of competence sharing in the scope of the supply of public goods and services, despite the validity of the decentralization principle, there seems to be a need, in some fields, to take particular care with certain negative consequences resulting from competition between the Federation members, as it eventually would lead to poorer standards (Enrich, 1996). In the European case, this would be particularly noticeable in the fiscal and environmental fields, where these consequences have recently been established, and would justify greater centralization and harmonization. In the second place, at the level of the redistribution function, a combination of some centralization (Tiebout, 1956) and a significant space for decentralization would be an optimal solution. Various reasons would justify this significant space for decentralization: reduced geographic mobility; failure of some sub-national programmes of redistribution

10 Fiscal Policy in the European Union

(Feldstein and Wrobel, 1998, for the case of the USA); extended aims of the regional redistribution function (King, 1984); and the added concern with the most disadvantaged that are closer (Pauly, 1973). This last aspect would have an increased importance in the European case, since it deals with different countries (and regions), with different traditions, values and histories (or rather, once again diversity justifying some degree of decentralization). In this context, a summary of the main theoretical and empirical references in this field seems to suggest that, in the case of the European Union, the redistribution policy would be maintained at the national level, particularly with regard to individual redistribution, while there would also be some space for inter-regional redistribution, namely via transfers through the community budget. Thirdly, the literature on fiscal federalism points towards an increased importance of the fiscal policy for the purposes of macroeconomic stabilization. It traditionally postulates the importance of a significant central budget, which, through the transfer mechanisms between the states/regions positively affected by asymmetric shocks and the states/regions negatively affected by the same shocks, seems to exercise an important degree of stabilization (Spahn, 1994). This has, in fact, been established through several studies, in the wake of the analyses of Sala-i-Martin and Sachs (1992), Italianer and Pisani-Ferry (1994) and Bayoumi and Masson (1995): these authors estimated, based on different methodologies, a significant degree of stabilization in the absorption of shocks by the North American federal budget. The same type of role was mentioned by subscribers to the theory of optimal currency areas (following the seminal works of Mundell, 1961 and Kenen, 1969). They considered that, once the monetary and exchange instruments were lost, an efficient response to the negative effects of specific and asymmetric shocks, in the context of a monetary union, would only be obtained through one of three mechanisms: broad flexibility of prices and salaries; strong mobility of labour; or fiscal transfers via a wide central budget. In the case that these do not exist (or, at least, only in a weak situation), the solution would be to move to national fiscal policies with high flexibility. As a result of these elements, and with regard to the EU, it becomes relevant to discuss the best way of pursuing aims of macroeconomic stabilization, knowing the difficulties at the level of labour mobility and the flexibility of some labour markets, as well as the significant difficulties of a political nature in promoting both a broadening of centralization and expansion of the Union budget.

Fiscal Federalism in the European Union

11

In the fourth place, at the level of the instruments of fiscal federalism, it is worth highlighting the existence of a set of relevant elements: the existence of certain guidelines (namely, the criteria defined by Musgrave, 1983) for a potential design of a ‘European’ fiscal system, despite criticisms of the traditional criteria and the obvious political difficulties (Alves, 2000); and the need to promote conditional transfers for internalizing spillover effects (Oates, 1999). There is also a need to take into account the problems, namely political ones, generated by transfers, whose aim is that of ‘fiscal equalization’ (Usher, 1995; McKinnon, 1997b) and, as such, the concern with a certain trade-off between the goals of greater homogeneity of the levels of economic growth and economic and social cohesion and the problems deriving from the existence of taxpayers and net receivers. Finally, there is a need for some care in constructing the mechanism(s) of ‘income sharing’, without the associated transfer system(s) being too broad, so as not to encourage increasing budgetary laxity. In the fifth place, some literature, following the pioneering work by Inman and Rubinfeld (1997) and considering both economic and political goals, seems to highlight the idea favourable to a largely decentralized federal system, indicating the possibility of strengthening citizens’ political participation, which could overcome the possible costs associated with a reduction in economic efficiency. This could be particularly relevant in the European case, where there is a significant challenge in the field of legitimacy and democratic participation. Finally, some literature also indicates that, under certain conditions, federalism could constitute the best system for preserving and developing the market economy (Weingast, 1995). Taking the European case, this might point towards a situation where the creation of a Federation would positively contribute towards strengthening European competitiveness. In particular, it points to the fact that there seems to be a need for special attention to the problem of fiscal discipline in subcentral governments (in this case, of the member states). In any case, it shows that it could be enough to combine the prohibition of monetary financing of the debt and bail-out behaviours on the part of the federal government with the non-existence of exaggerated inter-governmental transfers and with the efficient functioning of credit markets to generate responsible behaviour in the sub-central fiscal authorities (McKinnon, 1997a). In other words, applied to the European case, it indicates that it might not be necessary to define rules like those resulting from the Maastricht Treaty and the Stability and Growth Pact (SGP; European Council, 1997).

12 Fiscal Policy in the European Union

3 Fiscal policy in the EU: the present situation Having analysed some major lessons that the theory of fiscal federalism may contain for the EU’s economic organization, and before moving on to assess the gap between the present situation and that which would result from an application of this theory, a brief description of the present organizational context of the euro zone is relevant. This is done below, firstly by approaching the issue of the European solution for fiscal policies, and then with a brief reference to the issue of community budget size. In terms of fiscal policy (and, therefore, the macroeconomic stabilization in view of specific or asymmetric shocks), the European solution was provided for in the Treaty on European Union (1992). This foresaw that fiscal policies would remain in the hands of national governments, albeit limited by compulsory rules (particularly, restricting the public deficit to no more than 3 per cent of GDP and the public debt to no more than 60 per cent of GDP), complemented by their coordination at the level of the Council. It also foresaw the prohibition of monetary financing of public deficits and instituted a clause of national responsibility for the public debt (no bail-out). This solution was designed to maintain a policy instrument that could be used at national level and, at the same time, to prevent excessive public deficits from being created and maintained (through restrictive rules) and to promote some coherence among the various national fiscal policies and between these and the single monetary policy (through a coordination mechanism). In 1997, the SGP came to reinforce this option, particularly in terms of limiting freedom of activity. Thus, its preventive mechanism assumed budgetary balance as a medium-term goal, allowing automatic stabilizers to act and opening up room for manoeuvre with some discretion in handling fiscal policy, namely when a less favourable economic evolution occurs. On the other hand, its corrective mechanism established in a more concrete way the mode of operation for ‘excessive deficit procedure’, in particular defining the sanctions to be imposed and clarifying situations of exception. This solution has been the target of wide discussion and great criticism, at both the academic and political level, particularly prior to 1995 and after 2000, periods in which there was greater economic difficulty. The discussion was mostly focused around the way fiscal discipline would be implemented and controlled (Buiter et al., 1993; Rubio and Figueras, 1998), although the need for such discipline was also questioned by

Fiscal Federalism in the European Union

13

some authors, namely those who called for more radical reforms along Keynesian lines (Arestis et al., 2001; Arestis and Sawyer, 2003). The need for fiscal discipline was linked to the need to preserve the stability of the monetary union, which would be compromised if countries promoted excessive public deficits (De Grauwe, 2005). This situation could determine significant external effects, namely via an increase in the Union’s interest rate and possible pressures on the central bank in the sense of making monetary policy more flexible. On the other hand, the possible incentive to free-riding behaviour determined by the fact that everyone in a single currency context could share the costs of bad budgetary behaviour, and the possibility of creating excessive deficits for political reasons, would constitute additional elements in favour of a solution that would promote fiscal discipline. But the way to implement and control fiscal discipline has been the subject of major controversy, both among defenders of the present rules (Buti et al., 2003; Begg et al., 2004), and defenders of relatively profound reforms (Casella, 1999; Creel, 2003; Collignon, 2004; Pisani-Ferry, 2004; Wyplosz, 2005). Criticism of the original SGP reached a peak in 2002 when the President of the European Commission at the time, Romano Prodi, classified it as ‘stupid’ (Prodi, 2002). The critical voices, which reappeared particularly by the beginning of this century, suggested greater flexibility in rules and a greater balance between nominal and real aims, namely because of: (i) the possibility, in a situation of economic crisis, of governments having to use restrictive fiscal policies, which would be counterproductive; (ii) the fact that continued situations of stagnation or poor economic growth were not considered as exceptions regarding the application of the excessive deficit procedure; (iii) the possibility that the time period for correcting excessive deficits was clearly too short; and (iv) the possibility of a differentiation of fiscal rules by taking into account the level of economic development and the economic dimension of countries (for example, considering a link between R&D expenses and economic growth, Afonso and Alves, 2008, argue for such differentiation). This discussion, and above all the economic difficulties felt by several countries at the beginning of the twenty-first century (particularly France and Germany, the ‘locomotives’ of the euro zone), led in 2003 to a suspension in the application of the SGP (European Council, 2003) and two years later to its reform (European Council, 2005). The major features of such reform included: (i) an extension of the deadline for correcting excessive deficits; (ii) the need for greater attention to the evolution of the structural deficit and the weight of public debt on the GDP as central

14 Fiscal Policy in the European Union

elements of fiscal sustainability in the medium and long run; (iii) the consideration of continued situations of low effective product growth (below potential) as exceptions to sanctions; and (iv) the inclusion of various ‘pertinent’ factors that could ease situations that would fit into the concept of excessive public deficit. These changes have allowed greater room for manoeuvre for national governments to deal with specific or asymmetric shocks, in this sense making the Pact more ‘flexible’, considering the classification of ‘ideal’ fiscal rules proposed by Kopits and Symansky (1998). However, the inclusion of a great diversity of ‘pertinent’ factors that may justify an excessive deficit could make the Pact less ‘enforceable’ (Alves and Afonso, 2007). In other words, new doubts have been raised as to its capacity to ensure the sustainability of public accounts in the euro zone, which are shared both by defenders of the original SGP (Buti et al., 2005), and by its critics (Buiter, 2005; Allington and McCombie, 2007). Another central element is the situation of the community budget in terms of its size, uses and outlooks. The first important thing to note is that the EU budget maintains a very small weight, representing little more than 1 per cent of the Union’s GDP, in a situation that contrasts greatly with those of federations (or similar political units) with a single currency, particularly with that of the United States. Secondly it is possible to argue that not only is the budget small, but also the major resources seem to be not truly ‘own’, contrary to what their name suggests. In fact, most of the revenue comes from the so-called ‘GNP resource’ and the weight of this resource has increased over the last few years, in view of the successive drop in the importance of customs duties and agricultural duties and the stagnation followed by the drop in relevance of the ‘VAT resource’. As this major resource would not fulfil the criteria for being characterized as a true ‘own resource’ (Cieslukowski and Alves, 2006), the EU remains in a situation of very limited financial autonomy. Finally, it is equally significant that a very substantial part of this small budget is devoted to one of the most controversial (and probably most unfair) policies in the Union: the Common Agricultural Policy (CAP). Despite its successive decline in importance in the budget, in particular after 1985, in favour of the policies for economic and social cohesion, the CAP continues to represent almost half of the community budget’s expenses. This leaves very little room for manoeuvre for an increased common intervention in other areas. These circumstances do not seem likely to change in the short run, as can be seen from the analysis of the financial perspectives for the period 2007–13 (Table 1.1), with a very low ceiling maintained for budget size

Fiscal Federalism in the European Union Table 1.1

15

Financial perspectives for the period 2007–2013

Values in euro billion, 2004 prices

Sustainable growth Preservation and management of natural resources Citizenship, freedom, security and justice EU as a global player Administration Compensations Total appropriations for commitments (% GNP) Own resources ceiling (% GNP)

2007

2013

Value

%

Value

%

51090 54972

42.4 45.6

57841 51145

45.7 40.4

1120

0.9

1910

1.5

6280 6720 419 1.10

5.2 5.6 0.3

8070 7680 0 1.00

6.4 6.1 0.0

1.24

1.24

Source: EU site, http://europa.eu.int.

(less than 1.3 per cent of the Union’s GNP) and significant expenses in the area of agriculture. Some reformulation of fundamental aims can be seen, particularly for competitiveness and cohesion, following the attempt at real implementation of the Lisbon Agenda, but keeping expenses with new challenges in the Union (such as foreign policy and common defence, the development of citizenship, the strengthening of the space of freedom, security and justice) at a very disappointing level.

4 Fiscal federalism in the EU: how advanced are we? If we compare the solution adopted in the context of the Maastricht Treaty and its subsequent reforms with the main ideas highlighted in the review of the literature on fiscal federalism, there seems to be quite a large gap between the present EU framework with regard to the performance in the field of macroeconomic stabilization (and, in general, between the definition and execution of fiscal policies) and what should occur in the context of a true Federation. This discrepancy is even greater in three areas: 1. The community budget is small – about 1 per cent of the Union’s GDP – and no significant changes seem to be expected in it for the next few years. Added to the fact that almost half the budget is destined for the Common Agricultural Policy, this means that it cannot contribute to the functions of macroeconomic stabilization that central budgets assume in the main federations.

16 Fiscal Policy in the European Union

2. Some more recent developments in the theory of fiscal federalism highlight the relevance of the creation of ‘hard budget constraints’ as a way of preserving and developing the market economy, in the framework of a highly decentralized federation (Weingast, 1995; McKinnon, 1997a). Therefore, there could be greater synchronization between the theory of fiscal federalism and the solution adopted for defining and executing fiscal policy in the EU, which is based particularly on the development and application of compulsory restrictive rules. However, both Weingast and McKinnon admit that it could be enough to combine the prohibition of monetary financing of the debt and bail-out behaviour on the part of the federal government with the non-existence of exaggerated intergovernmental transfers and with more efficient functioning of the credit markets in order to generate responsible behaviour of the sub-central fiscal authorities. In the European case, however, as these prohibitions exist, compulsory rules have also been adopted. Nevertheless, as mentioned above, such rules are not consensual and do not seem to be transparent enough or applicable in such a way as to promote fiscal discipline (or ‘hard budget constraints’). 3. The coordination of non-monetary policies in the EU seems to be insufficient, in a situation that is hardly favourable to obtaining coherence among the various national fiscal policies and between these policies and the common monetary policy. In the context of a federal economic organization, this coherence would be largely guaranteed, even in the context of broad decentralization of policies, since there would at least be institutions that guaranteed the definition of common fundamental aims and of general paths to follow, as well as effectively overseeing compliance with the rules. Considering this discrepancy and admitting that fiscal policies could still be mobilized for macroeconomic purposes (Solow, 2004) and that it seems difficult, in the short (and possibly medium) run to pursue this aim in a (traditional) framework of centralization, with a wider community budget, the debate mentioned in the previous section and the lessons of the fiscal federalism theory could indicate the following lines of evolution, which differ according to the time horizon: 1. In the medium to long run, and in order to bring about and consolidate the politico-institutional transformation that would lead to the creation of a ‘Federation of Nation-States’, it would be difficult not to

Fiscal Federalism in the European Union

17

keep moving towards an effective reform of the community budget. Such reform would enable its expansion to levels compatible with the new demands and new challenges facing the Union and, in particular, to develop an effective mechanism for stabilization regarding adverse shocks. Studies recently carried out by the European Commission (2004) could, in this field, constitute a good starting point, both from the point of view of the type of instruments that could shape new own resources for the community budget, and from that of a summary of the main problems that would arise in order to implement them. Such problems would be associated, on the one hand, with questions of fiscal harmonization and, on the other hand, with the necessary substitution of national fiscal burden (as well as national public expenses by EU expenses), as the only way in which the acceptance of European citizens could be considered. It is worth noting, however, that an enlargement of the European budget does not need to rely only on such substitution, as a more general strategy towards economic growth would lead to an increase in fiscal resources. Recently the President of the European Commission argued for a reopening of the budgetary debate with no taboos (Barroso, 2007). In any case, it is not to be expected, nor is it politically desirable (since it would possibly imply a situation of excessive centralization), that the community budget should be, even in the long run, similar in size to the central budgets of certain existing federations (e.g. that of the United States). In fact, some studies even prior to the adoption of the Maastricht programme for the single European currency, already anticipated this situation: MacDougall (1977) mentioned around 5–7 per cent of the GNP; Lamfalussy (1989) referred to about 3 per cent of the GNP. In either case, however, these values are very different from those seen today. 2. In the short run, and bearing in mind the political and economic difficulties arising from what was mentioned in the previous point, a more efficient resolution of the fundamental issues raised (namely, efficiency in combating the negative effects of specific and asymmetric shocks, the balance between nominal and real macroeconomic aims, the strengthening of the coherence of policy mix and the maximization of overall well-being) would seem to suggest three more immediate actions: (a) The most immediate would be a more credible reform of the Stability and Growth Pact, as a relevant mechanism for the supervision

18 Fiscal Policy in the European Union

and maintenance of fiscal discipline. Concerning this point, the 2005 reform seems to have raised the SGP’s degree of flexibility, but also to have decreased its credibility in terms of its sanctionary nature, in view of the excessive number of causes that may prevent its application. In this sense, an adjustment of the existing characteristics, namely considering a more restrictive and explicit list of ‘escape clauses’, together with the integration of some measures proposed over the last few years by different economists would seem to be essential elements. These measures would include the proposal to evolve towards a pact based on public debt sustainability also as an incentive to effective coordination. (b) The coordination of national fiscal policies, requiring a marked change in their institutional framework, should be strengthened and there should be an increasing coherence between national fiscal policies and the common monetary policy. As mentioned by Pisani-Ferry (2002), it would be important to adopt a code of conduct for economic policy and to establish a compulsory agreement of reciprocal consultation for members of the euro zone and the Commission before taking the relevant macroeconomic policy decisions. It would also be important to transform the Eurogroup into an executive entity with decisionmaking capacity through a qualified majority and to transform the (national) stability programmes into true instruments of coordination and supervision. Finally, it would be important to create constructive dialogue between the Eurogroup and the European Central Bank, allowing useful and coherent interaction in terms of structural reforms and macroeconomic policy. In this line, a stronger way of change would follow the arguments of Hein and Truger (2005) and Hein and Niechoj (2007). They argued for an alternative policy model, based on the coordinated development of monetary, wage and fiscal policies, as a way to achieve steady growth and high employment whilst maintaining price stability. In such a case, an effective coordination would even eliminate the need for fiscal discipline rules in order to avoid the external negative effects from excessive deficits on one country. (c) While a more significant expansion of the community budget is not possible, the effectiveness of the stabilizing response to shocks of a specific or asymmetric nature may require a limited mechanism of shock absorption to be created.

Fiscal Federalism in the European Union

19

Following Goodhart and Smith (1993), in order to be efficient and not to place excessive difficulties of implementation at various levels, such a mechanism should meet several requirements. Firstly, it should be limited so that community action would only occur in the event of serious economic difficulties. Secondly, it should be temporary so that it would not promote dependence and maintenance of the status quo rather than stabilization: this means that its only source of activation should be negative changes in economic activity, and it should be suspended as soon as these changes cease. In the third place, its impact should only be produced during the stage of minor economic growth, meaning that the mechanism should be based on an indicator closely associated with fluctuations in real income. Finally, the principle of subsidiarity should be effective, meaning that help should only occur where the deceleration of economic activity is caused by specific national factors. In this context, it should also be noted that, since the beginning of the 1990s, several attempts have been made to create a ‘European scheme of fiscal transfers’, which have been aimed at these kinds of goals (e.g. Melitz and Vori, 1993; Italianer and Pisani-Ferry, 1994; Hammond and von Hagen, 1998). Although these have been unsuccessful in terms of policy adoption, some of them contain significant possibilities to be explored in the framework that we have just defended. Also with regard to this, and in a similar way to that mentioned in section 2 relative to ‘income sharing mechanism(s)’, if this transfer system existed, it should not be too extensive, otherwise it could put fiscal discipline at risk, by encouraging increasingly lax behaviour at the level of public accounts. In addition, as a result of the lessons of fiscal federalism, evolution should be marked, at a more general level, by making legal provisions for a suitable and transparent sharing of competences at the various levels of government. In this context, this would ideally result from a true European Constitution, which would fully provide for the principles of decentralization and of subsidiarity, explicitly placing only the clearly supranational fields (e.g. defence, foreign policy, monetary policy and fiscal competition) as ‘federal’ government competences. In these fields, the unanimous decision rule should be replaced with a simplified rule of qualified majority: this would mean that decisions

20 Fiscal Policy in the European Union

would be more likely to pass in the central organs of the Union and would reduce the possibility of creating blocking minorities, thus raising the Union’s decision-making and intervention capacity, as well as contributing to a reduction in its ‘democratic deficit’.

5 Conclusion The objective of this chapter has been to compare the present process of definition and implementation of fiscal policies in the EU with the main conclusions of fiscal federalism theory, in order to draw possible lessons for the EU’s future evolution, namely towards a possible ‘Federation of Nation-States’. This comparison supports the idea that the major conclusions of such theory are broadly compatible with the emergence of a largely decentralized ‘Federation’ in Europe, including the possibility of running significant positive welfare effects. However, the comparison also shows that there is still a large gap between the present system and the one that would result from a wide application of the fiscal federalism theory. The gap is particularly relevant in the domain of macroeconomic stabilization, where three elements should be stressed: (i) a deficient definition and capacity of application of fiscal rules (even after the reform of the SGP); (ii) an insufficient degree of coordination of national fiscal policies; and (iii) the lack of financial autonomy for the centre, as the community budget has a very limited dimension. It seems unlikely that this situation will change in the near future, as the new Lisbon Treaty does not include important alterations in this field, in line with what already occurred in the failed Constitutional Treaty. However, it is important to modify this situation, with changes happening at two temporal levels. Firstly, in the short run, a more credible reform of the SGP is required, in order to keep it as a relevant mechanism for the maintenance of fiscal discipline. In our view, fiscal discipline is still an important matter within the monetary union, as some countries (especially the bigger economies, see Afonso and Alves, 2008) could hurt the general welfare if they created excessive deficits. Secondly, as some of the literature on fiscal federalism, and/or some post-Keynesian studies, seems to suggest, a strengthening of the coordination of national fiscal policies (requiring a marked change in its institutional framework), together with a more profound coordination between these and the monetary policy, would be the most important change in the short run. The creation of a limited mechanism of shock absorption would also be relevant, as it would help to increase the effectiveness

Fiscal Federalism in the European Union

21

of response to asymmetric shocks, whose occurrence may have been reduced, but not eliminated, by the creation of the monetary union. In the medium to long run, however, only a significant increase in the dimension of the EU budget would give the EU the degree of financial autonomy necessary to face its new challenges and demands. In addition, as a result of the lessons of fiscal federalism, evolution should be marked, at a more general level, by making legal provisions for a suitable and transparent sharing of competences at the various levels of government. This would ideally result from a true European Constitution, which would fully provide for the principles of decentralization and subsidiarity, explicitly placing only the clearly supranational fields (e.g. defence, foreign policy, monetary policy and fiscal competition) as ‘federal’ government competences. In these fields, the unanimous decision rule should be replaced with a simplified rule of qualified majority. This would mean that decisions would be more likely to pass in the central organs of the Union and would reduce the possibility of creating blocking minorities. Thus the Union’s decision-making and intervention capacity would increase, together with an important contribution to a reduction in its ‘democratic deficit’. The unsuccessful project of the Constitutional Treaty included several changes that would put the present situation closer to the one defended in this chapter, namely the elimination of the three-pillar institutional structure, the increase in the number of areas with majority voting, the alteration of the rules for qualified majority and the attempt to present a distribution of competences between the members and the Union. Some of these changes have been included in the Lisbon Treaty, presently in ratification. However, the non-existence of significant modifications in the economic field, the maintenance of some critical aspects of the Constitutional Treaty (Alves, 2007), the elimination of measures that could lead people to imagine a reinforcement of political union, and even the lack of transparency in the negotiation process still leave the situation far from what would be desirable. References Afonso, O. and R. H. Alves, ‘Endogenous Growth and European Fiscal Rules’, Applied Economics (2008 forthcoming). Allington, N. and J. McCombie, ‘Nonsense Upon Stilts: an Assessment of the European Stability and Growth Pact and How it May be Reformed’, in John McCombie and Carlos Rodriguez (eds), The European Union (Basingstoke: Palgrave Macmillan, 2007), 26–49. Alves, R. H., Políticas Fiscais Nacionais e União Económica e Monetária na Europa, 2nd edition (Lisboa: Bolsa de Valores de Lisboa e Porto, 2000).

22 Fiscal Policy in the European Union Alves, R. H., ‘European Union and (Fiscal) Federalism’, in John McCombie and Carlos Rodriguez (eds), The European Union (Basingstoke: Palgrave Macmillan, 2007), 154–72. Alves, R. H. and O. Afonso, ‘The New Stability and Growth Pact: More Flexible, Less Stupid?’, Intereconomics, 42(4) (2007): 218–25. Arestis, P. and M. Sawyer, ‘Macroeconomic Policies of the Economic and Monetary Union: Theoretical Underpinnings and Challenges’, International Papers in Political Economy, 10 (2003): 1–50. Arestis, P., K. McCauley and M. Sawyer, ‘An Alternative Stability Pact for the European Union’, Cambridge Journal of Economics, 25 (2001): 113–30. Bahl, R. and J. Linn, Urban Public Finance in Developing Countries (Oxford: Oxford University Press, 1992). Barroso, J. M. (2007) ‘Making the Right Choices for Europe’, available at: http://ec. europa.eu/commission_barroso/president/pdf/speaking_points_budget_en.pdf Bayoumi, T. and P. R. Masson, ‘Fiscal Flows in the United States and Canada: Lessons for Monetary Union in Europe’, European Economic Review, 39 (1995). Begg, I., M. Buti, M. Weale, H. Enderlein and W. Schelkle, ‘Symposium Reforming Fiscal Policy Co-ordination under EMU: What Should Become of the Stability and Growth Pact?’ Journal of Common Market Studies, 42(5) (2004): 1023–59. Buiter, W., ‘The “Sense and Nonsense of Maastricht” Revisited: What Have We Learnt about Stabilization in EMU?’ CEPR Discussion Paper No. 5405 (2005). Buiter, W., G. Corsetti and N. Roubini, ‘Excessive Deficits: Sense and Nonsense in the Treaty of Maastricht’, Economic Policy: a European Forum, April (1993): 58–100. Buti, M., S. Eijffinger and D. Franco, ‘Revisiting EMU’s Stability Pact: a Pragmatic Way Forward’, Oxford Review of Economic Policy, 19(1) (2003): 100–11. Buti, M., S. Eijffinger and D. Franco, ‘The Stability Pact Pains: a Forward-Looking Assessment of the Reform Debate’, CEPR Discussion Paper No. 5216 (2005). Casella, A., ‘Tradable Deficit Permits: Efficient Implementation of the Stability Pact in the European Monetary Union’, Economic Policy, 29 (1999): 323–61. Cieslukowski, M. and R. H. Alves, ‘Financial Autonomy of the EU after the Enlargement’, FEP Working Papers, 217, July (2006). Collignon, S., ‘The End of the Stability and Growth Pact?’, International Economics and Economic Policy, 1 (2004): 15–19. Creel, J., ‘Ranking Fiscal Policy Rules: the Golden Rule of Public Finance versus the Stability and Growth Pact’, Documents de Travail de l’OFCE 2003–4, Observatoire Français des Conjonctures Economiques, July (2003). De Grauwe, P., The Economics of Monetary Integration, 6th edition (Oxford: Oxford University Press, 2005). Enrich, P. D., ‘Saving the States from Themselves: Commerce Clause Constraints on State Tax Incentives for Business’, Harvard Law Review, 110 (1996): 378–461. European Commission, ‘Financing the European Union’, Commission Report on the Operation of the Own Resources System, COM (2004) 505. European Council, ‘Resolution on the Stability and Growth Pact’, Official Journal of the European Communities, C 236/1 (1997). European Council, ‘2546th Council Meeting Economic and Financial Affairs (Press Release)’, 14492/1/03 REV 1 (en), Brussels (2003). Available at: http://www. consilium.europa.eu/ueDocs/cms_Data/docs/pressData/en/ecofin/ 78051.pdf

Fiscal Federalism in the European Union

23

European Council, ‘Presidency Conclusions, European Council Brussels’ (2005), Concl 1, 7619/1/05. Available at: http://ue.eu.int/ueDocs/cms_Data/docs/ pressData/en/ec/84335.pdf Feldstein, M. and M. V. Wrobel, ‘Can State Taxes Redistribute Income?’, Journal of Public Economics, 68(3) (1998): 369–96. Flatters, F., V. Henderson and P. Mieszkowski, ‘Public Goods, Efficiency and Regional Fiscal Equalization’, Journal of Public Economics, 3 (1974): 99–112. Goodhart, C. and S. Smith, ‘Stabilization’, The Economics of Community Public Finance, European Economy, 5 (1993): 417–55. Hammond, G. and J. von Hagen, ‘Insurance against Asymmetric Shocks: an Empirical Study for the European Community’, The Manchester School, 66(3) (1998): 331–53. Hein, E. and T. Niechoj, ‘Guidelines for Sustained Growth in the European Union? The Concept and Consequences of the Broad Economic Policy Guidelines’, in John McCombie and Carlos Rodriguez (eds), The European Union (Basingstoke: Palgrave Macmillan, 2007), 5–25. Hein, E. and A. Truger, ‘Macroeconomic Coordination as an Economic Policy Concept – Opportunities and Obstacles in the EMU’, in Eckhard Hein, Torsten Niechoj, Torsten Schulten and Achim Truger (eds), Macroeconomic Policy Coordination in Europe and the Role of the Trade Unions (Brussels: ETUI, 2005), 19–67. Inman, R. and D. Rubinfeld, ‘The Political Economy of Federalism’, in Dennis Mueller (ed.), Perspectives on Public Choice: a Handbook (Cambridge: Cambridge University Press, 1997), 73–105. Italianer, A. and J. Pisani-Ferry, ‘The Regional Stabilisation Properties of Fiscal Arrangements’, in Jorgen Mortensen (ed.), Improving Economic and Social Cohesion in the European Community (New York: St. Martin’s Press, 1994), 155–94. Kenen, P. B., ‘The Theory of Optimum Currency Areas: an Eclectic View’, in Robert Mundell and Alexander Swoboda (eds), Monetary Problems of the International Economy (Chicago: University of Chicago Press, 1969). King, D., Fiscal Tiers: the Economics of Multi-Level Government (London: Allen & Unwin, 1984). Kopits, G. and S. Symansky, ‘Fiscal Policy Rules’, IMF Occasional Paper 162 (1998). Lamfalussy, A., ‘Macro-Coordination of Fiscal Policies in an Economic and Monetary Union in Europe’, in Report on Economic and Monetary Union in the European Community, Serviço das Publicações Oficiais das CE (1989), 91–125. MacDougall, D., Report of the Study Group on the Role of Public Finance in European Integration, 2 vols, Commission of the European Communities, Economic and Financial Series (1977). McKinnon, R., ‘EMU as a Device for Collective Action Retrenchment’, American Economic Review, 87 (1997a): 227–9. McKinnon, R., ‘Market-Preserving Fiscal Federalism in the American Monetary Union’, in Mario Blejer and Teresa Ter-Minassian (eds), Macroeconomic Dimensions of Public Finance: Essays in Honour of Vito Tanzi (London: Routledge, 1997b), 73–93. Mélitz, J. and S. Vori, ‘National Insurance against Unevenly Distributed Shocks in a European Monetary Union’, Recherches Economiques de Louvain, 59(1–2) (1993): 81–104.

24 Fiscal Policy in the European Union Mundell, R. A., ‘A Theory of Optimum Currency Areas’, American Economic Review, 51(4) (1961): 657–65. Musgrave, R., ‘Who Should Tax, Where and What?’, in Charles McLure (ed.), Tax Assignment in Federal Countries (Camberra: Australian University Press, 1983). Oates, W., Fiscal Federalism (New York: Harcourt Brace Jovanovich, 1972). Oates, W., ‘On the Welfare Gains from Fiscal Decentralization’, Working Paper 98–05, University of Maryland Economics Department (1998). Oates, W., ‘An Essay on Fiscal Federalism’, Journal of Economic Literature, 37(2), September (1999): 1120–49. Oates, W. and R. M. Schwabb, ‘The Theory of Regulatory Federalism: the Case of Environmental Management’, in Wallace Oates (ed.), The Economics of Environmental Regulation (Cheltenham: Edward Elgar, 1996), 319–31. Pauly, M., ‘Income Redistribution as a Local Public Good’, Journal of Public Economics, 2 (1973): 35–58. Pisani-Ferry, J., ‘Fiscal Discipline and Policy Coordination in the Eurozone: Assessment and Proposals’, Note for the GEA meeting of 16 April (2002). Pisani-Ferry, J., ‘Reforming the SGP: Does it Matter? What Should be Done?’ in Roger Liddle and Maria João Rodrigues (eds), Economic Reform in Europe: Priorities for the Next Five Years (Policy Network, 2004). Prodi, R., Interview with Le Monde, 17 October 2002. Rubio, O. and D. Figueras, ‘Federalismo y Unión Monetaria en Europa’, Instituto de Estudios Fiscales, P.T. no. 10/98 (1998). Sala-i-Martin, X. and J. Sachs, ‘Fiscal Federalism and Optimum Currency Areas: Evidence for Europe from the United States’, in M. Canzoneri, V. Grilli and P. Masson (eds), Establishing a Central Bank: Issues in Europe and Lessons from the US (Cambridge: Cambridge University Press, 1992). Schuman, R., Penser l’Europe (Nigel Edit, 1963). Shah, A., The Reform of Intergovernmental Fiscal Relations in Developing and Emerging Market Economies (Washington, DC: World Bank, 1994). Solow, R., ‘Is Fiscal Policy Possible?’, in R. Solow (ed.), Structural Reform and Macroeconomic Policy (Basingstoke: Palgrave Macmillan, 2004). Spahn, P. B., ‘Fiscal Federalism: a Survey on the Literature’, in Jorgen Mortensen (ed.), Improving Economic and Social Cohesion in the European Community (New York: St. Martin’s Press, 1994), 145–54. Tiebout, C., ‘A Pure Theory of Local Expenditures’, Journal of Political Economy, 64 (1956): 416–24. Usher, D., The Uneasy Case for Equalization Payments (Vancouver: The Fraser Institute, 1995). Weingast, B., ‘The Economic Role of Political Institutions: Market-Preserving Federalism and Economic Development’, Journal of Law, Economics and Organisation, 11 (1995): 1–31. Wyplosz, C., ‘Fiscal Policy: Institutions versus Rules’, National Institute Economic Review, 191 (2005): 70–84.

2 Is There Compatibility between the Stability and Growth Pact and Automatic Fiscal Stabilizers? Sabrina Rostaing-Paris

1 Introduction The automatic stabilization issue has been permanently at the forefront of the scene, since Robert Solow invited the academic community to reconsider it in 2002. Looking at the criticisms raised against fiscal activism, rule-based fiscal policy relying on the working of automatic stabilizers provides several clear advantages. State-contingent tax revenues and expenditures dampen economic fluctuations practically with no information and implementation lags. Moreover, the impact lag of automatic stabilizers is generally considered to be relatively short. In principle, if automatic stabilizers are allowed to operate symmetrically over the cycle, they do not contribute to structural deterioration in budgetary positions. Automatic fiscal stabilizers are components of government budgets and they operate to smooth the business cycle. For example, in a recession fewer taxes are collected, which operates to support private incomes and damps the adverse movements in aggregate demand. Conversely, during a boom more taxes are collected, counteracting the expansion in aggregate demand. This stabilizing property is stronger if the tax system is more progressive. Another automatic fiscal stabilizer is the unemployment insurance system: in a downswing, the growing payment of unemployment benefits supports demand and vice versa in an upswing. The Stability and Growth Pact (SGP), which constitutes the cornerstone of the fiscal discipline for all the member states of the Economic and Monetary Union (EMU), goes in this direction by recommending to the countries to allow automatic fiscal stabilizers to play freely and to give up the use of pro-cyclical discretionary policies as soon as the budget is 25

26 Fiscal Policy in the European Union

balanced or in surplus. The fiscal rules laid down in the Maastricht Treaty and the Amsterdam Council ask each member state of the EMU to reach a budget balance equilibrium or surplus in the medium term, and not to run over a 3 per cent of GDP deficit. Various reasons can be found to justify such a fiscal discipline. The first concern is about sustainability of public finances, that would be hampered when a government has only a few chances of being re-elected and which imposes severe constraints on the composition of public expenditure for the future government (Alesina and Tabellini, 1990; European Commission, 1997). This might occur when politicians are opportunists and sometimes make budget deficits in order to increase their chances of being elected (Milesi-Ferretti, 1996), and when governments are generous throughout the cycle and not only during recession (Buti and van den Noord, 2004). The second concern is about the existence of negative externalities within a monetary union, because the growing debt of a member state would lead to a rise in the interest rate of the whole zone (Brück and Zwiener, 2004), increasing the weight of the debt of the other countries (Creel et al., 2002) and depressing their economic activity (Carton, 2005). The third concern is about the need for an effective policy-mix, between the fiscal policies of each country and between the fiscal policies of each member state and the common monetary policy. The fourth concern is about the fiscal preparation of demographic weight, with the ageing of the European population and its cost. The next concern is about the credibility and the independence of the European Central Bank, because in the event of a too important debt in a country, the ECB should respond to avoid inflationary pressures (Buti et al., 1998). Moreover, the fiscal theory of price level1 (FTPL) suggests that, as long as the discretionary fiscal policy ensures the solvency of the public debt, the monetary policy cannot have the last control on the price level (Buti et al., 2001; Buti and van den Noord, 2004). Therefore, in order to guarantee full independence of the central bank, a monetary union needs an institutional mechanism which imposes fiscal discipline. Finally, the last concern is about the free stabilization of the activity, which can be linked to the theory of fiscal cycles, developed in 1930 by the Swedish economists Myrdal and Ohlin. However, compliance with the rule prevails over that of stabilization. The use of the automatic stabilizers must then be compatible with the requirements of the Stability and Growth Pact. In order to reach such a sound fiscal position, each country has to implement structural reforms. Therefore, the fiscal consolidation should have an impact on the tax and unemployment benefits systems. Indeed, a fiscal

The SGP and Automatic Fiscal Stabilizers 27

consolidation should be done through a reduction of the tax burden and of social expenditures in order to guarantee a sustainable debt in the long term. From a theoretical point of view, there should be no trade-off between economic efficiency and stabilization. Fiscal reforms should improve both efficiency and stabilization of the economy. Indeed, several authors (Brunila et al., 2002; Buti et al., 2003; Martinez-Mongay and Sekkat, 2005) show that such a trade-off depends on the typology of the shocks hitting the economy. Fiscal stabilization is desirable in the case of demand shocks because it allows the smoothing of both output and inflation, while in the case of supply shocks, output stabilization may come at the cost of temporarily higher inflation (in the case of temporary supply shocks) or may delay structural adjustment (in the case of permanent supply shocks). Therefore, a reform of the tax system should not only improve efficiency but also, in the case of supply shocks, improve the cyclical stabilization. The reason for this is that the tax system can have distortive effects on economic activity (Gali, 1994). The studies that have been carried out on this subject have shown that an optimal tax burden exists (see, for example, Buti and van den Noord, 2003), i.e. a level of taxation below which its reduction can not only improve economic efficiency, but also make automatic stabilizers more efficient, depending on the nature of economic shocks. The empirical literature reveals that in the case of a balanced budget rule, a default of the stabilization of the business cycle could arise if the economy of a member state lazes or falls into recession before having reached the ‘close to balance or surplus’ position (Brück and Zwiener, 2004). Bartolini et al. (1995), using a structural macroeconometric model, showed that automatic stabilizers were restrained by restrictive budgetary policies aiming at fiscal consolidation. Therefore, automatic stabilizers could not smooth the business cycle. In a similar vein, Paul van den Noord (2000) pointed out that the cyclical behaviour of tax receipts could have changed during the 1990s because of the reforms of the tax systems. Hence, the automatic stabilization properties of tax systems should have been reduced. The simulations carried out with the INTERLINK model suggest that in the European Union, if automatic stabilizers had been able to work freely, i.e. without being counterbalanced by restrictive discretionary adjustments, the fiscal deficits of 1999 would have been on average six times higher than their observed level. As there is no consensus among economists that the compliance with the rules of fiscal discipline necessarily requires a reduction of automatic

28 Fiscal Policy in the European Union

stabilizers, we have sought to discover whether the current size of automatic stabilizers is compatible with the compliance to the rules of fiscal discipline imposed by the SGP. Most studies focus on the efficiency of the automatic stabilizers. They try to estimate the degree of automatic stabilization. This chapter tries to determine the trade-off between fiscal consolidation and the presence of automatic stabilizers in economies as mentioned above. For that purpose, we first estimate the size of automatic fiscal stabilizers for each member state of the Economic and Monetary Union, with the exception of Luxembourg and Slovenia, from 1992 to 2006. We focus on the budget sensibility to the business cycle. As previous works are based on macroeconometric models which are not available, we use a simple univariate regression, already applied by J. B. Taylor in 2000 with US data. To our knowledge, Favarque et al. (2005) are the only authors to have studied the size of automatic stabilizers in the European Union countries by using Taylor’s regression. Differently to their work, we do not use exactly the same variables. Our econometric method gives the total size of automatic stabilizers, without making a distinction between the categories of stabilizers. International institutions, such as the OECD2 and the European Commission,3 specify five elasticities having an impact on the budget balance, and by summing these elasticities, estimate total fiscal stabilizers for each country. Our study focuses on the total size of stabilizers and not on the evolution of each category. Once we have estimated the size of the automatic stabilizers, we are able to compare our results with a reference value that allows compliance with the rules of the SGP. We believe that our approach could provide a new analysis of the automatic stabilization issue in the framework of the Stability and Growth Pact. The rest of the chapter is organized as follows. Section 2 develops the effects of fiscal consolidation. Section 3 presents the data used. Section 4 explains the estimation method developed in two stages. Section 5 reveals the different results found and their interpretation. Section 6 concludes.

2 The effects of fiscal consolidation First, among the different Keynesian views, fiscal consolidation can have opposite effects on the economy. Thus, fiscal consolidation can only be expansionist for the expansionary fiscal contractions theory. And it can only be neutral within the neo-Ricardian framework.

The SGP and Automatic Fiscal Stabilizers 29

2.1 Contrary Keynesian effects We first study the standard Keynesian effects of a fiscal policy. Then we look at the role of the fiscal policy from the New Consensus point of view. The standard Keynesian argument is that fiscal contraction has a temporary restrictive effect through the channel of the aggregate demand, in a model with rigid prices and wages. In the simplest static model with fixed prices, an exogeneous reduction of public expenditure (or a contraction of the disposable income following an increase in taxes) will cause the traditional Keynesian effects on demand via the multiplier mechanism. In the short term, the Keynesian effects prevail and restrictive fiscal policies have contractionary effects on private consumption and economic activity. Indeed, the Keynesian multiplier rests on the demand–production–income interaction. The multiplier of public expenditure is equal to 1/(1 − c), c being the marginal propensity to consume. Thus, public expenditure causes an equal rise of production and income, which is spent in proportion c, whence there is a new rise of production and income, always spent in proportion c, etc. In the case of a monetary union, where intra-zone trade is strongly developed, the budget deficit of a country can have not only beneficial effects in the country which makes the deficit but also in the other countries of the zone through the trade channel. Thus, the other countries will benefit from the economic revival of the country making the deficit. By adding into the Keynesian analysis a proportional tax revenue according to the coefficient t, the escape in tax leads then to a reduced multiplier: 1/(1 − c + ct), and its magnitude is less than that of the multiplier of expenditure. The multiplier applicable to the tax is negative: a fall of tax thus increases revenue through the demand channel. Thus, in the case of a fiscal consolidation, a rise of tax and/or a fall of the public expenditure reduces the total income of the economy. On the other side, economists belonging to New Consensus Macroeconomics, for whom money supply has an endogenous nature, consider that monetary policy is the main instrument of macroeconomic policy while fiscal policy is not a powerful macroeconomic instrument any more (Arestis and Sawyer, 2003). The New Consensus view is characterized by the existence of a supply-side equilibrium (the non-accelerating inflation rate of unemployment, the NAIRU). When unemployment falls below the NAIRU, the rate of domestic inflation increases. The aim of monetary policy is price stability. In the long run, monetary policy affects the inflation rate but does not have any impact on output, while, in the short run, monetary policy can have an output stabilization

30 Fiscal Policy in the European Union

objective (Fontana and Palacio-Vera, 2005). However, in the long run, economic activity fluctuates around the supply-side equilibrium. Therefore, there is no trade-off between inflation and unemployment in the long run. In the case of a New Consensus model, fiscal policy is ineffective. The budget balance can vary along the business cycle thanks to the automatic fiscal stabilizers but it should be balanced on average over the cycle. Moreover, fiscal policy must not prevent monetary policy keeping the price level under control. Therefore, for the New Consensus ideology, a fiscal consolidation should not have any depressive effect on economic activity and allows monetary policy to work freely in order to fulfil the inflation targeting. 2.2 Successful fiscal consolidation We first present the theory of expansionary fiscal contractions. Then we expose the experiences of fiscal consolidation which had a positive impact on the economic activity. (a) The expansionary fiscal contractions theory Contrary to traditional Keynesian thought, the expansionary fiscal contractions theory regards as possible that fiscal contractions have expansionary effects and that expansionary fiscal policy has depressive ones. Thus, the expansionary (depressive) effects of the fiscal contractions (expansions) would result from the impact of the reductions (rise) of public expenditure. The expansionary fiscal contractions theory is reinforced by Danish and Irish experiences in the 1980s, as reported by Giavazzi and Pagano in 1990. Following their study, other expansionary fiscal contractions in OECD countries have been discussed in the literature and are summarized in Hemming et al. (2002). The majority of the literature considers OECD countries in the studies. Two periods of studies appear: from the first half of 1960 to the first half of 1990, and from 1970 to the first half of 1990. The conclusions provided are not homogeneous, but this is not surprising, as the definition of fiscal contractions is different from one paper to the next. However, other OECD countries besides Denmark and Ireland experienced expansionary fiscal contractions (Australia, Belgium and Canada, among others). According to this theory, a credible and permanent programme of reduction of government expenditure and taxes will stimulate the rise of private demand, thanks to expectations of a permanent debt reduction. The private expenditure will increase sufficiently to compensate for

The SGP and Automatic Fiscal Stabilizers 31

the direct effects of the fiscal contraction, and then, the most important impact of the reduction of the deficit can be positive rather than negative (Barry and Devereux, 2003). But in order to have an expansionary fiscal consolidation, the level of government expenditure or public debt must be very high and unsustainable (Blanchard, 1990; Perotti, 1999; Briotti, 2005). In such a context, if the fiscal authorities send the signal of an important and permanent modification of the budget balance, that will involve expections of durable debt fall (Giavazzi and Pagano, 1996), because, in the case of a high ratio of debt to GDP, it would seem that there is a stronger probability of having anti-Keynesian effects than standard ones (Afonso, 2006). Bertola and Drazen (1993) and Sutherland (1997) consider that when public expenditure is sufficiently high, an important fiscal adjustment is expected to be set up in the future through a reduction of the public expenditure, thus increasing agents’ permanent income and consumption. Alesina and Ardagna (1998) show that a fiscal consolidation can be expansionary via interest rates. In the case of high levels of debt, investors ask for a risk premium to hold long-term liabilities. Therefore, a successful fiscal consolidation would lead to a reduction in real interest rates, then to a reduction in risk premium. And the reduction in real interest rates has expansionary effects on economc activity. Lastly, Alesina et al. (2002) show that a fiscal consolidation will involve a rise in investment, via wages reduction. In a case of a lack of labour force following an increase in public employment, real wages go up while private profits decline and investment falls. Hence, if the fiscal consolidation takes place through a reduction in public employment, investment should rise and wages should be lower because of a higher unemployment rate. (b) Fiscal consolidation experiences According to Giavazzi and Pagano (1990), Denmark and Ireland experienced expansionary fiscal contractions in the 1980s. In the 1970s Denmark suffered from high budget deficits, strong inflation rates and a depreciation of its currency. Between 1980 and 1982, the national debt accelerated, rising from 29 per cent to 65 per cent of GDP, the unemployment rate being 4.2 percentage points higher than the rate in 1979. In the last quarter of 1982, the long-term interest rates had reached 22 per cent, and inflation 10 per cent. Therefore, a fiscal programme to recover a sound fiscal position was set up. In four years, the full-employment primary budget4 passed to 10 per cent of GDP, of which 2.8 per cent came

32 Fiscal Policy in the European Union

from a reduction in government consumption, 0.4 per cent from cuts in public investment, and the remainder from the raising of taxes net of transfers. The primary budget balance then improved by 15.4 per cent. On the monetary side, the Danish krone was anchored to the German mark, thus guaranteeing the credibility of its value. The fiscal contraction was accompanied by an average growth rate of 3.6 per cent of GDP between 1983 and 1986. This growth was the result of a fast increase in private consumption and investment. At the same time, complementary monetary and exchange rate policies were implemented. The German mark became a nominal anchor of the Danish krone, and the long-term interest rates fell sharply. On the other hand, in Ireland, a fiscal consolidation took place in 1982 and had a depressive effect on consumption and private investment, whereas the budget deficit of full employment had been reduced by more than seven percentage points of GDP in two years. Vis-à-vis this failure, the new government made a new attempt in 1987, but it favoured a reduction of public expenditure and investment rather than a rise of taxes as in 1982. In two years, the primary full-employment budget deficit was cut down by 7 per cent of GDP once again, the growth rate went up and the ratio of debt to GDP started to decline for the first time since the beginning of the 1970s. In that case, fiscal consolidation was reinforced by an exchange rate policy which stimulated domestic demand and led to a fall in interest rates. Giavazzi and Pagano (1990) thus showed that anti-Keynesian effects on private consumption and investment took place, but only in Denmark. However, in the case of Denmark and Ireland, the restrictive fiscal policy took place after a devaluation of the currency, which then involved gains of competitiveness with respect to the other countries, and the pegging of their currency to the German mark. The authors conclude that the important expected increase in the disposable income and wealth of private agents is the explanation for the anti-Keynesian behaviour of consumption. Giavazzi and Pagano’s later (1996) empirical study relates to a panel of nineteen OECD countries from 1970 to 1992. They study the variables of income, private consumption, investment, taxation and transfers, debt and public expenditure. The authors show that if the primary structural balance remained between more or less 5 per cent of potential GDP, the effects of a fiscal expansion/consolidation on private consumption and investment would conform to the Keynesian analysis. On the other hand, if the variable exceeded this 5 per cent of GDP value, then fiscal expansion/consolidation would have anti-Keynesian effects. This

The SGP and Automatic Fiscal Stabilizers 33

negative correlation would be more obvious in the case of important consolidations than in that of important fiscal expansions. The authors conclude that the episodes of fiscal consolidation not only concerned Denmark and Ireland, but many other countries (Australia, Finland, Japan and Sweden) could also have made such an experiment. Thus, such anti-Keynesian behaviours could apply to various countries. However, to test these behaviours econometrically, Giavazzi and Pagano increase taxes and reduce public expenditure at the same time. They thus limit the role of automatic fiscal stabilizers through the reduction of public expenditure. The rise of taxes, through tax base or tax rates, will make it possible to avoid more strongly an overheating in the high phase of the cycle or in the event of expansionary shock on the economy but do not leave sufficient room for manoeuvre to taxpayers in a downswing or in the event of negative shock on the economy. In an article of 2000, Giavazzi et al. estimate national equations of savings rates for 18 industrialized countries, from 1970 to 1996, and 101 developing countries, from 1970 to 1994. In the case of the OECD countries, the authors distinguish effects of fiscal policy according to whether one is in a normal period, in contraction, or in expansion. In particular, in normal situations, a rise of 1 per cent of net taxes will involve an increase of 0.44 per cent in the private savings rate; in a period of expansion of 0.22 per cent; and in a period of contraction of 0.03 per cent. In the same way, the impact of a rise of 1 per cent of public expenditure involves a fall of private savings rate of 0.9 per cent in a normal period; a fall of 0.71 per cent in a period of expansion; and a fall of 0.5 per cent at the time of a consolidation. These results reveal the existence of Ricardian behaviours in the situation of fiscal contraction, whereas it is not the case during the two other periods. In this article, the authors indicate that the majority of the fiscal consolidations were carried out through a rise of net taxes rather than through a fall of public expenditure. Van Aarle and Garretsen (2003) take up the study by Giavazzi and Pagano (1996), but adapt it to the European countries over the transition period towards monetary union from 1990 to 1998, and then over a second period from 1970 to 2000. The authors were not able to show the existence of non-linear effects between taxes and transfers on one side and private consumption on the other during the phase of transition towards EMU. On the other hand, they found a short-term non-linear effect between public expenditure and consumption. But for Hjelm (2002) and Schclarek (2005) none of the preceding results is valid. Indeed, while working on panel data, they find that positive or negative fiscal shocks involve standard Keynesian effects in nineteen

34 Fiscal Policy in the European Union

OECD countries. Thus, the effects of fiscal contractions in Denmark and Ireland would not be transposable to the other countries. The success of such a consolidation would be due to specific factors. The expansionary effects of fiscal contractions in Denmark and Ireland would have been generated by the depreciation of the currency preceding the consolidation which would have led the confidence of the consumers on their future income. 2.3 The Ricardian equivalence hypothesis On the basis of the initial intuition of Ricardo, Barro (1974) contributed to refute the thesis of the effectiveness of fiscal action on the economic activity variations. The Ricardian equivalence hypothesis implies the same effect of an increase in taxes and an increase in debt to finance public expenditure. The economy is therefore influenced only by the quantity of government expenditure and not by the method of financing this expenditure. This theory proposes that a fiscal expansion generates the expectation of future fiscal contractions, in order to compensate initial looseness. Thus, public expenditure will be fully balanced by a more important private saving, resulting from households and firms taking precautionary measures. Consequently, the fall in the private aggregate demand will reduce, or bring back even to zero, the expected expansionary multiplier effects. In other words, in a case of perfect Ricardian equivalence, a tax cut, leading to an increasing public debt, will be fully offset by higher private saving. Indeed, economic agents will anticipate higher taxes in the future to refund the debt. Therefore, their permanent income and aggregate demand are not affected (Hemming et al., 2002).5 This is even more the case when the government has adopted a fiscal rule. Thus, if the fiscal expansion does not have the discounted effects, one can suppose that a fiscal contraction will not have any effects either. However, in order for the Ricardian equivalence assumption to hold, a certain number of conditions must be satisfied, like perfect capital markets, absence of constraints of liquidity, and altruistic agents able to form rational expectations. The robustness of the Ricardian equivalence can thus be called into question when these conditions are not fulfilled, and remains very much a matter for discussion among the scientific community.

3 Data and descriptive statistics The empirical analysis realized in this study is based on annual data of the primary cyclical budget balance and of the output gap, both of them

The SGP and Automatic Fiscal Stabilizers 35

in percentage of potential GDP. While the output gaps come from the database of the OECD,6 the primary cyclical budget balances arise from our calculation of the result of the difference between the primary total balances and primary cyclically adjusted balances of each country. The budget balances and the primary structural balances are also from the OECD database. The study covers the period 1992 to 2006, concerning each member state of the Economic and Monetary Union, with the exception of Luxembourg and Slovenia, i.e. eleven countries. In order to realize the estimates, we chose to use the primary (total, structural and cyclical) balances instead of (total, structural and cyclical) budget balances because ‘the primary budget balance reflects, more than the budget balance, the will of the Government and the Parliament’;7 and ‘because the paid interests depend, on the one hand, on the level of debt and, on the other hand, on the level of the long-term interest rates, two variables which, in the short run, escape government’s control’.8 Indeed, part of the debt burden is derived from the monetary policy, and is not affected by the budget policy. In Figures 2.1(a)–(k) we can see that, for each country, the progress of the two variables, expressed as percentage of potential GDP, is very similar. So, the primary cyclical balance fluctuates according to the business cycle. As the cyclical balance is the difference between the budget balance and the structural balance, and as the changes in the structural balance usually represent discretionary adjustments decided by the government, the changes in the cyclical balance are the result of the action of fiscal stabilizers which answer to the cycle in an automatic way. Therefore, these figures present a positive correlation between the primary cyclical balance and the output gap between 1992 and 2006, even though the magnitude of the output gap is higher than that of the primary cyclical balance. This correlation can also be observed in the scatterplot (Figure 2.2). The higher the output gap, the more the primary cyclical balance improves. Moreover, by comparing each country, we see very close fluctuations in the primary cyclical balances and in the output gaps, with the exception of Finland, Greece and Ireland. Indeed, by focusing on the primary cyclical balance, on average, this variable first decreased from 1992 to around 1997, then increased until 2000 and lastly has slumped since. As a result, the primary cyclical balance is worse in 2006 than it was in 1992. But the fall was different across countries, from 2.9 per cent of potential GDP in Portugal to 0.4 per cent of potential GDP in Italy. The increase

1992

2006

(h) Italy

(k) Spain

⫺4

⫺5

(i) The Netherlands

(f) Greece

⫺4

⫺4

Figure 2.1(a)–(k)

2006

⫺12

Primary cyclical balance Output gap 1999

0

2006

⫺0.5

1999

⫺4.5

1992

4

2006

3

1999

3

1992

(c) Finland

2006

⫺3

1999

⫺3

1992

⫺2

2006

⫺1

1999

0.5

1992

⫺0.5

2006

1

1999

2

1992

4

2006

(e) Germany

2

1992

(b) Belgium

1999

⫺4 1992

⫺7 2006

⫺3 1999

⫺2

1992

0

2006

⫺1

1999

⫺1

1992

1

2006

(j) Portugal 4

1999

(g) Ireland 5

1999

(d) France 1

1992

In % of potential GDP

In % of potential GDP

In % of potential GDP

36

(a) Austria 4

Primary cyclical balance and output gap in European countries from 1992 to 2006

The SGP and Automatic Fiscal Stabilizers 37

Primary cyclical balance (in % of potential GDP)

1 0.5

Greece

Ireland

Finland

0 Italy Belgium Germany France ⫺1 The Netherlands Austria

Spain

⫺0.5

⫺1.5 ⫺2 ⫺2.5 ⫺4

Portugal ⫺3

0 ⫺2 ⫺1 1 Output gap (in % of potential GDP)

2

Figure 2.2 Primary cyclical balance vs. output gap

of the cyclical balance in Finland, Greece and Ireland was respectively of 5.2, 0.5 and 0.9 per cent of potential GDP.

4 Two-step estimation procedure In order to achieve our goal, we need to first estimate the size of the automatic stabilizers and then to compare these results with the rules of the Stability and Growth Pact. 4.1 Estimates of automatic fiscal stabilizers Taylor (2000) proposed a rule of fiscal policy derived from his rule of monetary policy of 1993. He described the budget balance according to the output gap and the structural budget balance: BB = f .OG + SBB

(1)

where f is a constant, OG is the output gap (variation of the real GDP from the potential GDP, expressed as a percentage of potential GDP), and where budget balance (BB) and structural budget balance (SBB) are expressed as a percentage of GDP. The first term of the right-hand side of equation (1), f.OG, represents the cyclical budget balance. According to Taylor, this equation represents a standard breakdown of the budget balance into two parts: a cyclical and a structural part. It allows also making the distinction between automatic stabilizers and discretionary policy. Indeed, the evolutions of the

38 Fiscal Policy in the European Union

structural balance show the effects of the discretionary measures taken by the governments, while the f.OG term represents the impact of automatic stabilizers on the budget balance. Thus, thanks to the estimates of this rule, Taylor could show how automatic fiscal stabilizers worked in the United States during the period 1960–99. For that purpose, he made a regression, with each component of the budget (budget balance, cyclical balance and structural balance) as dependent variables and the output gap as explanatory variable. Other studies have already broken down the budget balance into a cyclical balance and a structural balance, in particular the European Commission (2001) and Favarque et al. (2005). Favarque et al. (2005) applied the fiscal rule of Taylor to the countries of the EMU. They did not take into account the primary balances and observed rigorously the rule suggested by Taylor. Their results for the recent period are similar to the results of other studies on the subject that used a different methodology (van den Noord, 2000). The main advantage of Taylor’s rule is that it is easy to use, while macroeconometric models are kept confidential by the various institutions. In addition, as Favarque et al. (2005) pointed out, the application of this rule is particularly simple, comprehensible and transparent. We thus apply the method of Taylor (2000) to the member states of the Economic and Monetary Union, with the exception of Luxembourg and Slovenia. We have the following relation: PCBt = γ0 + γ1 OGt + εt

(2)

with PCBt the primary cyclical balance in the year t, γ0 as an intercept, OGt as the output gap9 at time t, and εt a disturbance. We start by studying the order of integration of the series. The unit root test of Philips-Perron concludes with the stationary of the series. Using time-series data, we apply the Ordinary Least Squares method (OLS) in order to estimate the coefficients of equation (2). An ARCH test was set up to test the presence of heteroskedasticity in the model, which shows the absence of heteroskedasticity. If and when necessary, the estimates were corrected by an autocorrelation of errors by the method of AR(1). The autocorrelation of errors is detected using a Breush-Godfrey test. 4.2 Application of Taylor’s fiscal rule to the Stability and Growth Pact and its discipline In essence, Taylor’s fiscal rule is just an identity which breaks down the budget balance (BB) into two components: the cyclical component

The SGP and Automatic Fiscal Stabilizers 39

( f.OG) and the structural balance (SBB). For the structural balance and an output gap given, this identity enables us to check whether automatic fiscal stabilizers (the f coefficient) have been compatible with the rules of the Stability and Growth Pact. Specifically, we shall be able to say whether automatic stabilizers prevented the budget balance from reaching equilibrium or surplus on the one hand and, on the other hand, avoided a budget deficit equal to or higher than 3 per cent of GDP. Lastly, we shall look at how automatic stabilizers should work for a structural balanced budget within the framework of a reformed Pact. By using the data of the structural balances and the output gaps for each country, we can compare the coefficient f estimated previously with what it should have been in order to have been compliant with the rules of the Pact. Using the breakdown of the budget balance, we shall see which conditions should make automatic stabilizers compatible with the fiscal rules. (a) Balanced or surplus budget rule The breakdown of the budget balance is: BB = f .OG + SBB

(1)

The rule of a balanced or surplus budget can be expressed as: BB ≥ 0

(3)

By combining equations (3) and (1), we get: f .OG + SBB ≥ 0 ⇔ f .OG ≥ −SBB

(4)

SBB if OG > 0 OG SBB if OG < 0 or f ≤ − OG ⇔f ≥−

Therefore, when the output gap is positive, the estimated size of auto  matic stabilizers must be higher than the reference value − SBB , whereas OG in the case of a negative output gap, the estimated f must be lower than the reference value. The compatibility of automatic fiscal stabilizers with the rule of the SGP thus depends on the sign of the output gap.

40 Fiscal Policy in the European Union

(b) The rule of budget deficit lower than 3 per cent of GDP The starting point is again the breakdown of the budget balance (equation (1)). The rule of a budget deficit smaller than 3 per cent of GDP can be expressed as: BB > −3

(5)

By combining equations (5) and (1), we get: f .OG + SBB > −3 ⇔ f .OG > −3 − SBB

(6)

3 + SBB if OG > 0 OG 3 + SBB if OG < 0 or f < − OG ⇔f >−

As previously, when the output gap is positive, the estimated size of   automatic stabilizers must be higher than the reference value − 3+SBB , OG whereas in the case of a negative output gap, the estimated f must be lower than the reference value. (c) Target of a structural budget balance Within the framework of a reformed Pact, a target of a balanced structural budget could be adopted. Such a rule would allow the budget balance to vary according to the economic activity thanks to the action of automatic stabilizers. Again following Taylor’s fiscal rule, we start from the breakdown of the budget balance (equation (1)). We add to the rule of a balanced structural budget the limit of the budget deficit which does not exceed 3 per cent of GDP. This can be expressed as follows: 

SBB = 0 BB > −3

(7)

By combining equations (7) and (1), we get: f .OG > −3 3 if OG > 0 OG 3 if OG < 0 or f < − OG

⇔f >−

(8)

The SGP and Automatic Fiscal Stabilizers 41

When the output gap is positive, the estimated size of automatic stabiliz 3  ers must be higher than the reference value − OG , whereas in the case of a negative output gap, the estimated f must be lower than the reference value. The compatibility of automatic fiscal stabilizers with the modified rule of the Stability and Growth Pact depends once again on the sign of the output gap. Table 2.1 summarizes the different conditions required to comply with the various rules of the SGP. Table 2.1 Summary table of compliance conditions When output gap is:

Rule of BB ≥ 0 needs:

Rule of BB > −3 per cent of GDP needs:

SBB OG SBB f ≤− OG

f >−

f ≥−

Positive Negative

3 + SBB OG 3 + SBB f <− OG

Rules of  SBB ≥ 0 BB > −3% of GDP need: 3 OG 3 f <− OG f >−

5 Estimation results This section gives first the results of the size of automatic stabilizers and then their comparison with the fiscal rules of the SGP. 5.1 The size of automatic stabilizers Table 2.2 gives the results of automatic fiscal stabilizers for each country. The quality of the estimations is good, as the percentage of the ajusted 2

variance (R ) is highly above 50 per cent. Moreover, the results of Fisher’s test (F-stat) consistently confirm the good quality of the estimations. Therefore, the explicative power of the model is suitable. As expected, the coefficient of the cyclical component is significant with the threshold of 1 per cent throughout all the countries. It is positive in all the countries, indicating that when the output gap decreases by 1 per cent, i.e. when the economic activity slows down, the cyclical budget balance grows weaker from 0.33 per cent in Greece to 0.71 per cent in Germany, thus showing a counter-cyclical budgetary action. But the sensitivity of the cyclical balance to the business cycle is very different from one country to another. Therefore, we can suppose that it should be easier for countries with small automatic stabilizers, i.e. Greece, Spain, Ireland and Portugal, to comply with the rule of balanced budget,

42

Table 2.2

Automatic fiscal stabilizers 1992–2006(a,b) Austria

Belgium

Finland

France

Germany

Greece

Ireland

Italy

The Netherlands

Portugal

Spain

Intercept

−.22* (−1.80)

−.22*** (−2.72)

−.19*** (−2.89)

−.3* (−1.79)

.42** (2.36)

−.11* (−2.02)

.10 (1.06)

.14 (.2)

−.07 (−.22)

−.24 (−.89)

−.10 (−.66)

OG

.42*** (6.96)

.49*** (8.54)

.52*** (42.62)

.42*** (5.56)

.71*** (6.27)

.31*** (12.61)

.37*** (14.55)

.53*** (7.81)

.47*** (4.06)

.40*** (6.35)

.34*** (7.49)

AR(1)

.55** (2.07)

–

–

.64*** (2.91)

–

–

.56** (2.15)

–

.67** (2.43)

.71** (2.33)

.65*** (2.99)

R2

.92

.85

.99

.90

.75

.92

.99

.82

.90

.95

.94

.91

.84

.99

.88

.73

.92

.99

.81

.89

.94

.93

F-stat (Prob)

70.2 (.000)

73.02 (.000)

1816.16 (.000)

52.14 (.000)

39.36 (.000)

158.94 (.000)

511.2 (.000)

60.99 (.000)

57.2 (.000)

106.76 (.000)

88.57 (.000)

Breush-Godfrey (Prob)

–

2.09 (.148)

.44 (.508)

–

1.79 (.181)

1.84 (.174)

–

.41 (.522)

–

–

–

R

2

(a) Ordinary Least Square method. (b) Dependent variable: primary cyclical balance. Note: Standard errors in parentheses. Asterisks indicate statistical significance at the 5 (**) and 1 (***) per cent level.

The SGP and Automatic Fiscal Stabilizers 43

while the budget balance should be not very sensitive to the business cycle in these countries. 5.2 Comparisons between estimated f and the reference values10 (a) The case of a balanced or surplus budget rule Over the period of the study, automatic fiscal stabilizers were not compatible with the rule of the net budgetary balanced position in many countries: in Germany, Austria, France, Greece, Italy and Portugal. However, all the countries of the EMU had in place, at different times, some automatic stabilizers, that were incompatible with the rule (see Table 2.3). Table 2.3 Periods of compatibility between automatic fiscal stabilizers and a balanced or surplus budget in EMU countries since 1992 Countries

Compatibility of automatic stabilizers with a balanced or surplus budget

Number of periods of compatibility

Austria Belgium

Incompatible over the whole period Compatible since 2000

0 7

Finland France Germany Greece Ireland

Compatible since 1998 Incompatible over the whole period Incompatible over the whole period Incompatible over the whole period Compatible from 1996 to 2001 and from 2003 to 2006 Incompatible over the whole period Compatible in 1999 and 2000 Incompatible over the whole period Compatible since 2003

9 0 0 0 10

Italy The Netherlands Portugal Spain Total

32

0 2 0 4

Balanced or surplus budget since 1992 Never 2000–2003, 2005–2006 1998–2006 Never 2000 Never 1996–2001, 2003–2006 Never 1999-2000 Never 2003, 2005–2006 31

Generally, in the periods of economic growth, when the output gaps were positive, stabilizers were not powerful enough to improve the budget balance clearly, in other words, stabilizers always had a size smaller than that which would have been necessary for the budget to be at least balanced. On the other hand, during the time of economic deceleration, characterized by negative output gaps, stabilizers were more important than they should have been for the budget to be balanced. The results we find are compatible with other studies which indicate that stabilizers have an asymmetrical role along the cycle, as they are more important at the time of the descending phases of the cycle than at the time of the expansion. Thus, automatic stabilizers prevent the

44 Fiscal Policy in the European Union

budget balances from improving sufficiently when they should, but let them become strongly degraded in order to stimulate demand in the downturn cycle. Over the period under survey, the countries for which stabilizers are incompatible with the rule of balanced budget never managed to reach a balance, with the exception of Germany which obtained a budget surplus of 1.3 per cent of GDP in the year 2000. For the other countries, the periods of incompatibility between stabilizers and the rule correspond to the periods of budget deficit. If we take as example the particular cases of Finland and the Netherlands, we notice that: • in Finland automatic stabilizers have been compatible with the rule

of the Pact only since 1998, and the Finnish budget balance has been in surplus since 1998; • in the Netherlands, automatic stabilizers were not compatible with the rule solely in 1999 and 2000, and the budget was in surplus in these two same years only. We can thus conclude that generally, there is no compatibility between automatic fiscal stabilizers and a balanced budget within the EMU, as Finland and Ireland are the two countries with the most compatibility periods. Therefore, automatic stabilizers were compatible with the rule of fiscal discipline over 32 fiscal years only, the budget was in balance only during 31 periods, whereas the study covers 165 periods. Finally, over the period, automatic fiscal stabilizers were compatible with the rule of fiscal discipline only in 19 per cent of the cases. However, during difficult times, the member states of the EMU are authorized to let their budget balance drift towards a deficit of 3 per cent of GDP. We shall now look at the compatibility between automatic fiscal stabilizers and a rule of budget deficit smaller than 3 per cent of GDP. (b) The case of a rule of budget deficit lower than 3 per cent of GDP By focusing on the examples of Germany and France, Figures 2.3(a)–(b) show that in Germany, automatic fiscal stabilizers were compatible with the rule of budget deficit lower than 3 per cent of the GDP in 1992, in 1994, and from 1997 to 2001. They would have been too hefty in 1993, in 1995–1996, and since 2002 to allow Germany to remain within the limit of authorized public deficit. In France, automatic stabilizers were compatible with the rule of budget deficit from 1997 to 2001 and have been since 2005. They would have been insufficient in 1992, then too hefty from 1993 to 1996 and from 2002 to 2004.

The SGP and Automatic Fiscal Stabilizers 45

In % of potential GDP and in nominal value

(a) Germany 5 2.5 0 ⫺2.5 ⫺5 1992

1996

2000

2004

(b) France In % of potential GDP and in nominal value

5 3 1 ⫺1 ⫺3 1992

1996 Estimated f

2000 Reference value

2004 og

Figure 2.3(a)–(b) Comparison between the size of automatic stabilizers and the reference value with respect of the sign of the output gap in Germany and France since 1992

If we compare our results with the previous study, automatic fiscal stabilizers seem rather more compatible with the rule of budget deficit lower than 3 per cent of GDP (Table 2.4). Ireland is an exception, in respect to the other countries, with stabilizers incompatible only in 1992 and compatible since then. Comparing the episodes of budget deficit lower or equal to 3 per cent of GDP with the episodes of compatibility of automatic stabilizers with the deficit rule, we notice that they generally tend to match each other. However, in several countries (Spain, Finland, France, Greece, Italy, the Netherlands and Portugal) the budget deficit sometimes reached or exceeded the limit of 3 per cent of GDP whereas automatic stabilizers were compatible that time with the rule of deficit (1996 in Finland, 1997 in France, 1999 in Greece, 1998 in Italy, 1994 in The Netherlands, 1997–8

46 Fiscal Policy in the European Union Table 2.4 Periods of compatibility between automatic fiscal stabilizers and a budget deficit lower than 3 per cent of GDP Countries

Compatibility of automatic fiscal stabilizers with a budget deficit lower than 3 per cent of GDP

Austria Belgium Finland France Germany Greece Ireland Italy The Netherlands Portugal Spain Total

Number of periods of compatibility

Budget deficit smaller than 3 per cent of GDP

Compatible in 1992 and since 1997 Compatible since 1997 Compatible since 1996 Compatible from 1997 to 2001 and since 2005 Compatible in 1992, 1994, and from 1997 to 2001 Compatible in 1999 and in 2006 Compatible since 1993 Compatible from 1997 to 2000

11

1992, 1997–2006

10 11 7

Compatible in 1994, from 1996 to 2002 and since 2004 Compatible from 1997 to 1999 Compatible since 1997

11

1997–2006 1997–2006 1998–2001, 2005–2006 1992, 1994, 1997–2001, 2006 2006 Since 1992 1997, 1999–2000 1993, 1996–2002, 2004–2006 1999, 2002 1997,1999–2006

90

7 2 14 4

3 10

86

in Portugal and in Spain). The deficit could then have been increased not only by automatic stabilizers but also by a counter-cyclical discretionary policy. Thus, automatic fiscal stabilizers were compatible with a deficit lower than 3 per cent of GDP in 54 per cent of the cases, that is to say nearly three times more often than with a balanced or in surplus budget rule. However, this result should not be surprising as it appears easier to comply with a rule of limited budget deficit than with a balanced budget rule. Therefore, every country, except Ireland, has at some point exceeded, or still exceeds now, the limit of budget deficit imposed by the Stability and Growth Pact. Although automatic stabilizers were compatible with such a rule of budgetary discipline over more than half of the period, it remains the case that in 46 per cent of cases they did not work out in respect of a budget deficit not exceeding 3 per cent of GDP. (c) The case of the rules of balanced or surplus structural budget and of budget deficit lower than 3 per cent of GDP In the case of such a rule, in most countries of the EMU, when the output gap has been negative (positive), the size of automatic fiscal stabilizers

The SGP and Automatic Fiscal Stabilizers 47 Table 2.5 Periods of compatibility bewteen automatic fiscal stabilizers and a balanced structural budget combined with a budget deficit limit in EMU countries since 1992 Countries

Compatibility of automatic fiscal stabilizers with a balanced structural budget and a budget deficit lower than 3 per cent of GDP

Number of periods of compatibility

Austria Belgium

Compatible over the whole period Compatible over the whole period

15 15

Finland

10

France Germany Greece Ireland

Compatible since 1997, incompatible before Compatible over the whole period Compatible over the whole period Compatible over the whole period Compatible over the whole period

Italy The Netherlands Portugal Spain

Compatible over the whole period Compatible over the whole period Compatible over the whole period Compatible over the whole period

15 15 15 15

Total

160

15 15 15 15

Balanced or surplus structural budget

2004 1998–1999, 2001–2006 1996–2006 Never Never Never 1994, 1996–2000, 2004–2006 Never 2005–2006 Never 2003–2006 34

has been lower (higher) than the reference value. Therefore, the automatic fiscal stabilizers have been largely compatible with a rule of a balanced structural budget and of a budget deficit lower than 3 per cent of GDP (Table 2.5). However, in Finland, from 1992 to 1996, f was higher than the reference value whereas the output gap was negative. But since 1997, when the output gap has been positive (negative), f has been higher (smaller) than the reference value. Finally, in the case of Finland, the automatic fiscal stabilizers have been compatible with the fiscal rule only since 1997. Thus, within the framework of the modified Pact, which lets the budget deficit reach 3 per cent of GDP and which imposes a balanced structural budget, automatic fiscal stabilizers become compatible with the objectives. But surprisingly, the compatibility between automatic stabilizers and the rule of budget discipline has not gone hand in hand with a balanced or surplus structural budget. From Table 2.5, we can see that there have been 160 periods of compatibility but only 34 periods of balanced or surplus structural budget, with some countries (France, Germany, Greece, Italy and Portugal) never having reached a balanced structural budget. Therefore, as seen previously, if the structural budget has not

48 Fiscal Policy in the European Union

been in balance during the period while the automatic stabilizers have had the required size to be compatible with the rule, this is because of discretionary policies introduced by the governments.

6 Conclusion While previous studies on the subject have tended to show how the rules of the Stability and Growth Pact could prevent the work of automatic stabilizers, this chapter has addressed a somewhat different issue. Ceteris paribus, progressive tax systems lead to a lower budget deficit in good times, while the deficit increases in recessions. Moreover, large and progressive tax systems usually go hand in hand with more generous systems of social protection. Although social benefit programmes mainly have an equity role, some of them also act as automatic stabilizers. Unemployment benefits constitute the clearest example. Tax systems and unemployment benefits work together to smooth the business cycle, as a counter-cyclical response to shocks hitting the economy. Once the Maastricht Treaty was approved by the future member states of the EMU, they had to make fiscal reforms, among other things, in order to enforce EMU criteria. The aim of these reforms was to reach sound public finances in each country in order to participate at stage three of monetary union on the one hand, and to maintain this position of sustainable budget balance. Therefore, a trade-off could have arisen between the size of automatic fiscal stabilizers and the compliance with the rules of the Stability and Growth Pact. Once the estimates of the size of automatic stabilizers for each country over the period 1992–2006 were realized, we compared these estimates to a value which would allow compliance with the rules of the SGP. Our results show that automatic stabilizers are not compatible with a rule of a balanced budget in 81 per cent of the cases, while they are compatible with a rule of a budget deficit lower than 3 per cent of GDP in 54 per cent of the cases. Most of the time, we can see that, in the periods of economic growth, when the output gaps were positive, stabilizers were not powerful enough to improve the budget balance clearly. On the other hand, during the times of economic deceleration, characterized by negative output gaps, stabilizers were more important than they should have been for the budget to be balanced. Finally, in a framework of a reformed Pact, automatic stabilizers are compatible with a rule of a balanced structural budget combined with a deficit budget lower than 3 per cent of GDP in 97 per cent of the cases.

The SGP and Automatic Fiscal Stabilizers 49

These results show that a trade-off between automatic fiscal stabilizers and the compliance of the rules of the Pact does exist. Many economists have criticized the Pact and suggested reforming it, even at the time of its adoption, at the Amsterdam Council, in 1997. The main criticisms have focused on the deficit budget threshold, which is considered as too rigid, and the target of balanced or surplus budget, which should lead to a null debt in the long term. One of the reforms proposed would be a target of a cyclically adjusted budget balance, which would allow the budget balance to vary according to the business cycles, therefore, budget deficits would be possible and the budget balance could play again its role of automatic stabilizer. This chapter endorses this view, showing that fiscal automatic stabilizers can exist without preventing sustainability of the public finances. Nevertheless, we must remain careful when interpreting the results. First, we used the primary cyclically adjusted balance to get the primary cyclical budget for each country, and the output gap to perform the regressions. However, structural balance and output gap are variables which do not exist in reality. Therefore, these estimations may be criticized. Second, we have only 15 observations for each country, which constitutes a small sample, whereas usually the OLS method requires bigger ones. Finally, this study has focused on the size of automatic stabilizers, but we now still have to know whether these automatic stabilizers are efficient, or, in other words, whether they help to smooth the business cycle.

Notes 1. The roots of the FTPL can be found in Barro (1974), Sargent and Wallace (1981), Leeper (1991), and Canzoneri and Diba (1996), among others. 2. The methodology of the OECD is described in van den Noord, P. (2000) and in Girouard and André (2005). 3. The methodology of the European Commission is presented in Bouthevillain and Garcia (2000). 4. The full-employment budget is based on receipts and expenditures that would prevail under conditions of full employment. This approach has been extensively used in the United States, but other countries (the Netherlands, West Germany) have also adopted similar measures as an aid to policy-making. 5. When Ricardian equivalence holds, the fiscal multiplier is zero in the case of lump-sum taxes. But with proportional or progressive taxes, if the increasing public spending is expected to be financed by higher future taxes, permanent income and consumption will decline. In this last case, fiscal multipliers even become negative (Hemming et al., 2002). 6. OECD Economic Outlook 80 database (OECD, 2006).

50 Fiscal Policy in the European Union 7. 8. 9. 10.

Bénassy-Quéré and Blanchard (2005: 143). Creel and Sterdyniak (1995). Where OG = Y−Y∗ , with Y the real GDP and Y* the potential GDP. Y∗ Graphs making it possible to check for each country the compatibility between automatic fiscal stabilizers and the various rules of fiscal discipline are available from the author.

References Afonso, A., ‘Expansionary Fiscal Consolidations in Europe: New Evidence’, European Central Bank, Working Paper Series, 675 (2006). Alesina, A. and S. Ardagna, ‘Tales of Fiscal Adjustment’, Economic Policy, 27 (1998): 487–545. Alesina, A., S. Ardagna, R. Perotti and F. Schiantarelli, ‘Fiscal Policy, Profits, and Investment’, American Economic Review, 92(3) (2002): 571–89. Alesina, A. and G. Tabellini, ‘A Positive Theory of Fiscal Deficits and Government Debt’, Review of Economic Studies, 57 (1990): 403–14. Arestis, P. and M. Sawyer, ‘Inflation Targeting: a Critical Appraisal’, The Levy Economics Institute, Working Paper, 388 (2003). Barro, R. J., ‘Are Government Bonds Net Wealth?’, Journal of Political Economy, 82 (1974): 1095–1117. Barry, F. and M. B. Devereux, ‘Expansionary Fiscal Contraction: a Theoretical Exploration’, Journal of Macroeconomics, 25(1) (2003): 1–23. Bartolini, L., A. Razin and S. Symansky, ‘G-7 Fiscal Restructuring in the 1990s: Macroeconomic Effects’, Economic Policy (1995). Bénassy-Quéré, A. and O. Blanchard, Politique économique (De Boeck, 2005), 143. Bertola, G. and A. Drazen, ‘Trigger Points and Budget Cuts: Explaining the Effects of Fiscal Austerity’, American Economic Review, 83(1) (1993): 11–26. Blanchard, O., ‘Can Severe Fiscal Contractions be Expansionary? Tales of Two Small European Countries: Comments’, in O. J. Blanchard and S. Fischer (eds), NBER Macroeconomics Annual (Cambridge, MA and London: MIT Press, 1990), 111–16. Bouthevillain, C. and S. Garcia, ‘Limites des méthodes d’évaluation du concept de déficit structurel’, Revue française d’économie, 15(1) (2000): 75–121. Briotti, M. G., ‘Economic Reactions to Public Finance Consolidation: a Survey of the Literature’, BCE Occasional Paper Series, 38 (2005). Brück, T. and R. Zwiener, ‘Fiscal Policy Rules for Stabilization and Growth: a Simulation Analysis of Deficit and Expenditure Targets in a Monetary Union’, German Institute for Economic Research, Discussion Paper, 427 (2004). Brunila, A., M. Buti and J. in’t Veld, ‘Fiscal Policy in Europe: How Effective are Automatic Stabilizers?’, European Commission, Economic Paper, 117 (2002). Buti, M., D. Franco and H. Ongena, ‘Fiscal Discipline and Flexibility in EMU: the Implementation of the Stability and Growth Pact’, Oxford Review of Economic Policy, 14(3) (1998): 81–97. Buti, M., C. Martinez-Mongay, K. Sekkat and P. van den Noord, ‘Automatic Fiscal Stabilizers in EMU: a Conflict between Efficiency and Stabilization?’, CESifo Economics Studies, 49(1) (2003): 123–40.

The SGP and Automatic Fiscal Stabilizers 51 Buti, M., W. Roeger and J. in’t Veld, ‘Stabilizing Output and Inflation: Policy Conflicts and Co-ordination under a Stability Pact’, Journal of Common Market Studies, 39(5) (2001): 801–28. Buti, M. and P. van den Noord, ‘What is the Impact of Tax and Welfare Reforms on Fiscal Stabilizers? A Simple Model and an Application to EMU’, European Commission, Economic Paper, 187 (2003). Buti, M. and P. van den Noord, ‘Fiscal Policy in EMU: Rules, Discretion and Political Incentives’, European Commission, Economic Paper, 206 (2004). Canzoneri, M. B. and B. Diba, ‘Fiscal Constraints on Central Bank Independence and Price Stability, CEPR Working Paper, 463 (1996). Carton, B., ‘Les externalités budgétaires dans la zone euro’, Economie et Prévision, 169 (2005): 297–302. Creel, J. and H. Sterdyniak, ‘Les déficits publics en Europe: causes, conséquences ou remèdes à la crise?’, Revue de l’OFCE, 54 (1995). Creel, J., T. Latreille and J. Le Cacheux, ‘Le Pacte de stabilité et les politiques budgétaires dans l’Union européenne’, Revue de l’OFCE, Hors série (2002): 245–97. European Commission, ‘Economic Policy in EMU’, Economic Papers 124 and 125 (1997), 1–43; 47–66; 94–104. European Commission, ‘Public Finances in EMU – 2001’, European Economy, 3 (2001). Farvaque, E., F. Huart and C. Vaneecloo, ‘Stabilization et transparence: La règle budgétaire de Taylor et le Pacte de stabilité’, Revue de l’OFCE, 92 (2005): 99–144. Fontana, G. and A. Palacio-Vera, ‘Are Long-Run Price Stability and Short-Run Output Stabilization All That Monetary Policy Can Aim For?’, Metroeconomica, 58(2) (2005): 269–98. Gali, J., ‘Government Size and Macroeconomic Stability’, European Economic Review, 38(1) (1994): 117–32. Giavazzi, F., T. Jappelli and M. Pagano, ‘Searching for Non-linear Effects of Fiscal Policy: Evidence from Industrial and Developing Countries’, European Economic Review, 44(7) (2000): 1259–89. Giavazzi, F. and M. Pagano, ‘Can Severe Fiscal Contractions be Expansionary? Tales of Two Small European Countries’, in O. J. Blanchard and S. Fischer (eds), NBER Macroeconomics Annual (Cambridge, MA and London: MIT Press, 1990), 75–111. Giavazzi, F. and M. Pagano, ‘Non-Keynesian Effects of Fiscal Policy Changes: International Evidence and the Swedish Experience’, NBER Working Paper 5332 (1996). Girouard, N. and C. André, ‘Measuring Cyclically-adjusted Budget Balances for OECD Countries’, OECD Economic Department, Working Paper 434 (2005). Hemming, R., M. Kell and S. Mahfouz, ‘The Effectiveness of Fiscal Policy in Stimulating Economic Activity: a Review of the Literature’, IMF Working Paper, wp/02/208 (2002). Hjelm, G., ‘Is Private Consumption Growth Higher (Lower) During Periods of Fiscal Contractions (Expansions)?’, Journal of Macroeconomics, 24(1) (2002): 17–39. Leeper, E., ‘Equilibria under “Active” and “Passive” Monetary Policies’, Journal of Monetary Economics, 27(1) (1991): 129–47.

52 Fiscal Policy in the European Union Martinez-Mongay, C. and K. Sekkat, ‘Progressive Taxation, Macroeconomic Stabilization and Efficiency in Europe’, European Commission, Economic Paper 233 (2005). Milesi-Ferretti, G. M., ‘Fiscal Rules and the Budget Process’, IMF Working Paper, wp/96/60 (1996). OECD, OECD Economic Outlook 80 (Paris: OECD, 2006). Perotti, R., ‘Fiscal Policy in Good Times and Bad’, Quarterly Journal of Economics, 114 (1999): 1399–1436. Sargent, T. J. and N. Wallace, ‘Some Unpleasant Monetarist Arithmetic’, Federal Reserve Bank of Minneapolis Quarterly Review, 5 (1981): 1–17. Schclarek, A., ‘Fiscal Policy and Private Consumption in Industrial and Developing Countries’, Department of Economics, Lund University (2005). Solow, R., ‘Peut-on recourir à la politique budgétaire? Est-ce souhaitable?’, OFCE Observation et diagnostics économiques, conférence présidentielle prononcée au XIIIème Congrès mondial de l’Association internationale des sciences économiques, September (2002), Lisbon. Sutherland, A., ‘Fiscal Crises and Aggregate Demand: Can High Public Debt Reverse the Effects of Fiscal Policy?’, Journal of Public Economics, 65(2) (1997): 147-62. Taylor, J. B., ‘Reassessing Discretionary Fiscal Policy’, Journal of Economic Perspectives, 14(3) (2000): 21–36. van Aarle, B. and H. Garretsen, ‘Keynesian, Non-Keynesian or No Effects of Fiscal Policy Changes? The EMU Case’, Journal of Macroeconomics, 25 (2003): 213–40. van den Noord, P., ‘The Size and Role of Automatic Fiscal Stabilizers in the 1990s and Beyond’, OECD Economics Department, Working Paper 230 (2000).

3 Stabilization Capacity and Fiscal Policy in the EMU Jorge Uxó González and M. Jesús Arroyo Fernández

1 Introduction Over the last two decades a ‘New Consensus’ has emerged within mainstream macroeconomics (Arestis, 2007), and one of its main features is the prominence of monetary policy over fiscal policy as a tool of stabilization policy. Bean (2007) states that interest rates are a flexible tool that can be changed instantaneously, while fiscal policy is seen as a less effective weapon because increases in government spending take time to initiate, and its effectiveness can be limited by Ricardian equivalence effects or the difficulty of reversing a fiscal expansion. In addition, the theoretical downgrading of fiscal policy has been reflected in practice by the introduction of legal rules limiting the use of expansionary fiscal measures, such as the Stability and Growth Pact (SGP) in the European Union. However, the vast majority of empirical evidence does not support the rejection of the use of fiscal policy.1 From a theoretical perspective, Blinder (2004) summarizes the most frequent criticisms of Ricardian equivalence, and Calmfors (2003: 5) states that ‘the Ricardian Equivalence results require very restrictive theoretical assumptions which are nor likely to apply in reality’. Arestis and Sawyer (2004: 21), finally, highlight that ‘fiscal policy remains a powerful instrument for regulating the level of aggregate demand’. The European and Monetary Union (EMU) is a good example of the relevance of this debate. The transfer of the monetary policy to the Eurosystem appears to reinforce the stabilizing role to be played by national fiscal policies, because the interest rates established by the European Central Bank (ECB) cannot be simultaneously adjusted to the cyclical conditions of all the euro area economies. Then, fiscal 53

54 Fiscal Policy in the European Union

policy should play a more active role in the European Monetary Union, with enough flexibility to tackle idiosyncratic shocks. Calmfors (2002) supports the same idea: ‘The general problem, as I see it, is that macroeconomic stabilization in response to asymmetric developments in the euro area will require an increased use of fiscal policy as a stabilization policy instrument at the national level.’ But the Stability and Growth Pact basically restricts this function to automatic stabilizers,2 with little room for the adoption of other discretionary anti-cyclical policies. Our main objective in this chapter is to study the problem of the possible loss of stabilization capacity by EMU economies and to consider whether a more active fiscal policy is necessary and convenient to solve this problem. More specifically, we present a proposal for the application of fiscal policy by national governments defined as an activist fiscal policy rule. First, we will review the concept of stabilization capacity and the framework in which monetary and fiscal policy is applied in EMU. Then, we will see that experience with the monetary policy applied to date in the EMU has already shown how different cyclical situations arise in different countries – Spain and Germany, for example – all of which cannot be solved at the same time by the ECB. This discussion will allow us to consider the stabilizing and destabilizing mechanisms which come into play in a monetary union when a shock takes place (specifically, the real interest rate and the real exchange rate). Finally, we will propose the application of a more active fiscal policy in EMU. This will be done using some fiscal policy rules and simulating their operation in a simple model of the European Monetary Union.

2 Short-run stabilization and the economic policy framework in EMU The creation of the monetary union has given rise to the problem of the possible loss of stabilization capacity of national economies, because of the redefinition of the principal instruments associated with this task (monetary and fiscal policy). The monetary policy framework mainly establishes that the ECB has the stability of prices as almost its only objective3 and it applies one monetary policy for all the EMU countries trying to stabilize the mean values of inflation, without taking into account explicitly the inflation differentials between member countries.4 This implies that the nominal interest rate might not be appropriate for some countries and, as a

Stabilization and Fiscal Policy in the EMU 55

consequence, the real interest rate could even have a destabilizing effect in those countries with a different rate of inflation than the EMU mean. In these circumstances, fiscal policy could compensate these effects if the member countries could set it discretionally in order to stabilize the economy in the short run. However, this is not totally possible in the new European fiscal policy framework, in which governments’ capacity to apply short-run stabilization policies is limited by the Stability and Growth Pact. It is a well-known fact that this mainly establishes that EMU countries should adjust their structural balances to close to equilibrium, and except for specific exceptions,5 the nominal deficit can be no greater than 3 per cent of the GDP.6 In this policy framework, the contribution of fiscal policy to the cyclical stabilization of national economies is confined almost exclusively to the use of automatic stabilizers, while discretionary fiscal policy, although not ruled out, is actually limited to certain situations.7 However, the reform of the Stability and Growth Pact approved in March 2005 introduced more flexibility for the application of short-term budgetary policies by national governments (European Council, 2005). Each state, for example, will present its own long-term budgetary target and an adjustment plan aimed at achieving it; even more importantly, it has accepted a broader definition of the concept of severe economic recession and other exceptional factors which justify the appearance of unsanctioned excess deficits (cyclical economic situation, sustainability of public debt, or the application of structural reforms related to the Lisbon Agenda, for instance). In addition to the SGP, some European countries have introduced national fiscal rules to favour even more the respect for long-term budgetary stability. The number of these national fiscal rules has increased over the past twenty years and they have tended to cover central and regional governments.8 The European Council, in fact, stated in March 2005 that ‘national budgetary rules should be complementary to the Member States’ commitments under the Stability and Growth Pact’. In order to analyse how this monetary and fiscal policy can reduce the ‘stabilization capacity’ in EMU, we must define it. We propose that the authorities are trying to achieve two economic policy objectives (the stabilization of the rate of inflation around a target and of the GDP around its potential9 ) and, to achieve these objectives, the authorities use the monetary and fiscal policies. In this context, the stabilization capacity is the ability of an economy to return to the equilibrium values of the inflation rate and the income level after a shock, as fast as possible and with the fewest deviations of these variables from their long-term

56 Fiscal Policy in the European Union

values. Then, this stabilization capacity decreases if, after a transitory shock, the economy does not return to its equilibrium or it takes longer to do so. With this definition of stabilization capacity in mind, the European monetary and fiscal framework is especially relevant when the countries present cyclical divergences because of specific shocks. If all these economies experience the same shock (a global deceleration, for instance), the single monetary policy could react to foster a rapid return to equilibrium. The problem arises, however, when the shock is asymmetric: by only affecting some economies, there would be no reaction by the monetary authority (the function of which is to monitor the mean economic situation) and the countries in question would take longer to return to normal, generating costs in terms of employment and income. The empirical studies conducted to test the likelihood of European economies experiencing specific or asymmetric shocks show that such situations cannot be ruled out.10 These characteristics of the monetary and fiscal policy framework are also important when the EMU countries present structural differences, for example different equilibrium real interest rates. In this situation, the same monetary policy cannot be correct for all countries because it has different effects in each of them. Looking at the different cyclical situations among the EMU countries since 1999, we can see in the following figures that this problem is more than a mere theoretical possibility. Although the inflation differentials between the member states decreased in the years prior to the establishment of the monetary union, this trend ceased in 1999 and there has even been a new increase in inflation dispersion11 (Figure 3.1). Moreover, in contrast with other monetary areas, another distinctive feature of euro area inflation differentials is their persistence (ECB, 2005). The differences in inflation rates have been highly persistent, with some countries registering above-average inflation rates since 1999, and others constantly below the mean (Figure 3.2). Specifically, since 1999 Ireland, Greece, Spain, Portugal and Italy have always had inflation rates above the mean, whereas the inflation rates in Germany and Austria were below the mean every year. As the ECB (2003) states correctly, diverging inflation rates might just be a sign of regional economies adjusting towards equilibrium. That is, if countries have entered the monetary union at an over- or undervalued exchange rate, below- or above-average inflation rates over a number of years might be required just to bring a region’s real exchange rate to equilibrium. But divergences can also be a sign of a different position in

57

1.4 1.2 1 0.8 0.6 0.4 0.2

2004m06 2004m11 2005m04 2005m09 2006m02 2006m07 2006m12

2001m12 2002m05 2002m10 2003m03 2003m08 2004m01

1999m01 1999m06 1999m11 2000m04 2000m09 2001m02 2001m07

0

Figure 3.1 Inflation dispersion in EMU Source: Eurostat Database.

4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 GE

FI

AU

FR

BE EMU IT

Figure 3.2 Average inflation rate (1999–2006) Source: Eurostat Database.

HO PO SP GR

IR

58 Fiscal Policy in the European Union

7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 IT

GE PO AU EMU BE

FR HO

FI

SP GR

IR

Figure 3.3 GDP rate of growth (average, 1999–2006) Source: Eurostat Database and the authors.

the business cycle. Because of this we also analyse the growth rates of EMU member countries that have registered significantly different values in the period 1999–2006.12 Figure 3.3 shows how some countries, such as Spain, Greece and Ireland again, have had higher growth rates, while others have presented smaller than average growth rates, such as Germany, Portugal and Italy. These differences remain when we take into account the potential rate of growth of each country (Figure 3.4). Finally, there is not even a declining trend in output gap dispersion between member countries13 (such as Spain and Germany) and EMU (Figure 3.5). Concluding, the inflation and the growth rates show clear evidence of the possibility that, in the European Monetary Union, there are countries with inflation and growth rates above mean persistently (such as Spain) and others, like Germany, under the mean. This reflects the importance of using the fiscal policy as a stabilization policy in the short term.14 In other words, what this evidence shows is that Germany would have needed a more expansive demand policy, while Spain would have needed a more restrictive demand policy, and this cannot be achieved only with a single monetary policy.

59

2 UEM

GE

SP

1.5 1 0.5 0 ⫺0.5 ⫺1 ⫺1.5 ⫺2

1999

2000

2001

2002

2003

2004

2005

2006

Figure 3.4 Differences between GDP and trend (rates of growth 1999–2006) Source: Eurostat Database and the authors.

3 2 1 0 ⫺1 ⫺2 ⫺3 ⫺4

IT

GE

PO

AU EMU BE

FR

HO

FI

SP

GR

IR

Figure 3.5 Differences between GDP and trend (rates of growth, sum 1999– 2006) Source: Eurostat Database and the authors.

60 Fiscal Policy in the European Union

3 Destabilizing and stabilizing mechanisms in a monetary union There is evidence that differences in local inflation and cycles are not unusual in large currency areas, and in the case of the European Monetary Union it is necessary to pay special attention to them because of the stabilizing and destabilizing mechanisms which are operating in the euro zone.15 On the one hand, the real interest rate, as a result of the nominal interest rate set by the ECB and the country’s inflation rate, generates a destabilizing effect on the national economy with a different inflation rate than the mean. On the other hand, the real exchange rate can have a stabilizing effect when, for example, the excess inflation in a member country generates a real appreciation, leading the deflationary tendencies that tend to reduce the initial inflation gap in the medium term. Next, we try to have some measure of the relevance of these destabilizing and stabilizing mechanisms to some specific countries. To get an initial idea of the destabilizing mechanism derived from the single monetary policy16 we analyse first how the nominal interest rate set by the European Central Bank has been adjusted to the cyclical situation of Spain and Germany since 1999, and how it is reflected in the real interest rate. Then, we will review the real exchange rate mechanism. Our analysis is based on the Taylor rule (Taylor, 1993), since it appears to appropriately represent the effective use of the interest rate by major central banks to stabilize the economy. According to this rule, the central bank establishes a reference, or neutral, interest rate (for which the economy achieves its potential production) and positions the intervention rate higher (lower) when inflation is greater (lower) than the target or the output gap is positive (negative). Formally, this simple rule can be represented as follows for the ECB: ˙ t − P˙ T ) + β2 OGUt iUt = r + P˙ T + (1 + β1 )(PU

(1)

˙ and OGU are the mean inflawhere r is the real neutral interest rate, PU tion rate and output gap values in the EMU, and P˙ T is the target inflation rate, established by the ECB as close to or below 2 per cent. Parameters β1 and β2 indicate the importance of the monetary authorities’ reaction to deviations in inflation and the output from the reference values. From Taylor (1993) estimations, it is common to value parameter β1 at 0.5 and parameter β2 at 0.5, so the equation can now be rewritten as: ˙ t − 2%) + 0.5OGUt iUt = r + 2% + 1.5(PU

(2)

Stabilization and Fiscal Policy in the EMU 61

In a paper on monetary policy rules (ECB, 2001), the ECB says that a monetary policy based on rules has important advantages over greater discretionarity by central banks,17 but it also argues that these types of simple rules also involve some important weaknesses.18 Therefore, to avoid these problems while retaining the advantages of systematic behaviour, the ECB has preferred to represent its monetary policy as a ‘strategy’ based on two pillars, with performance more ‘governed’ by rules rather than strictly ‘linked’ to such rules. Nonetheless, the ECB (2004: 57) describes the performance of a central bank in a similar way to the Taylor rule: ‘A central bank pursuing a stability-oriented policy world then set its interest rate instrument so as to move real short-term interest rates to that level below or above their natural level that is necessary to counter the effects of these shocks to price developments.’ And there is much empirical literature according to which the actual performance of the leading central banks follows this type of simple rule.19 Therefore, it continues to be useful to analyse the monetary policy actually applied by the ECB using this tool. If we accept this methodology, we find that the single monetary policy applied by the ECB could be inappropriate for a country belonging to the monetary union for one of these three reasons: 1. The ECB could be setting an interest rate different than the optimal rate determined by the Taylor rule. 2. The Taylor rule’s parameters could have different values in one country than in the whole monetary union. For example, this could be the case of the neutral interest rate or of the parameters β1 and β2 . These two parameters are derived both from the preferences of the authorities and from some structural circumstances determining the efficacy of the monetary policy transmission mechanism.20 3. The optimal interest rate could be different for a country because its inflation and output gap are deviating from the monetary union average. Since the purpose of our analysis is mainly to study how the ECB has adapted the evolution of the intervention rate to short-term changes in the cyclical situation of the EMU economies, we can avoid the problems of estimating the long-term equilibrium rate and concentrate on the ‘cyclic component’ of the monetary policy rule (that is, in the third of the above possibilities). Following Doménech et al. (2002), this is the name we give to the weighted sum of the deviation of inflation from the target, plus

62 Fiscal Policy in the European Union

the output gap, using the coefficients proposed by Taylor himself. For the monetary union: ˙ t − 2%) + 0.5OGUt CYCLIC COMPONENT = 1.5(PU

(3)

Ultimately, this component measures how much the ECB intervention rate must deviate from its long-term equilibrium value as a consequence, in turn, of the distance between output and inflation from their longterm rates (the potential GDP and a 2 per cent inflation rate). If the evolution of the interest rate effectively adapts to the changes in this short-term ‘cyclical component’, it can be considered that the ECB does apply a stabilizing monetary policy, at least from the point of view of the whole monetary union. But this expression refers to the mean inflation and output gap in the EMU and the problem we are considering here is precisely that the development of these variables may have differed in each country in the euro zone, leading to a bad adjustment between the optimal interest rate for the EMU and the optimal rate for each of these countries. We refer to this poor adjustment as the Interest Rate Misalignment (IRM),21 calculated by the sum of the deviations in inflation and in output gap from the monetary union average, which is defined for country i as follows: ˙ t − P˙ i ) + 0.5(OGUt − OGi ) IRMti = 1.5(PU t t

(4)

In other words, this index measures the difference between the optimal interest rate for the EMU and a specific country, resulting from the different performance of the cyclical component, and assuming the same monetary policy rule and the same long-term equilibrium interest rate for the entire euro zone.22 A positive value of the index means that the interest rate applied by the ECB as the result of the mean cyclical situation in the EMU is too restrictive for the current cyclical conditions in country i, and the opposite would be true if the index is negative. Looking at the case of Germany, Figure 3.6 shows that the index has been positive since 1999 because of the negative evolution of Germany’s output gap and its lower inflation rates. Spain, in contrast, has a negative index, mainly because of its positive inflation differential from the mean and its higher rate of growth. This reveals a bad adjustment between the entire union’s monetary policy and the policy that would have been appropriate for these countries according to their own circumstances, and it therefore leads us to add the use of active fiscal policy for anticyclical purposes.

Stabilization and Fiscal Policy in the EMU 63 Output gap

Inflation rate

2.00

4.50

1.50

4.00

1.00

3.50

0.50

3.00 2.50

0.00

2.00

⫺0.50

EMU

Spain

06-I 06-III

04-I 04-III 05-I 05-III

02-I 02-III 03-I 03-III

99-I 99-III

06-I

06-III

05-I

05-III

04-I

04-III

03-I

03-III

02-I

02-III

01-I

01-III

00-I

00-III

0.00 99-I

0.50

⫺2.00 99-III

1.00

⫺1.50

00-I 00-III 01-I 01-III

1.50

⫺1.00

Germany

Nominal interest rates misalignment

3 2 1 0 ⫺1 ⫺2 ⫺3

Spain

06-I

06-III

05-III

05-I

04-III

04-I

03-I

03-III

02-III

02-I

01-I

01-III

00-III

00-I

99-III

99-I

⫺4

Germany

Figure 3.6 Different cyclical position and single monetary policy in EMU: Spain and Germany Source: Eurostat Database and the authors.

Indeed, the problem lies not only in the fact that the nominal interest rate established by the ECB may not be appropriate for all the countries if they are in different cyclical positions, but that, as a result, the real interest rate may have a destabilizing effect. We can see this by supposing that a country (Germany) suffers a demand shock and its output gap is below the European average. The ECB reduces the nominal interest rate but only to stabilize the euro zone GDP and inflation. As a consequence, the German real interest rate would reduce by less and its GDP will be lower. The IMF (2004: 108) considers this question, concluding that a change in a country’s output gap does not give rise to a change in its real interest rate in the desired direction, but tends to cause greater disequilibria: ‘The effect [of the national output gap] on real interest rates has tended to be procyclical, with rising inflation in booming economies, especially Ireland, the Netherlands and Portugal, leading to lower real interest

64 Fiscal Policy in the European Union

Real interest rates 4.00 3.00 2.00 1.00 0.00 ⫺1.00

EMU

Spain

06-III

06-I

05-III

05-I

04-III

04-I

03-III

03-I

02-III

02-I

01-III

01-I

00-III

00-I

99-III

99-I

⫺2.00

Germany

Figure 3.7 Real interest rates Source: Eurostat Database.

rates, stimulating domestic demand even further. Similarly, in countries experiencing protracted downturns, such as Germany, falling inflation tended to result in relatively high real rates.’ In Figure 3.7 we represent the real interest rate in the EMU, Spain and Germany, confirming this destabilizing effect: the real interest rate has fallen less in Germany than in Spain, although the first has a slower inflation and growth rate and it would have needed a stronger monetary impulse. Of course, the loss of stabilization capacity by EMU economies would be lower if monetary union economies had alternative adjustment mechanisms other than monetary policy, among which traditional Optimal Monetary Area theory focuses on flexible salaries and geographic mobility. Empirical studies, however, also show that such mechanisms are weak in European economies (Mongelli, 2002) and it would be difficult to increase their efficacy in the short term.23 Another possible stabilizing mechanism is based on the effect that inflation differentials within the EMU have on the real exchange rate. If a country experiences a specific shock, and its inflation rate is no longer the same as that of the union (but higher, for instance) the resulting variation in the real exchange rate will act as a stabilization mechanism (in this case, a real appreciation would decrease exports to the rest of the

Stabilization and Fiscal Policy in the EMU 65

130 120 110 100 90 80 70 60 1999

2000

2001

Spain/EMU

2002

2003

Spain/Germany

2004

2005

2006

Germany/EMU

Figure 3.8 Real exchange rates ULCE (1999 = 100) Source: European Commission (2007).

EMU, reduce income and, ultimately, moderate inflationist tension). If it is strong enough, this mechanism will compensate for the lack of an independent monetary policy. For example, as a consequence of the lower German rate of inflation that has taken place between 1999 and 2006, its competitiveness has increased by 12.4 per cent vis-à-vis the euro zone and by 23.6 per cent vis-à-vis Spain. In contrast, Spain has experienced a real appreciation of an 8.3 per cent relative to EMU24 (Figure 3.8). However, this mechanism may not be strong enough to ensure the stabilization of national economies subject to asymmetric shocks, or to do so quickly enough.25 For example, the IMF (2004: 109) says that ‘the effect of inflation differentials on external competitiveness should be stabilizing, since higher inflation leads to real appreciation (and vice versa). In practice, however, the stabilizing effect of inflation differentials through this channel does not seem to have been particularly strong.’ Concluding, this section shows that given the lack of cyclical synchronization among the European economies and relatively weak adjustment mechanisms, the common monetary policy should be accompanied by an alternative mechanism enabling national authorities to recover this stabilization capacity, and the natural candidate is fiscal policy. In the

66 Fiscal Policy in the European Union

rest of this chapter we consider the possibility of a more active fiscal policy performing this function.

4 A model of the monetary union including fiscal policy In this section we present a simplified model of a monetary union formed by two countries which can experience specific supply and demand shocks. Our main objective is to use the model to analyse whether the introduction of some appropriate ‘activist’ fiscal policy rule could enhance the stabilization capacity of national economies in the EMU, in comparison with the Stability and Growth Pact framework. We use different rules to systematize the behaviour of monetary and fiscal authorities. This expression is used here not in the sense of legal restrictions. Instead of this, each rule is a simple equation determining the value of the economic policy instrument: interest rate and budgetary balance. They are contingent or activist rules because the instrument will change according to the evolution of some indicators of the state of the economy. Fiscal rules have been less developed in the literature26 than monetary rules, but their use enables us to compare the effects on the economy of alternative fiscal authorities’ behaviour, or discuss the policy mix using the same method for analysing monetary and fiscal policy. The model has three blocks of equations, one for each country (represented by numbers 1 and 2) and the third for the monetary union (represented by the letter U). All the parameters are positive or zero and, to simplify, we assume that their value is identical in both countries. This means that there are no structural differences between them, enabling us to isolate the effect of specific shocks in one country on the evolution of macroeconomic variables, our primary interest. For the same reason, we assume that the fiscal policy applied by the two governments follows the same rule, with identical coefficients, although they differ insofar as each of them targets the evolution of its own national variables. Both economies also have the same size. COUNTRY 1: ˙ t = P1 ˙ t−1 + aOG1t−1 + z1t P1 ˙ t − d BB1t + g1t OG1t = −b(r1t − r ) − c RE ˙ t r1t = iUt − P1 ˙ t − P˙ T ) BB1t = BB1T + (α1 + α2 )OG1t + α3 (P1

(1.1) (1.2) (1.3) (1.4)

Stabilization and Fiscal Policy in the EMU 67

COUNTRY 2: ˙ t = P2 ˙ t−1 + aOG2t−1 + z2t P2

(2.1)

˙ t − d BB2t + g2t OG2t = −b(r2t − r) + c RE

(2.2)

˙ t r2t = iUt − P2

(2.3)

˙ t −P ) BB2t = BB2 + (α1 + α2 )OG2t + α3 (P2 T

˙T

(2.4)

MONETARY UNION: ˙ t = 0.5P1 ˙ t + 0.5P2 ˙ t PU OGUt = 0.5OG1t + 0.5OG2t ˙ t + β1 (PU ˙ t − P˙ T ) + β2 OGUt iUt = r + PU

(U.1) (U.2) (U.3)

The first equation for each country (1.1 and 2.1) shows the inflation ˙ which remains constant only when the output gap (OG) is zero rate (P), and there are no supply shocks. z represents these possible inflationist shocks originating on the supply side of the economy,27 and they could be specific for each country. The effect of the output gap on inflation occurs with a delay of one period, meaning that the possible variations in aggregate demand first lead to income variations and only later to changes in inflation. Parameter a represents this effect. Equations (1.2) and (2.2) represent two IS lines. The level of aggregate demand, and therefore the output gap, depends on three different effects and a random variable: 1. First, on the difference between the real interest rate (r) and the real equilibrium interest rate (r).28 The parameter b measures this real interest rate effect. 2. Second, on the real exchange rate effect (RE, defined as the relative prices of country 1 to country 2) derived from possible inflation differentials. In this case the relevant parameter is c. 3. Finally, we have the fiscal policy effect, or the variations of the output gap derived from the different values of the budgetary balance (BB). If this surplus increases, the aggregate demand falls. In this case, the influence of the fiscal policy on aggregate demand is measured by the parameter d. We suppose than d = 0, so fiscal policy can be used to influence aggregate demand.

68 Fiscal Policy in the European Union

4. Of course, as in the first equations, demand shocks can also occur, represented by the random variable g, and they can be specific for each country. The third pair of equations (1.3 and 2.3) show that the nominal interest rate is now set by the ECB for the entire union (iU) and the national real interest rate is determined by the difference between this single interest rate and each economy’s inflation rate. Therefore, the same monetary policy can give rise to different real interest rates in each of the two economies. The last equations of the national blocks (1.4 and 2.4) represent a general fiscal policy rule, in which the budgetary balance can depend on the output gap and on the inflation rate. More specifically, we have formulated two alternative fiscal policy rules in our model: • The first one represents a strict application of the Stability and Growth

Pact (SGPFR) by the European governments. In this rule the government defines a cyclically adjusted budgetary balance target (BBT ) and deviates from it exclusively by the action of automatic stabilizers, represented by α1 . The value of the BBT would ‘anchor’ the fiscal policy at a budgetary balance value which is sustainable in the long term. As one of the fiscal objectives set in the SGP is to reach ‘equilibrium or surplus’ in the medium term, we suppose that BBT = 0 for both countries. Of course, the nominal deficit has to be less than 3 per cent of GDP every year. This fiscal policy framework can be represented by the following equations (for example, for country 1): BBt = BBT + α1 OGt BB1T = 0 BB1t ≥ −3% • The second alternative is a More Active Fiscal Rule (MAFR). As in the

previous case, this rule also has a long-term target for the budgetary balance, but it introduces four main changes: 1. We consider that national fiscal authorities display an active fiscal policy in order to stabilize their economies. So, their fiscal policy applies to the role of automatic stabilizers but also to the discretionary behaviour by the authorities. This ‘discretional reaction’ is represented by α2 .

Stabilization and Fiscal Policy in the EMU 69

2. Governments modify the budgetary balance according not only to the output gap (as before) but also to the inflation rate. This means that the fiscal policy rule adopts the same form as the Taylor rule for monetary policy. In our opinion, there are good reasons for believing that this may be reasonable fiscal authority behaviour. In Spain, for instance, a restrictive fiscal policy has been applied, and at least in some part this behaviour is justified by the higher Spanish inflation.29 For the same reasons, Germany should probably have applied a more expansive fiscal policy during the economic downturn. 3. In this fiscal policy framework, the long-term budgetary target could even be different to zero and different between both countries. Actually, they should be different if each country also has different equilibrium real interest rates. In this case, the single monetary policy would be much too expansive to one country and much too restrictive for the other, so they should use their fiscal policies to compensate this negative effect even in the long term.30 4. Finally, the 3 per cent ceiling is nonsense, because the long-term target is enough to guarantee the sustainability of public finances. When the national government applies this fiscal policy rule, it takes the same form that the Taylor rule does for the monetary policy, displaying the same BBT rule as the equilibrium real interest rate.31 In Table 3.1 we show the parameters’ values of the general fiscal policy rule corresponding to each of these possible scenarios. Finally, the model is completed with a third block representing the monetary union. In this case, both inflation and the output gap are the mean of the two countries, and the ECB applies a standard Taylor rule. Of course, the relevant values to the ECB are the average for the euro zone. Table 3.1

Alternative fiscal policy rules

˙ t − P˙ T ) General Fiscal Policy Rule: BB1t = BB1T + (α1 + α2 )OG1t + α3 (P1 Parameters

SGPFR

MAFR

α1 α2 α3 BBT BB

>0 0 0 0 ≥−3%

>0 ≥0 >0 It could be different from 0 No limits

70 Fiscal Policy in the European Union

5 A restrictive demand shock in one country under different fiscal policies Once we have briefly presented the model, we can use it to compare the dynamic evolution of the national economies belonging to the EMU when one of them suffers a specific demand shock. We will suppose that national fiscal authorities follow the same policy rule. We start with an initial equilibrium in which the two economies are at their potential income levels and the inflation rate is 2 per cent. Then, we assume that country 1 suffers a negative demand shock and the national income falls. Although the monetary policy and the fiscal policy take an expansive stance (the ECB reduces the nominal interest rate and the national government increases the deficit) the output gap becomes negative. Because of this, in the following period the inflation rate will be less than 2 per cent. Then, in order to return to its initial equilibrium, the economy needs a period in which the output gap remains positive. But once the GDP departs from its potential, its dynamic evolution depends, apart from fiscal policy, on the relative importance of the two mechanisms we have described before: • As a consequence of the falling inflation in country 1, the average

inflation will also fall in the whole monetary union, but less than proportionally. So, although the ECB will reduce the nominal interest rate, it will do it by less than the reduction in the national inflation rate, because it sets the monetary policy on the average EMU values. So, the real interest rate will rise in country 1 instead of fall, and it will be destabilizing. • Country 1 will gain competitiveness vis-à-vis the other EMU countries as a consequence of its lower inflation. This could compensate, at least partially, the real interest rate effect, improving the net exports of the country and increasing the aggregate demand. The final outcome will depend on the degree to which the real appreciation compensates for the contractive effect of a rise in the real interest rate. But this condition may not always be met and, if there is no fiscal policy, the economies will depart from equilibrium without any tendency to return to it: in country 1, the national income will not grow but fall even more, and there will be further inflation deceleration in the following periods. Can an active fiscal policy solve this problem? This depends on the concrete rule applied by the government, as shown in Figure 3.9.32

Stabilization and Fiscal Policy in the EMU 71 0.4

2.2 2.0

0.0

1.8 ⫺0.4

1.6 1.4

⫺0.8

1.2

⫺1.2

1.0 ⫺1.6

0.8

⫺2.0

0.6 2

4

6

8

10 12 14 16 18 20 22 24

Inflation SGPFR

2

4

6

8

10 12 14 16 18 20 22 24

Output gap SGPFR

Inflation MAFR

3.2

0.0

2.8

⫺0.2

2.4

⫺0.4

2.0

⫺0.6

1.6

⫺0.8

Output gap MAFR

⫺1.0

1.2 2

4

6

8

10 12 14 16 18 20 22 24

Real interest rate SGPFR

Real interest rate MAFR

2

4

6

8

10 12 14 16 18 20 22 24

Budgetary balance SGPFR

Budgetary balance MAFR

Figure 3.9 A restrictive demand shock when the competitiveness effect is weak (country 1)

In Figure 3.933 we show the evolution of the inflation rate, the output gap, the real interest rate and the budgetary balance in the two alternative scenarios when the real exchange rate effect is weak. As we can see, the inflation rate and the output gap will be ever decreasing in the case of the SGP. In contrast, the fiscal policy will guarantee the return to the initial equilibrium with the MAFR. With both rules, there is fiscal deficit in the second period (t = 5) but the deficit is higher with the MAFR because of the low level of the inflation rate (α3 > 0). Because of this, the activist fiscal policy neutralizes the negative effect of the real interest rate and maintains the GDP above potential until the inflation rate reaches the 2 per cent target. However, if the fiscal policy depends only on the automatic stabilizers, when the GDP recovers (t = 5) the budgetary deficit becomes less negative (precisely because α1 > 0 and α3 = 0) and this fiscal policy is not able to generate a positive OG. So, in the third period the inflation rate falls again, going away from the 2 per cent target. It is important to note that

72 Fiscal Policy in the European Union Table 3.2

Stabilizing and destabilizing mechanism and model parameters

Mechanism

Efffect

Real interest rate Real exchange rate Fiscal policy

Destabilizing Stabilizing Stabilizing

IS parameters b c d

Economic policy rules parameters β1 , β2 α1 , α2 , α3

this result (the only fiscal policy rule able to stabilize the economy is the MAFR, not the SGPFR) does not depend on the precise value of the parameter measuring the actuation of automatic stabilizers (α1 ). Looking at the structure of the model, each one of the three different mechanisms determining the evolution of GDP and inflation (real interest rate, real exchange rate and budgetary balance) is related to a parameter of the IS equation and, also, to the behaviour of the monetary and fiscal authorities (see Table 3.2). We can obtain the relative parameters’ values that assure the return of the economy to its initial equilibrium if a shock takes place.34 This combination of parameters (or ‘stability condition’) assures that when an economy departs from its equilibrium and the inflation rate is below (above) the target, the output gap will be positive (negative) in the following periods. This will provoke a rising (falling) inflation until it returns to the central bank target. The aggregate demand equation (Romer, 2000) represents the relation between the inflation rate and the value of the output gap in the same period, and it will allow us to determine when the stability condition is met. The aggregate demand for country 1 can be derived by substituting in the IS (equation 1.2) the value of the real interest rate (equations (1.3) and ˙ t (U.3)), the rate of change of the real interest rate (difference between P1 ˙ t ) and the budgetary balance (equation (1.4)). Taking for simplicity and P2 the value of OG2 as given, we have the following equation:

OG1t =

˙ t − c(P1 ˙ t − P2 ˙ t ) + bβ1 P˙ T − 0.5b(1 + β1 )P2 ˙ t −0.5b(β1 − 1)P1 T OBJ ˙ ˙ − 0.5bβ2 OG2t − dBB − dα3 (P1t − P ) + g1t 1 + 0.5bβ2 + d(α1 + α2 )

(1.5)

For the model to be stable, the necessary condition is that the slope of the aggregate demand curve (that is, the variation of the OG when the rate

Stabilization and Fiscal Policy in the EMU 73 Table 3.3

Stability conditions with different fiscal policy rules

Fiscal policy rule

Stability condition

Automatic Stabilizers (SGPFR)

c>

Activist Fiscal Policy Rule (MAFR)

b 2 b dα3 c> − 2 2

of inflation changes) must be negative, and this may not be the case for certain combinations of the value of the parameters. More specifically, this slope is determined by the following expression: ∂OG1t −0.5b(β1 − 1) − c − dα3 = ˙ t 1 + 0.5bβ2 + d(α1 + α2 ) ∂P1

(1.6)

Providing β1 < 1, which is the case we can consider significant,35 the stability condition can be met, in the first place, if the value of parameter c is high enough, relative to b, to compensate for the rise in the real interest rate. Nevertheless, if the fiscal authorities actually display a more active fiscal policy, they can compensate the relative weakness of this competitiveness effect reinforcing the stabilizing mechanisms. But we can see that the only relevant parameter to solve the problem when c is too low is the reaction of the budgetary balance to the inflation rate (α3 ), not to the OG36 (α1 or α2 ). So, the automatic stabilizers won’t be able in any case to improve the stabilization capacity in the EMU when the national economies suffer specific shocks. In fact, this condition is necessary, but not sufficient. Solving the system formed by difference equations representing the simultaneous evolution of the inflation rate and output gap of each of the two economies we have this stability condition with the two fiscal rules. With the SGPFR, c must be higher than b/2. However, with the MAFR the stability condition is less strict, and a positive value of α3 allows the model to be stable even with a lower value of parameter c than b/2 (Table 3.3). The advantages of the active fiscal policy rule in relation to the Stability and Growth Pact are not limited only to the case when the competitiveness effect is too weak to neutralize the increasing real interest rate derived from the single monetary policy. If the real depreciation in country 1 has a strong enough effect on net exports and this is translated into a GDP increase, the economy will

74 Fiscal Policy in the European Union 2.1

0.4

2.0 0.0 1.9 ⫺0.4

1.8 1.7

⫺0.8

1.6

⫺1.2

1.5 1.4

⫺1.6

1.3

⫺2.0

1.2 2

4

6

8

2

10 12 14 16 18 20 22 24

Inflation SGPFR

2.4

4

6

8

10 12 14 16 18 20 22 24

Output gap SGPFR

Inflation MAFR

Output gap MAFR

0.2

2.3 0.0 2.2 ⫺0.2

2.1 2.0

⫺0.4

1.9

⫺0.6

1.8 1.7

⫺0.8

1.6

⫺1.0

1.5 2

4

6

8

10 12 14 16 18 20 22 24

Real interest rate SGPFR

Real interest rate MAFR

2

4

6

8

10 12 14 16 18 20 22 24

Budgetary balance SGPFR

Budgetary balance MAFR

Figure 3.10 A restrictive demand shock when the competitiveness effect is strong (country 1)

return to its initial equilibrium with both fiscal policy rules. However, it will do so with a different path. As we can see in Figure 3.10, when the government applies the MAFR the fiscal policy becomes more expansive than when it only allows the automatic stabilizers to operate. This is so because once the output gap becomes positive as a consequence of the higher impact of the competitiveness effect, the fiscal policy takes a restrictive stance because of the automatic stabilizers (α1 > 0, so the budgetary balance improves). Then, fiscal policy weakens the efficacy of the stabilizing effect, instead of reinforcing it. However, in the second fiscal policy rule this effect does not take place, because the inflation is still under the 2 per cent target and the government maintains a public deficit. Consequently, the equilibrium will be reached faster and, in accordance with our definition, the stabilization capacity will be also higher in this case.

Stabilization and Fiscal Policy in the EMU 75

6 Conclusions The establishment of the EMU has required the definition of a new framework for economic policy, which has been focused on the centralization of monetary policy and maintaining other economic policies under the national authorities. More specifically, fiscal policy is applied under the Stability and Growth Pact, which almost restricts its operation to the automatic stabilizers. This framework gives rise to some important problems. One of them, which has been considered in this chapter, is the loss of stabilization capacity that the creation of a monetary union involves for national economies. This problem is reflected in the fact that the effects of idiosyncratic shocks can be more persistent in the European national economies, increasing the standard deviation of the economic policy objectives (inflation and GDP) around their long-term targets, or that the economy doesn’t return to equilibrium or it takes longer to do so after a transitory shock. And this can be more evident when the countries present cyclical divergences, frequent specific shocks or structural differences. In these situations, the same monetary policy cannot be correct for all the countries. We have shown that the differences in inflation and growth rates have been actually highly persistent between the member countries. And we have summarized the operation of the stabilizing and destabilizing mechanisms in a monetary union through the calculation of the Interest Rate Misalignment indices and the evolution of the real interest and real exchange rates. Indeed, the stabilizing mechanisms (mainly, the real exchange rate) may not be strong enough to ensure the stabilization of national economies subject to asymmetric shocks or to do so quickly enough. This problem, and the lack of cyclical synchronization among the European economies, justifies the need to complement the common monetary policy with a more active fiscal policy than that which is derived from a strict application of the Stability and Growth Pact. More precisely, we have presented a simplified model of a monetary union formed by two countries which can experience specific supply and demand shocks and simulated the dynamics with some appropriate ‘activist’ fiscal policy rule (MAFR) and inside the Stability and Growth Pact framework (SGPFR). This fiscal rule we propose takes the same form as the Taylor rule for the monetary policy, and the authorities react by modifying the budgetary balance when the output gap is different from zero or when the inflation rate departs from its target.

76 Fiscal Policy in the European Union

When a restrictive demand shock takes place and the competitiveness effect is too weak, the inflation rate and the output gap will be ever decreasing in the case of the SGPFR, while the economy will return to the initial equilibrium with the MAFR. And, this does not depend on the precise value of the parameter measuring the actuation of automatic stabilizers, but only on the reaction of fiscal authorities to the inflation gap. Even more importantly, if we consider the case in which the competitiveness effect through a real depreciation is stronger, although the economy will return to its equilibrium with both fiscal policy rules, it will reach it faster with the Taylor-type rule. We conclude, then, that the MAFR can improve the stabilization capacity of the national economies. Because of this, we think that there is a case for the reform of the SGP to allow the national governments to adapt better their national fiscal policies to their specific national circumstances.

Annex: values of parameters used in the simulations To illustrate our arguments better, we have simulated the dynamic evolution of the two economic policy objectives (inflation rate and output gap) and the two economic policy instruments (real interest rate, set by the ECB, and budgetary balance, set by the national authorities). These simulations are developed using the following parameters’ values (Table 3.4): • Monetary policy rule parameters. We have given these parameters the value

originally proposed in Taylor (1993). So, the ECB changes the nominal interest rate by the same amount (0.5) when the inflation rate deviates from its target (β1 ) and when income differs from its potential value (β2 ). A change in these parameters, at least in a reasonable interval, would not alter the dynamic results of this chapter, although it would affect the speed with which each economy reaches equilibrium. We have also given the equilibrium interest rate the same value as Taylor – 2 per cent – and the inflation target is also established at 2 per cent. • Fiscal policy rule parameters. First, we have the parameters measuring the reaction of fiscal authorities to the inflation rate and to the output gap. We have given them the same values as in the monetary policy rule (0.5). However, as we have shown in the text, its value only influences the speed with which the economy returns to (or departs from) the equilibrium, not the stability of the

Stabilization and Fiscal Policy in the EMU 77 Table 3.4 Value of model parameters used in the simulations Parameter a (response of inflation to output gap) b (response of output gap to real interest rate) c (response of output gap to real exchange rate) d (response of output gap to the budgetary balance) β1 (response of nominal interest rate to inflation) β2 (response of nominal interest rate to output gap) Equilibrium real interest rate Target inflation rate α1 (automatic response of budgetary balance to output gap) α2 (discretionary response of budgetary balance to output gap) α3 (response of budgetary balance to inflation) Budgetary balance target

Simulated values 0.4 0.4 0.1; 0.3 1 0.5 0.5 2% 2% 0.5 0.0 0.5 0%

model. Another important value is the budgetary balance long-term target. For simplicity, we have supposed in the simulations that BBT = 0 in both countries, although it could be different in the MAFR. • Parameters measuring the effect of stabilizing mechanisms on income. These are parameter b and parameter d. Their relative value is particularly important because it determines whether the economies return to equilibrium or not after a shock without fiscal policy. We have therefore established a fixed value for parameter b (0.4) and we consider a case when parameter d is high enough (0.3) and too low (0.1). With regards to the specific values chosen, we have based them on the calibration performed by Galí (1998) and Aarle et al. (2004) of models with equations very similar to ours. According to these papers, both parameters will have less than unit values, and a reasonable value for c could be 0.4. However, we insist that should this value change, it would be important to know its value in relation to parameter c.The choice of this value, therefore, does not condition our results, which will always depend on the ratio between b and c. • Other structural parameters. The last parameter to be analysed measures the acceleration in inflation when income is at other than its potential level. As in the previous case, based on Galí (1998) and Aarle et al. (2004), it has been given a value of 0.4. If this parameter were to change value it would not alter the dynamic analysed qualitatively (in particular, the model’s stability conditions would be the same) but quantitatively, in the sense that it would change the speed with which, if applicable, equilibrium is reached.

78 Fiscal Policy in the European Union

Notes 1. Hemming et al. (2002) offer a survey of this empirical literature and report an average estimate tax multiplier of around 0.5 2. Tamborini (2003) describes what he calls the ‘Brussels consensus on macroeconomic policies in the EMU’ as a ‘policy-mix between common monetary policy and national automatic stabilizers’. 3. According to Article 105 of the European Union Treaty, ‘the primary objective of the ESCB shall be to maintain price stability. Without prejudice to the objective of price stability, the ESCB shall support the general economic policies in the Community with a view to contributing to the achievement of the objectives of the Community as laid down in Article 2.’ This establishes a hierarchical mandate for the ECB, while other central banks such as the Federal Reserve have a dual mandate with the price stability and growth and employment at the same level. 4. As the ECB (2005: 61) says: ‘Monetary policy is conducted by the Governing Council of the ECB with the primary objective of maintaining price stability in the euro area as a whole. Monetary policy does not, therefore, directly address differences in inflation rates or other economic developments which – because of the variety of economic structures or policies in place – may emerge across the euro area.’ 5. In March 2005 new exceptional circumstances were approved allowing deficits over 3 per cent GDP. 6. Compliance with the SGP in the first six years of the EMU has been very low: five EMU countries have had an excessive deficit in 2005 and only four countries have registered a surplus. 7. For the European Commission (2002: 113) ‘the underlying philosophy of the SGP is sceptical on fiscal fine-tuning: stabilization should be achieved by the operation of automatic stabilizers and discretionary fiscal policy, while not ruled out altogether, should be confined to a limited set of circumstances where its usefulness is well established. Hence, it should be subject to a careful case-by-case examination by the country concerned and by the Eurogroup given the potential spillovers.’ 8. For a review of national fiscal rules in Europe see European Commission (2006) and Fernández Llera (2005). 9. In this chapter we assume the New Consensus on macroeconomics approach to the potential output. This is the level of income compatible with a constant rate of inflation and is mainly determined by supply-side forces. When the level of income reaches its potential value, the rate of unemployment is the NAIRU. In this approach it is assumed that stabilization policies have via changes in aggregate demand only short-run effects. For a different view see Fontana and Palacio-Vera (2007). 10. A classic paper in this respect is Bayoumi and Eichengreen (1993). Subsequent debates have focused, however, on the possibility of deeper economic integration either favouring the productive specialization of European economies (increasing the likelihood of an asymmetric shock even further) or generating a tendency towards closer approximation of productive structures. 11. The three factors primarily explaining these differentials are the different reaction to common shocks (oil prices and euro exchange rate), cyclical differences and the Balassa-Samuelson effect.

Stabilization and Fiscal Policy in the EMU 79 12. Roubini et al. (2007) remark that, first, there is only slight evidence of per capita GDP convergence; second, the growth differentials have been persistent over that time period; and third, there are ambiguous and conflicting empirical signals about the degree of business cycle synchronization since the inception of the EMU. 13. For the ECB (ECB, 2007) the dispersion of real GDP growth rates across the euro area countries has largely reflected lasting trend growth differences and, to a lesser extent, cyclical differences and the breakdown of the variance of real GDP growth, since the beginning of the 1990s, points to a large decrease in the contribution to dispersion from the cyclical component and, simultaneously, to a large increase in the contribution from trend growth differences. 14. For Dullien and Fritsche (2006: 2) ‘given both inflation and growth trends, the common interest rate of the European Central Bank has been perceived to be inappropriate both for fast-growing as well as slow-growing countries. The overall impression of economic divergence has been underlined by a deterioration of public finances in the slow growth countries, notably Italy and Germany.’ 15. Papademos (2007: 1) highlights that the difference between the economic divergences in the euro area from those in the United States is the persistence of the inflation and growth performance across euro area countries. And he says that ‘this feature, which is not characteristic of the United States, may suggest that the underlying adjustment mechanisms in the euro area economies are not functioning as smoothly and in a timely manner, with potentially undesirable implications for the dynamics of economic activity and employment’. 16. Roubini et al. (2007) argue that the real interest rate channel has led to a destabilizing adjustment in Spain, Portugal, the Netherlands, Greece and possibly Italy. 17. These advantages include the clear expression of monetary policy’s commitment to price stability, avoiding a possible inflationist bias, more effective guidance for public expectations and reinforcement of the credibility and transparency of central bank activities. 18. These limitations have to do with the loss of information derived from reducing the equation to so few variables, the difficulty of systematizing all the contingencies that monetary policy may face and the relative uncertainty of some of the variables which have to be used in decision-making processes. In the Taylor rule, this is particularly important in the case of the equilibrium interest rate and the output gap, which are not directly observable. 19. There are a lot of papers estimating the ECB’s reaction function. For example, see Gerdesmeier and Roffia (2003). 20. See Carlin and Soskice (2006) for a formal determination of the optimal value of these parameters. 21. The IMF (2004) calculates a similar index of the adjustment of the ECB monetary policy to the different cyclical situations of the EMU members using different specifications of the Taylor rule and estimating what are called ‘monetary gaps’. 22. Our argument would be strengthened even further if this were not the case. Hayo (2006: 1) estimates Taylor rules for each EMU country with data

80 Fiscal Policy in the European Union

23.

24.

25.

26.

27. 28.

29.

previous to 1999 and then he obtains the interest rate that the national central banks have set since 1999, given actual inflation rates and output gaps. He finds that ‘for almost all EMU member countries euro area interest rates tend to be below the national target interest rates, even after explicitly accounting for a lower real interest rate in the EMU period, with Germany being the only exception’. Since 2000, European countries have been trying to apply a structural reform programme – the Lisbon Strategy – which could increase the efficacy of these mechanisms. If this was the case, it would reduce the need to apply other policies – fiscal, for instance – to solve the problem of asymmetric shocks in the EMU, although it would not completely disappear. However, the progress made in the Lisbon Strategy in the first few years has not been as great as expected, as shown in Kok (2004). To analyse the competitiveness we use the Harmonized Competitiveness Indictors that the European Commission (2007) has recently developed. These indicators, which are based on consumer price indices or costs, are calculated on the basis of weighted averages of the bilateral exchange rates of each euro area country vis-à-vis the currencies of its trading partners. We compare the real exchange rates based on unit labour costs in the whole economy (ULCE) as an approximation to the evolution of competitiveness. We analyse this through the evolution of unit labour cost because this is a key determinant of inflation and of the changes in competitiveness. Depending on the index, the loss of competitiveness of Spain vis-à-vis the euro zone ranges between 5 per cent and 12 per cent. Looking specifically at the Spanish experience, López-Salido et al. (2005: 31) say that ‘the terms of trade effect does not play a strong stabilizing role as has been suggested sometimes in the literature . . . We have found that, in general, the deterioration of the terms of trade that inflation differentials cause does not have a quantitatively relevant stabilizing effect as its impact on output is relatively small in the short term.’ There are, however, some articles estimating fiscal rules to represent the behaviour of different authorities (Galí and Perotti, 2003; García-Serrador et al., 2008). Other papers, more similar to ours, propose different fiscal policy rules and analyse their possible effects on the economy using a macroeconomic model. Some examples are Taylor (1995, 2000), Aarle et al. (2004), Setterfield (2005), Alonso and García (2004) and Alonso et al. (2007). Z is a random variable with zero mean and known variance. When the real interest rate is set at its equilibrium value the output gap is equal to zero. This equilibrium value is defined for budgetary balance and constant real exchange rate. Buti and Martínez Mongay (2005: 31) refer to the fiscal policy being applied in Spain as follows: ‘Is there enough fiscal tightening? With credit growing above 15 per cent, coupled with high household indebtedness and a large and persistent inflation differential, risks seem to be on the downside. With a view to preventing economic imbalances, in particular overheating, from deepening further, fiscal policy should be tightened’ (our emphasis). Or the International Monetary Fund (2006) says that the task of fiscal policy in Spain during 2006 should be to ‘dampen the aggregate demand’ because of high inflation.

Stabilization and Fiscal Policy in the EMU 81 30. Taylor (2000) and Seidman (2003) specifically suggest that this balance should be equal to zero. However, according to Calmfors (2002), ‘the exact target should depend on a number of factors: the present debt situation, the requirements posed by the demographic situation and the size of automatic stabilizers’. 31. Setterfield (2005) calls this fiscal policy rule a ‘pseudo Taylor rule’. 32. In the Annex we explain the parameters’ values we have used in the simulations. 33. The evolution of country 2 would be just the contrary of country 1. For example, when the inflation is falling in country 1 because of negative OG, the competitiveness is improving in country 2, giving rise to a positive OG and a rising inflation. 34. See Uxó and Arroyo (2007) for a deeper analysis of this condition. 35. Taylor (1993) gave this parameter a value of 0.5. 36. Given that the denominator is always positive, the sign of equation (1.6) does not depend on the values of α1 or α2 .

References Aarle, B. van, H. Garretsen and F. Huart, ‘Monetary and Fiscal Policy Rules in the EMU’, German Economic Review, 5(4) (2004). Alonso, L. A. and P. García, ‘Central Bank Independence: Taylor Rules and Fiscal Policy’, Documentos de Trabajo de la Facultad de Ciencias Económicas y Empresariales, Universidad Complutense de Madrid, 0401 (2004). Alonso, L. A., L. Corchón and V. Guzmán, ‘Instability and Trade in Currency Areas’, Economic Letters, 94(1) (2007): 71–5. Arestis, P. (ed.), Is There a New Consensus in Macroeconomics? (Basingstoke: Palgrave Macmillan, 2007). Arestis, P. and M. Sawyer, ‘Fiscal Policy: a Potent Instrument’, The New School Economic Review, 1(1) (2004). Bayoumi, T. and B. Eichengreen, ‘Shocking Aspects of European Monetary Integration’, in F. Torres and F. Gaviazzi (eds), Adjustment and Growth in the European Monetary Union (Cambridge: Cambridge University Press, 1993). Bean, C., ‘Is There a New Consensus in Monetary Policy?’, in P. Arestis (ed.), Is There a New Consensus in Macroeconomics? (Basingstoke: Palgrave Macmillan, 2007). Blinder, A., ‘The Case Against the Case Against Discretionary Fiscal Policy’, CEPS Working Paper 100 (2004). Buti, M. and C. Martínez Mongay, ‘Country Study: Spain in the EMU: a Virtuous Long-lasting Cycle?’, European Commission, Occasional Paper 14 (2005). Calmfors, L., ‘Fiscal Policy as a Stabilization Policy Tool in the EMU’, EPRUNetwork Conference, 22 May 2002. Calmfors, L., ‘Fiscal Policy to Stabilize the Domestic Economy in the EMU: What Can We Learn from Monetary Policy?’, CESIFO Economic Studies, 49(3) (2003). Carlin, W. and D. Soskice, Macroeconomics: Imperfections, Institutions and Policies (Oxford: Oxford University Press, 2006). Doménech, R., M. Ledo and D. Taguas, ‘Some New Results on Interest Rate Rules in the EMU and in the US’, Journal of Economics and Business, 54(4) (2002).

82 Fiscal Policy in the European Union Dullien, S. and U. Fritsche, ‘How Bad is Divergence in the Euro-Zone? Lessons from the United States of America and Germany’, University of Hamburg, Department of Economics and Politics Discussion Papers, Macroeconomics and Finance Series 5, 2006. European Central Bank, ‘Issues Related to Monetary Policy Rules’, Monthly Bulletin, October (2001). European Central Bank, Inflation Differentials in the Euro Area: Potential Causes and Policy Implications, Frankfurt (2003). European Central Bank, ‘The Natural Real Interest Rate in the Euro Area’, Monthly Bulletin, May (2004). European Central Bank, Monetary Policy and Inflation Differentials in a Heterogeneous Currency Area’, Monthly Bulletin, May (2005). European Central Bank, ‘Output Growth Differentials in the Euro Area: Sources and Implications’, Monthly Bulletin, April (2007). European Commission, Public Finances in EMU – 2002 (2002). European Commission, Public Finances in EMU – 2006 (2006). European Commission, ‘Price and Cost Competitiveness’, http://ec.europa.eu/ economy_finance_publications/price-cost-competitiveness/(2007). European Council, ‘Mejoras en la implementación del Pacto de Estabilidad y Crecimiento’, Conclusiones de la Presidencia, Annex II, 22 and 23 March (2005). Fernández Llera, R., ‘Estabilidad Presupuestaria y autonomía financiera: ¿disyuntiva o conjunción necesaria?’, Seminarios de Economía Pública, Instituto de Estudios Fiscales (2005). Fontana, G. and A. Palacio-Vera, ‘Are Long-Run Price Stability and Short-Run Output Stabilization All That Monetary Policy Can Aim For?’, Metroeconomica, 58(2) (2007). Gerdesmeier, D. and B. Roffia, ‘Empirical Estimates of Reaction Function for the Euro Area’, ECB Working Paper, 206 (2003). Galí, J., ‘La política monetaria europea y sus posibles repercusiones sobre la economía española’, Seminar on El euro y sus repercusiones sobre la economía española, Fundación BBVA (1998). Galí, J. and R. Perotti, ‘Fiscal Policy and Monetary Integration in Europe’, NBER Working Paper, 9773 (2003). García-Serrador, A., M. J. Arroyo, R. Mínguez and J. Uxó, ‘Estimation of a Fiscal Rule for EMU Countries (1984–2005)’, Applied Economics, 40, April (2008). Hayo, B., ‘Is European Monetary Policy Appropriate for EMU Member Countries? A Counterfactual Analysis’, Marburg Papers on Economics, 10-2006 (2006). Hemming, R., M. Kell and S. Mahfouz, ‘The Effectiveness of Fiscal Policy in Stimulating Economic Activity: a Review of the Literature’, IMF Working Papers, 02208 (2002). International Monetary Fund, ‘Has Fiscal Behavior Changed under the European Economic and Monetary Union?’, World Economic Outlook, September (2004). International Monetary Fund, Spain: Staff Report for the 2006 Article IV Consultation (2006). Kok, W., ‘Facing the Challenge: the Lisbon Strategy for Growth and Employment’, Report of the High Level Group chaired by Wim Kok, available at http://europa.eu.int/growthandjobs/pdf/kok_report_en.pdf (2004). López-Salido, J. D., F. Restoy and J. Vallés, ‘Inflation Differentials: the Spanish Case’, Bank of Spain, Working Papers, 0514 (2005).

Stabilization and Fiscal Policy in the EMU 83 Mongelli, F. P., ‘New Views on the Optimum Currency Area Theory: What is EMU Telling Us?’, ECB Working Papers, 138 (2002). Papademos, L., ‘Inflation and Competitiveness Divergences in the Euro Area Countries: Causes, Consequences and Policy Responses’, The ECB and its Watchers IX, Frankfurt am Main, September (2007). Romer, D., ‘Keynesian Macroeconomics without the LM Curve’, Journal of Economic Perspectives, 14(2) (2000): 149–69. Roubini, N., E. Parisi-Capone and C. Menegatti, ‘Growth Differentials in the EMU: Facts and Considerations’, Roubini Global Economics Service (2007). Seidman, L. S., Automatic Fiscal Policies to Combat Recessions (New York: M. E. Sharpe, 2003). Setterfield, M., ‘Is There a Stabilizing Role for Fiscal Policy in the New Consensus?’, paper presented at the Meetings of the Eastern Economic Association, New York City (2005). Tamborini, R., ‘The “Brussels Consensus” on Macroeconomic Stabilization Policies in the EMU: a Critical Assessment’, paper presented at the 1st EUI Alumni Conference, Governing EMU: Political, Economic, Legal and Historical Perspectives (2003). Taylor, J. B. ‘Discretion versus Policy Rules in Practice’, Cranegie-Rochester Conference Series on Public Policy, 39 (1993). Taylor, J. B., Macroeconomic Policy in a World Economy (New York: W.W. Norton, 1995). Taylor, J. B., Reassessing Discretionary Fiscal Policy, Stanford University (2000), available at http://www.stanford.edu/∼johntayl/Papers/Reassessing+Revised.pdf Uxó, J. and M. J. Arroyo, ‘Alternative Fiscal Policy Rules and the Stabilization Problem in EMU: Theory and Simulations’, in P. Arestis (ed.), Advances in Monetary Policy and Macroeconomics (Basingstoke: Palgrave Macmillan, 2007).

4 Fiscal Adjustment and Composition of Public Expenditures in the EMU Jesús Ferreiro, M. Teresa García-del-Valle and Carmen Gómez

1 Introduction The rules and norms set out in the Maastrich Treaty and in the Stability and Growth Pact (SGP) have involved for the European Monetary Union (EMU) member states the implementation of an orthodox strategy of fiscal policy based on the correction of fiscal imbalances and cuts in the current sizes of public revenues and expenditures. This strategy is based on a more general view according to which demand-side policies do not affect economic activity in the long run, that is, fiscal policy does not influence the path of potential output. However, in the last few years, fiscal policy has been gaining in relevance. In this sense, the Lisbon Strategy and the current Broad Economic Policy Guidelines (BEPG) and, even, the recent reform of the Stability and Growth Pact, all accept the potential positive impact that fiscal policy can have on economic activity in the long run, both in terms of the level of potential output and the long-term rate of growth of potential output. This positive impact would come from the composition of public expenditure, not from the size of public expenditures or revenues or from the fiscal balance. The key would be the recomposition of public expenditures, increasing the share of the ‘productive’ expenditures. These would be those expenditures with a positive impact on the productive factors (capital and labour) endowment and the productivity of productive inputs. Consequently, public expenditures, that is, the composition of public expenditures, can be instrumented with the aim of influencing the level and the rate of growth of economic activity. The EMU’s orthodox strategy of macroeconomic policy gives a special role to monetary policy. Fiscal policy in the EMU is determined by the rules and norms set out in the Maastricht Treaty and the Stability and 84

Fiscal Adjustment and Public Expenditures

85

Growth Pact (SGB), according to which the working of national fiscal policies in the euro zone is based on the following principles: 1. The need to implement sound and sustainable fiscal policies. 2. The emphasis on reducing the size of public deficits and the stocks of public borrowing. 3. The reduction of the current levels of public expenditure and taxation, both measured in percentage of GDP. The aim of this chapter is to analyse whether membership of the EMU has involved a change in the national public finances towards those productive public expenditures, and whether, therefore, national fiscal policies in the EMU could be enhancing the long-term economic growth path. The chapter is structured as follows. Section 2 shows the theoretical bases of the fiscal policy in the EMU. Section 3 studies the evolution of public expenditure in EMU member states, focusing the analysis on the evolution of those items that are defined in the literature as productive public spending, both in terms of the economic and functional classifications. Section 4 concludes.

2 The role of fiscal policy in the European Monetary Union: theoretical bases The European Monetary Union has involved the implementation of economic policy based on what is usually named the New Consensus in Macroeconomics (Arestis and Sawyer, 2004). In terms of the macroeconomic policies, that view gives a special role to monetary policy, downgrading fiscal policy to a secondary role, always subordinated to the former: thus, fiscal policy is focused on the generation and maintenance of the right environment for the effective working of monetary policy. These theoretical foundations have led to a fiscal policy determined by the rules and norms set out in the Maastricht Treaty and the Stability and Growth Pact. In this sense, the working of national fiscal policies in the euro zone is based on the three principles we mentioned above, namely, the need to implement sound and sustainable fiscal policies; the objective to reduce the size of public deficits and the stocks of public borrowing; and to reduce current levels of public expenditure and taxation, measured as a percentage of GDP. These basic principles are based on an orthodox theoretical background. According to this basis, first, in the long run fiscal policy cannot

86 Fiscal Policy in the European Union

affect the economic activity (as measured by potential output), and, second, the only possible impact of fiscal policy is generated in the short run. Consequently, an active fiscal policy can only be implemented on a short-term perspective, focused on the correction of cyclical disequilibria (acting on the output gap) through the working of built-in stabilizers. The long-term effects of fiscal policy, mainly those related to the existence of permanent deficits, would arise not from the traditional Keynesian effects but from the non-Keynesian effects of fiscal policy (Afonso, 2001; European Commission, 2004; Giavazzi et al., 1999; Hemming et al., 2002). The literature about the non-Keynesian effects of fiscal policy has given rise to the development of a number of studies focusing on the expansionary impact of fiscal consolidation. These studies deny the existence of expansionary effects generated by the traditional expansionary Keynesian fiscal policies based on the generation of fiscal deficits. Moreover, they argue that fiscal consolidation episodes have an expansionary impact on economic activity both in the short term and in the long term.1 Thus, after the consolidation episode the rate of economic growth is well above that registered before the fiscal adjustment (Afonso, 2001, 2006; Alesina and Perotti, 1995, 1997; Alesina et al., 1998, 2002; Briotti, 2004, 2005; European Commission, 2003, 2007; Giavazzi and Pagano, 1990; Giavazzi et al., 1999, 2000; Giudice et al., 2003, 2007; Kumar et al., 2007; McDermott and Wescott, 1996; Aarle and Garretsen, 2003). The subordination of fiscal policy to monetary policy and the rejection of discretionary management of fiscal instruments requires that fiscal policy adopt a passive role. On the one hand, the objective of fiscal policy is the creation of a sound economic environment via the reduction/removal of fiscal imbalances that gives rise to low inflation rates and that contributes to the effective working of the monetary policy set by the European Central Bank. On the other hand, the counter-cyclical fiscal policy must be based on the working of built-in stabilizers, and, consequently, only cyclical fiscal deficits are allowed, although their size cannot be higher than 3 per cent of GDP.2 This strategy of fiscal policy is based on a more general view according to which demand-side policies do not affect economic activity in the long run, that is, fiscal policy does not influence the path of potential output. However, in the last few years, fiscal policy has become more relevant. In this sense, the Lisbon Strategy and the current Broad Economic Policy Guidelines (BEPG) accept the potential positive impact that fiscal policy can have on economic activity in the long run, both in terms of the level of potential output and the long-term rate of growth of potential

Fiscal Adjustment and Public Expenditures

87

output. This positive impact would come from the composition of public expenditure, not from the size of public expenditures or revenues or from the fiscal balance. The key would be the recomposition of public expenditures, increasing the share of the ‘productive’ expenditures. These would be those expenditures with a positive impact on the productive factors (capital and labour) endowment and the productivity of productive inputs. Consequently, public expenditures, or rather, the composition of public expenditures, can be instrumented with the aim of influencing the level and the rate of growth of economic activity. Public expenditure items would be included in the set of instruments belonging to the supply-side or structural policies. Their objective would be, from a longrun perspective, to affect the level and the rate of growth of potential output via their impact on the aggregate supply. That involves the existence and working of a new transmission channel of fiscal policy different from the traditional Keynesian channels based on the impact of fiscal measures on aggregate demand and its components. These arguments are now being accepted by the European institutions. Thus, in the Report of the Commission and the ECOFIN Council, it is argued that public budgets can contribute to foster economic growth and employment through three channels: supporting a stable macroeconomic framework via sound public finances, making tax and benefit systems more employment friendly and redirecting public expenditures towards physical and human capital accumulation. Broadening the focus from achieving budgetary stability towards putting the emphasis on the overall contribution which public finances can make to growth and employment marks a new step in the fiscal agenda in EMU. (Council of the European Union, 2001: 1) More recently, in the Broad Economic Policy Guidelines for the 2005–8 period, in guideline number three it is stated that: To promote a growth- and employment-orientated and efficient allocation of resources Member States should, without prejudice to guidelines on economic stability and sustainability, re-direct the composition of public expenditure towards growth-enhancing categories in line with Lisbon strategy. (European Commission, 2005a: 40)

88 Fiscal Policy in the European Union

Furthermore: Well-designed tax and expenditure systems that promote an efficient allocation of resources are a necessity for the public sector to make a full contribution towards growth and employment, without jeopardising the goals of economic stability and sustainability. This can be achieved by redirecting expenditure towards growth-enhancing categories such as research and development (R&D), physical infrastructure, environmental friendly technologies, human capital and knowledge. (European Commission, 2005a: 41) In this sense, the recent reform of the Stability and Growth Pact gives the composition of public expenditures a key role in the evaluation of the situation of public finances. The Commission’s Report that has to evaluate the existence of an excessive deficit ‘shall appropriately reflect developments in the medium-term economic position (in particular potential growth, prevailing cyclical conditions, the implementation of policies in the context of the Lisbon agenda and policies to foster research and development and innovation) and developments in the medium-term budgetary position (in particular, fiscal consolidation efforts in “good times”, debt sustainability, public investment and the overall quality of public finances)’.3 The public policy endogenous growth models are the theoretical basis of this new fiscal policy strategy. These models are, actually, an extension of endogenous growth theory. According to this theory, economic growth is determined by the accumulation of productive factors (physical and human capital) and by technical progress (Aghion and Howitt, 1992; Barro, 1990; Barro and Sala-i-Martin, 1995; Grossman and Helpman, 1991; Jones, 1995; Lucas, 1988, 1990; Rebelo, 1991; Romer, 1986, 1990). Public policy endogenous growth models focus on the role that fiscal policy can play in enhancing or retarding economic growth. According to Gemmell and Kneller (2001), fiscal policies can affect growth through several ways: production externalities, productivity growth, productivity differences between the public and private sector, fiscal effects on factor accumulation, crowding-out and redistribution. These models argue that shifting the revenue stance away from distortionary forms of taxation and towards non-distortionary forms, and switching expenditures from unproductive to productive forms are growth-enhancing (Angelopoulos et al., 2006; Aschauer, 1989; Baier and Glomm, 2001; Barro, 1990, 1991; Barro and Sala-i-Martin, 1992, 1995; Chamley, 1986; Devarajan et al., 1996; Gemmel and Kneller, 2001; Gupta el al., 2005; Jones et al., 1993; King and Rebelo, 1990; Kneller et al.,

Fiscal Adjustment and Public Expenditures

89

1999, 2001; Kocherlatoky and Yi, 1997; Mendoza et al., 1997; Romero de Avila and Strauch, 2003; Zagler and Dürnecker, 2003). Therefore, fiscal policy can determine both the level of output and the steady-state growth rate (Kneller et al., 1999; Lamo and Strauch, 2002). As we have mentioned above, productive expenditures are growthenhancing whilst non-productive expenditures are growth-retarding. Since in endogenous growth models the (potential or long-term) economic growth is determined by factor endowments and technical progress, the public policy endogenous growth models define as ‘productive’ expenditures those that, by complementing private sector production and generating positive externalities to firms, have a positive effect on the marginal productivity of capital and labour, and ‘unproductive’ expenditures would be those that give direct utility to households (Angelopoulos et al., 2006; Devarajan et al., 1996; European Commission, 2004). Although the empirical evidence in support of endogenous growth through fiscal policy is mixed, for a number of studies ‘productive’ expenditures would include spending on the following items: public investment, R&D, education, active labour market policies, health, defence, public order and general administrative costs, transport and communication (Afonso et al., 2005; Angelopoulos et al., 2006; Aschauer, 1989; Atkinson and van den Noord, 2001; Barro, 1990, 1991; Bleaney et al., 2001; Bloom et al., 2001; Devarajan et al., 1996; Easterly and Rebelo, 1993; Fölster and Henrekson, 1998; Gemmel and Kneller, 2001; Gwartney et al., 1998; Kneller et al., 1999, 2001; Lamo and Strauch, 2002; Nourzad and Vrieze, 1995; Romero de Avila and Strauch, 2003; Sanchez-Robles, 1998; Thöne, 2003). Nonetheless, the above conclusion must be taken with caveats. Empirical studies show that the impact of public expenditures depends on the kind of taxation that finances that expenditure, that is, the impact depends on the expenditure being financed with a distortionary or a nondistortionary taxation (Kneller et al., 1999, 2001; Kocherlatoky and Yi, 1997; Romero de Avila and Strauch, 2003). As Kneller et al. (1999: 174) argue: ‘Non-distortionary tax-financed increases in productive expenditures are predicted to have a positive impact upon the growth rate, whereas with distortionary-tax financing the predicted growth effect is ambiguous. Finally, non-productive expenditures financed by a distortionary tax have an unambiguously negative growth effect, but a zero effect is expected if non-distortionary tax finance is used.’ Another caveat is the existence of non-linear relations between economic growth and the size of public spending (Cameron, 1978). Thus,

90 Fiscal Policy in the European Union

although public spending may have a positive impact on growth, the impact can be negative if expenditure exceeds a certain threshold, this threshold being different for each type of spending (European Commission, 2002; Gupta et al., 2005). Finally, empirical studies face two problems. The first problem is that national statistics on government outlays are not fully comparable (Florio, 2001). The second one is that data corresponding to the theoretical classification of public outlays into ‘productive’ or ‘unproductive’ are not available at the macroeconomic level. That forces us to use data coming from national accounts, either on the basis of the economic or the functional classification, and to assume that in each case all the spending in a certain category is either productive or unproductive (European Commission, 2004). That means, for instance, that spending on subsidies or wages and salaries can be considered as unproductive, regardless of whether those outlays finance spending on education and R&D or spending on recreation. In contrast, all the spending on public investment is considered as productive regardless of their ultimate use.

3 An analysis of the impact of fiscal adjustments in the EMU on the quality of public expenditures Table 4.1 shows the evolution of fiscal deficits and public expenditures since 1990. After the EMU, fiscal deficits have significantly fallen in all EMU member states, with the exception of Germany, whose fiscal deficit is 0.2 per cent lower. With the exceptions of Luxembourg and Portugal, fiscal adjustment has come with a decrease in public expenditure. According to the expansionary fiscal adjustment models, the quality of fiscal expenditures composition is related to the impact of fiscal adjustment on long-term economic growth. Thus, those fiscal adjustments based on a cut in public outlays may have an expansionary effect, mainly if there is a simultaneous cut in taxation. Figure 4.1 shows the evolution, in percentage points of GDP, of public expenditures and taxation in EMU countries between the periods 1990–8 and 1999–2005. Three groups of countries emerge from that analysis. In the first group (Portugal and Luxembourg) there has been both a rise in taxation and public expenditures. In the second group (formed by Belgium, France, Greece and Spain) the falling expenditures have come with higher taxation. And, finally, in the third group (Austria, Finland, Germany, Ireland, Italy and the Netherlands) there has been a fall in both public expenditures and taxation.

91 Table 4.1

Evolution of fiscal deficit and public expenditure (% GDP) Fiscal deficit

Public expenditure

1990–1998 1999–2005 Difference 1990–1998 1999–2005 Difference (1) (2) (3 = 2 − 1) (4) (5) (6 = 5 − 4) Belgium Germany∗ Ireland Greece Spain∗∗ France Italy Luxembourg Netherlands Austria Portugal Finland

5.03 2.75 1.29 9.14 4.48 4.13 8.08 −2.15 2.98 3.41 5.46 2.82

−4.67 −0.20 −2.86 −4.25 −4.03 −1.46 −5.19 −0.25 −2.26 −2.06 −1.84 −6.33

0.35 2.54 −1.56 4.89 0.46 2.67 2.88 −2.40 0.72 1.35 3.62 −3.51

52.42 47.94 41.30 45.44 42.58 52.97 52.97 39.71 52.22 54.03 42.95 58.43

−2.40 −0.61 −7.88 −0.40 −3.69 −0.31 −5.23 0.66 −6.49 −2.98 2.02 −8.85

50.02 47.33 33.42 45.03 38.90 52.66 47.74 40.37 45.73 51.05 44.97 49.58

Source: AMECO Database. Data based on ESA95. A − sign means a fiscal surplus. ∗ Data for Germany start in 1991. ∗∗ Data for Spain start in 1995.

Greece

4

Portugal 3

Belgium

2

France

Revenues

1 Italy

0

Austria

⫺1 ⫺2

Spain

Luxembourg Germany

Finland

⫺3 ⫺4 ⫺5 ⫺6

Netherlands

Ireland ⫺10 ⫺9 ⫺8 ⫺7 ⫺6 ⫺5 ⫺4 ⫺3 ⫺2 ⫺1 Expenditures

0

1

2

3

Figure 4.1 Evolution of public expenditures and taxation between the periods 1990–8 and 1999–2005 (% GDP) Source: Our calculations based on AMECO.

92 Fiscal Policy in the European Union

In this sense, for most EMU member states fiscal adjustment processes should be considered as expansionary ones, since in most cases they are based on expenditure cuts. Therefore, the quality of fiscal adjustments, and, consequently, the quality of public expenditures size, would have improved. However, the literature on endogenous growth models stresses the argument that what really matters is not so much the size of total public outlays but their composition. Tables 4.2(a), 4.2(b) and 4.2(c) show the evolution of public expenditures between 1990–8 and 1999–2005 in the twelve member states of EMU, focusing the analysis on the economic classification of public spending. As mentioned above, the empirical and theoretical literature assumes that only public investment and social benefits, depending on the size of the latter, can be considered as positive public expenditures items. In the case of public investment (defined in generic terms as the sum of capital transfers and gross capital formation), this item has only increased in Ireland. In the rest of the countries, this kind of spending has declined, ranging from 0.1 per cent GDP in Greece to 1.43 per cent GDP in Germany. The cases of Germany, France and Portugal are particularly remarkable. In Germany and France the fall in public investments more than offsets the increase in the rest of public outlays, thus leading to a fall in total public expenditures. In the case of Portugal, the size of the fall in public investment is smaller than the increase in the rest of public outlays, and, therefore, total public expenditure increases. In the case of spending on social protection, measured as the sum of expenditures in social benefits and transfers in kind, in two cases, Belgium and Luxembourg, these items have remained nearly unchanged, with a decline of only some hundredths of GDP percentage points; in four countries, Ireland, Spain, Netherlands and Finland, this expenditure declines; and, finally, it increases in the other six economies. As we mentioned, it is assumed that the economic impact of social expenditures depends on their size, and, thus, the impact would be positive if the size is ‘low’ and negative with an excessive ‘level’ of social expenditures. Figure 4.2 can help to evaluate the economic impact on the long-term economic growth of the evolution of this kind of expenditure and, therefore, the quality of public expenditures in the euro zone. The countries above the bisectrix are the countries where social expenditures have declined after EMU, and in the countries below the bisectrix social expenditures have increased. We can detect three groups of countries. In the first group, formed by Greece, Ireland, Spain and Portugal, the size of social expenditures is

Table 4.2(a) Evolution of the size of public expenditures (in % GDP) between the periods 1990–8 and 1999–2005 by kind of expenditure (economic classification) Belgium Germany∗ Ireland Greece Spain∗∗ France Italy Intermediate consumption Compensation of employees Subsidies Property income Social benefits other than social transfers in kind Social transfers in kind Other current transfers Capital transfers Gross capital formation Acquisitions less disposals of non-financial nonproduced assets Total general government expenditure

Luxembourg Netherlands Austria Portugal Finland

0.31 0.11 −0.05 −4.00 −0.77

−0.14 −0.87 −0.44 −0.26 1.66

−0.39 1.51 −0.45 −0.46 0.18 −1.57 0.07 −2.29 0.17 −0.95 −0.41 −0.35 −0.11 −0.08 −0.49 −4.35 −4.95 −2.63 −0.32 −5.26 −3.08 1.08 −3.08 −0.13 0.44

0.18 −1.37 −0.67 −0.21 −0.71

0.49 −0.72 −0.25 −2.50 −4.61

−0.93 −1.88 −0.01 −0.39 0.17

0.66 1.06 −0.09 −3.63 2.05

−0.17 −2.25 −1.26 −1.07 −3.83

0.75 0.22 −0.09 −0.07 −0.01

0.35 0.13 −0.76 −0.67 −0.34

−0.03 0.00 −0.02 0.54 0.33 0.17 0.19 0.15 0.45 0.27 0.06 −0.51 −0.44 −0.31 0.09 1.39 0.41 −0.36 −0.21 −0.24 −0.03 0.02 −0.08 0.00 −0.23

0.61 0.10 −0.10 −0.02 −0.30

0.49 0.63 −0.83 0.12 0.19

0.93 0.23 0.54 −1.48 −0.10

1.12 0.34 −0.30 −0.24 −0.17

0.49 0.42 −0.68 −0.40 −0.12

−3.58

−1.42

−8.16 −2.49 −9.37 −0.31 −5.22

−2.45

−7.00

−2.98

0.72

−8.85

Source: EUROSTAT. ∗ Periods compared are (1991–8) and (1999–2005). ∗∗ Periods compared are (1995–8) and (1999–2005).

93

94 Table 4.2(b) Evolution of the size of public expenditures (in % GDP) between the periods 1990–8 and 1999–2005 by kind of expenditure (economic classification) Belgium Germany∗ Ireland Greece Spain∗∗ France Italy Total general government expenditure Property income Capital transfers + gross capital formation Social benefits and transfers in kind Compensation of employees Subsidies Other current transfers Acquisitions less disposals of non-financial non-produced assets Intermediate consumption Other expenditures

Luxembourg Netherlands Austria Portugal Finland

−3.58

−1.42

−8.16 −2.49 −9.37 −0.31 −5.22

−2.45

−7.00

−2.98

0.72

−8.85

−4.00 −0.16

−0.26 −1.43

−4.35 −4.95 −2.63 −0.32 −5.26 1.45 −0.10 −0.80 −0.52 −0.14

−0.21 −0.12

−2.50 −0.71

−0.39 −0.94

−3.63 −0.54

−1.07 −1.08

−0.03

2.01

−3.10

0.76

−0.09

−4.11

1.11

3.18

−3.34

0.11 −0.05 0.22 −0.01

−0.87 −0.44 0.13 −0.34

−1.57 0.07 −2.29 0.17 −0.95 −0.41 −0.35 −0.11 −0.08 −0.49 0.17 0.19 0.15 0.45 0.27 −0.03 0.02 −0.08 0.00 −0.23

−1.37 −0.67 0.10 −0.30

−0.72 −0.25 0.63 0.19

−1.88 −0.01 0.23 −0.10

1.06 −0.09 0.34 −0.17

−2.25 −1.26 0.42 −0.12

0.31 0.00

−0.14 0.00

−0.39 0.00

0.18 0.00

0.49 0.00

−0.93 −0.03

0.66 0.00

−0.17 0.07

Source: EUROSTAT. ∗ Periods compared are (1991–8) and (1999–2005). ∗∗ Periods compared are (1995–8) and (1999–2005).

1.08 −3.10

0.41

1.51 −0.45 −0.46 0.00 0.00 0.02

0.18 0.55

Table 4.2(c) Evolution of the size of public expenditures (in % GDP) between the periods 1990–8 and 1999–2005 by kind of expenditure (economic classification) Kind of expenditure

Increase

Decline

Intermediate consumption

Belgium, Greece, Netherlands, Italy, Luxembourg, Portugal Belgium, Greece, France, Portugal France

Germany, Ireland, Spain, France, Finland

Compensation of employees Other taxes on production Subsidies Property income Current taxes on income, wealth, etc. Social benefits other than social transfers in kind Social transfers in kind

Finland

Capital transfers

Germany, Greece, Italy, Austria, Portugal Portugal, Finland, Netherlands, Austria, Italy, Luxembourg, France, Belgium,Germany Belgium, Germany, Ireland, Greece, Portugal, Finland, Austria, Netherlands, Italy, Luxembourg, Spain, France Austria, Italy, Ireland

Gross capital formation

Netherlands, Ireland, Greece

Acquisitions less disposals of non-financial nonproduced assets

Netherlands, Greece

Other current transfers

Germany, Ireland, Spain, Italy, Luxembourg, Netherlands, Austria, Finland Austria Belgium, Germany, Ireland, Greece, Spain, France, Netherlands, Austria, Portugal, Finland Belgium, Germany, Ireland, Greece, Spain, France, Netherlands, Austria, Portugal, Finland

Belgium, Ireland, Spain, France, Luxembourg, Netherlands, Finland Spain, Ireland

Portugal, Finland, Netherlands, Luxembourg, Spain, France, Greece, Belgium, Germany Portugal, Finland, Austria, Italy, Luxembourg, Spain, Finland, Belgium, Germany Austria, Portugal, Finland, Luxembourg, Italy, Spain, Belgium, Germany

95

96 Fiscal Policy in the European Union

30.00

Germany

25.00

Before

Netherlands Austria Belgium

Finland

20.00

Luxembourg Italy Spain

15.00 Ireland

Greece Portugal

10.00 10.00

15.00

20.00

25.00

30.00

After Figure 4.2 Size of public social expenditures before EMU (1990–8) and after EMU (1999–2005) (% GDP)

lower than the average. Out of these four countries, in Ireland and Spain, the two countries with the lower size of social expenditures, this spending declined after the EMU, whilst in Greece and Portugal it increased. Given the low size of this expenditure, we can assume that under the threshold where higher values have a negative impact, such a behaviour should have had, ceteris paribus (i.e. keeping unchanged the size of total expenditures and revenues and the fiscal balance) a negative impact on the quality of public expenditures in Ireland and Spain and a positive one in Greece and Portugal. The second group comprises four countries, the Netherlands, Finland, Luxembourg and Italy. After the EMU, social expenditures remained unchanged in Luxembourg, slightly increased in Italy and declined in the Netherlands and Finland. In this group, the conclusion about the quality of the public expenditures depends on whether the size of social expenditures is above/below the threshold for which social expenditures would negatively affect economic growth. The last group of countries is formed by Belgium, Austria and Germany. These countries are those with the highest size of social expenditures after EMU. Whilst in Belgium this spending has remained unchanged,

Fiscal Adjustment and Public Expenditures

97

in Austria and Germany, social expenditure has increased after EMU. In these countries there is a risk that the size of social expenditures is excessive, and therefore, an increase in that item of spending could have a negative impact on growth, thus worsening the quality of their public finances. Table 4.3 shows the composition of public expenditures in EMU countries according to the functional classification of public spending (with the exception of Spain due to the lack of data before 1999). The kind of expenditures refers to the COFOG classification. Table 4.4 summarizes the data shown in Table 4.3. It is clear that there is no common pattern of behaviour of public expenditures, with the only exception of general public services. This item, which is mainly formed by the public debt interest payments, declines in all the EMU countries, except in Luxembourg, an outcome that is consistent with the decline in the size of public deficits and outstanding public debt previously detected. In any case, the impact of public expenditure on economic growth and its quality depends, as we mentioned, on the size of the items that can be considered as ‘productive’ or ‘unproductive’. Based on the literature analysed in the previous section, we have opted to make the following division of public spending items into productive and unproductive ones: • Productive expenditure: defence, public order and safety, economic

affairs (includes sectoral R&D and transport and communication), environmental protection, housing and community amenities, health and education. • Unproductive expenditure: general public services (includes public debt interest payments), recreation, culture and religion, social protection. Tables 4.5(a) and 4.5(b) show the evolution of public expenditures in EMU countries using the previous classification of productive and productive expenditures according, now, to the functional classification of public spending. Assuming that the distinction made between productive and productive items is right, the only clear-cut conclusions we can reach are that the quality of public expenditures in Portugal should have improved in Greece and Belgium and worsened in Germany and Luxembourg. However, the situation is more ambiguous in the other countries, depending on the absolute size of the changes in the items. Figure 4.3 enables a better analysis of the evolution of public expenditures. As we can see, leaving aside the cases of the four countries

98 Table 4.3 Public expenditures (in % GDP) by kind of expenditure (functional classification) Country

Variable

1990–8∗ (1)

BELGIUM

Total expenditures General public services Defence Public order and safety Economic affairs Environment protection Housing and community amenities Health Recreation, culture and religion Education Social protection

GERMANY

IRELAND

GREECE

1999–2005 (2)

Difference (3 = 1 −2)

52.42 12.40 1.59 1.37 5.30 0.64 0.33

49.69 9.95 1.18 1.63 4.78 0.72 0.33

−2.74 −2.45 −0.40 0.26 −0.52 0.08 0.00

5.88 0.86

6.65 1.13

0.76 0.27

5.99 18.06

5.90 17.42

−0.09 −0.64

Total expenditures General public services Defence Public order and safety Economic affairs Environment protection Housing and community amenities Health Recreation, culture and religion Education Social protection

48.75 6.68 1.47 1.62 5.42 0.93 0.87

47.29 5.89 1.17 1.65 3.93 0.56 1.04

−1.47 −0.79 −0.30 0.03 −1.49 −0.37 0.17

6.14 0.80

6.25 0.69

0.12 −0.11

4.36 20.47

4.24 21.86

−0.12 1.40

Total expenditures General public services Defence Public order and safety Economic affairs Environment protection Housing and community amenities Health Recreation, culture and religion Education Social protection

42.42 8.19 1.05 2.75 5.26 0.50 1.47

33.37 4.06 0.62 1.47 4.82 0.54 1.60

−9.06 −4.13 −0.44 −1.28 −0.44 0.04 0.14

6.05 0.46

6.59 0.53

0.54 0.07

4.98 11.71

4.22 8.92

−0.77 −2.79

Total expenditures General public services

45.03 15.66

44.63 9.06

−0.41 −6.60 (Continued)

99 Table 4.3

(Continued)

Country

1990–8∗ (1)

1999–2005 (2)

Difference (3 = 1 −2)

3.66 0.30 3.50 0.30 0.24

3.40 1.04 5.25 0.55 0.41

−0.25 0.74 1.74 0.25 0.17

1.91 0.15

4.07 0.34

2.16 0.18

3.08 16.23

2.89 17.63

−0.19 1.40

53.85 8.16 2.46 1.23 3.51 0.61 1.55

52.64 7.42 2.02 1.28 3.13 0.74 1.72

−1.21 −0.74 −0.44 0.05 −0.39 0.13 0.18

6.63 1.10

6.96 1.30

0.33 0.20

6.52 22.07

6.33 21.73

−0.20 −0.35

Total expenditures General public services Defence Public order and safety Economic affairs Environment protection Housing and community amenities Health Recreation, culture and religion Education Social protection

52.95 13.74 1.28 2.03 4.85 0.75 1.06

47.69 9.40 1.28 1.95 3.84 0.81 0.73

−5.25 −4.34 0.00 −0.08 −1.01 0.06 −0.33

5.88 0.80

6.32 0.85

0.44 0.05

5.12 17.43

4.71 17.80

−0.41 0.37

Total expenditures General public services Defence Public order and safety Economic affairs

39.71 4.28 0.61 0.75 4.98

40.39 4.71 0.27 0.97 4.23

0.68 0.44 −0.34 0.22 −0.75

Variable

Defence Public order and safety Economic affairs Environment protection Housing and community amenities Health Recreation, culture and religion Education Social protection FRANCE

ITALY

LUXEMBOURG

Total expenditures General public services Defence Public order and safety Economic affairs Environment protection Housing and community amenities Health Recreation, culture and religion Education Social protection

(Continued)

100 Table 4.3 (Continued) Country

1990–8∗ (1)

1999–2005 (2)

Difference (3 = 1 −2)

1.26 0.96

1.13 0.83

−0.12 −0.13

4.74 1.47

4.74 1.84

0.00 0.37

4.49 16.17

4.74 16.91

0.25 0.74

Total expenditures General public services Defence Public order and safety Economic affairs Environment protection Housing and community amenities Health Recreation, culture and religion Education Social protection

52.60 9.86 1.78 1.41 4.81 0.87 2.35

45.74 8.33 1.54 1.62 4.83 0.84 1.08

−6.86 −1.53 −0.25 0.20 0.02 −0.03 −1.27

3.67 1.16

4.08 1.40

0.41 0.23

5.00 19.09

4.98 17.06

−0.03 −2.04

Total expenditures General public services Defence Public order and safety Economic affairs Environment protection Housing and community amenities Health Recreation, culture and religion Education Social protection

54.49 8.92 0.98 1.50 4.72 0.88 1.02

51.04 7.68 0.90 1.43 4.86 0.38 0.77

−3.45 −1.24 −0.08 −0.06 0.14 −0.51 −0.25

7.65 1.13

7.04 1.04

−0.61 −0.10

6.09 21.60

5.93 21.01

−0.15 −0.59

Total expenditures General public services Defence Public order and safety Economic affairs Environment protection

42.95 8.86 1.82 2.06 5.75 0.46

44.89 6.38 1.42 1.89 4.94 0.63

1.94 −2.47 −0.40 −0.17 −0.81 0.17

Variable

Environment protection Housing and community amenities Health Recreation, culture and religion Education Social protection NETHERLANDS

AUSTRIA

PORTUGAL

(Continued)

Fiscal Adjustment and Public Expenditures

101

Table 4.3 (Continued) Country

Variable

Housing and community amenities Health Recreation, culture and religion Education Social protection FINLAND

Total expenditures General public services Defence Public order and safety Economic affairs Environment protection Housing and community amenities Health Recreation, culture and religion Education Social protection

1990–8∗ (1)

1999–2005 (2)

Difference (3 = 1 −2)

0.96

0.82

−0.14

5.04 0.87

6.72 1.12

1.68 0.25

5.85 11.28

7.18 13.79

1.33 2.51

58.43 7.42 1.95 1.45 7.21 0.31 0.70

49.58 6.82 1.54 1.40 4.78 0.31 0.37

−8.85 −0.60 −0.41 −0.05 −2.44 0.00 −0.33

6.30 1.36

6.21 1.20

−0.09 −0.16

6.80 24.92

6.00 20.95

−0.80 −3.97

Source: AMECO Database. ∗ For Germany the period analysed is 1991–8; for France, Netherlands and Austria, the period is 1995–8.

mentioned above, Portugal is located at section II of the figure. In this section, both kinds of spending increase, but the increase in productive expenditure is higher than that of unproductive expenditure, thus leading to a ‘better’ composition of public expenditure. In section V, productive and unproductive expenditure declines, and therefore, the economic impact depends on the size of the decline of both kinds of expenditures. The higher the decline of unproductive spending and the lower the decline of productive expenditures, the more positive is the impact on economic growth and the higher the quality of public expenditures. That means that the closer a country is to the point (0, −7) the more the positive change in the quality of public expenditure. This suggests that for this group of countries, a better evolution of public spending would have taken place in the Netherlands, Italy and Ireland. Obviously, this conclusion must be taken with caveats, since it depends on the size of public expenditures before the adjustment.

102 Fiscal Policy in the European Union Table 4.4 Evolution of public expenditure between 1990–8 and 1999–2005 Kind of expenditure

Increase

Decline

General public services

Luxembourg

Belgium, Germany, Ireland, Greece, France, Italy, Netherlands, Austria, Portugal, Finland

Defence

Belgium, Germany, Ireland, Greece, France, Luxembourg, Netherlands, Austria, Portugal, Finland

Public order and Belgium, Germany, Greece, safety France, Luxembourg, Netherlands

Ireland, Italy, Austria, Portugal, Finland

Economic affairs Greece, Netherlands, Austria Belgium, Germany, Ireland, France, Italy, Luxembourg, Portugal, Finland Environment protection

Belgium, Ireland, Greece, France, Italy, Portugal

Germany, Luxembourg, Netherlands, Austria

Housing and community amenities

Germany, Ireland, Greece, France

Italy, Luxembourg, Netherlands, Austria, Portugal, Finland

Health

Belgium, Germany, Ireland, Greece, France, Italy, Netherlands, Portugal

Austria, Finland

Recreation, culture and religion

Belgium, Ireland, Greece, France, Italy, Luxembourg, Netherlands, Portugal

Germany, Austria, Finland

Education

Luxembourg, Portugal

Belgium, Germany, Ireland, Greece, France, Italy, Netherlands, Austria, Finland

Social protection Germany, Greece, France, Belgium, Ireland, Netherlands, Italy, Luxembourg, Portugal Austria, Finland

4 Conclusions As we explained at the beginning of the chapter, the strategy of fiscal policy in EMU has changed from a fiscal policy only based on the need to reduce fiscal deficits via cuts in public expenditures to a strategy in which a key role is played by the decline in public expenditures resulting from a recomposition of public spending towards those items considered as productive expenditures.

103 Table 4.5(a) Evolution of productive and unproductive public expenditure (% GDP) Country

Kind of expenditure

Belgium

Productive Unproductive Productive Unproductive Productive Unproductive Productive Unproductive Productive Unproductive Productive Unproductive Productive Unproductive Productive Unproductive Productive Unproductive Productive Unproductive Productive Unproductive

Germany Ireland Greece France Italy Luxembourg Netherlands Austria Portugal Finland

1990–8∗ (1)

1999–2005 (2)

Difference (3 = 2 − 1)

21.10 31.32 20.80 27.95 22.06 20.36 13.00 32.04 22.51 31.34 20.98 31.97 17.79 21.92 19.91 30.12 22.83 31.66 21.94 21.01 24.73 33.70

21.19 28.50 18.85 28.44 19.86 13.51 17.60 27.02 22.19 30.45 19.65 28.05 16.92 23.46 18.92 26.78 21.31 29.73 23.6 21.29 20.61 28.97

0.08 −2.82 −1.96 0.49 −2.20 −6.85 4.61 −5.01 −0.33 −0.89 −1.33 −3.92 −0.87 1.54 −0.99 −3.34 −1.52 −1.93 1.66 0.28 −4.12 −4.73

Source: AMECO Database. ∗ For Germany the period analysed is 1991–8; for France, Netherlands and Austria, the period is 1995–8.

Table 4.5(b) Evolution of productive and unproductive public expenditure (% GDP) Productive expenditures

Unproductive expenditure

Increase Decline

Increase

Decline

Portugal Greece, Belgium

Germany, Luxembourg Ireland, Italy, Netherlands, Austria, Finland, France

Unproductive

104 Fiscal Policy in the European Union

7 6 5 4 3 2 1 0 ⫺1 ⫺2 ⫺3 ⫺4 ⫺5 ⫺6 ⫺7

I VIII II

Luxembourg VII Portugal

Germany France VI

III

Belgium

Austria

Netherlands Finland

Italy Greece

V

Ireland

IV

⫺7 ⫺6 ⫺5 ⫺4 ⫺3 ⫺2 ⫺1

0

1

2

3

4

5

6

7

Productive Figure 4.3 Evolution of productive and unproductive public expenditures according to the functional classification (% GDP)

Our study has shown that the first of these premises is fulfilled. After the creation of the EMU, all the countries have reduced the size of their fiscal deficits, and in ten out of the twelve countries public expenditures have declined (the exceptions being Portugal and Luxembourg). The situation, however, is not the same when the composition of public expenditures is analysed. In terms of the economic classification of public spending, in all the countries, with the single exception of Ireland, public investment has declined, and the situation of spending on social expenditures is unclear. In the case of the functional classification, a generalized move to a higher quality of public spending, that is, a higher size of productive expenditures, is not detected. Only in three countries (Belgium, Greece and Portugal) is a rise in the size of productive expenditures detected, whilst in eight countries this kind of expenditure declines. Given the evolution of unproductive expenditures, the impact on long-term economic growth of this behaviour of public expenditures depends dramatically on the assumption made about whether the initial size of these expenditures can be considered ‘excessive’ – a question which it is difficult to

Fiscal Adjustment and Public Expenditures

105

test empirically. In any case, the only clear-cut conclusion that can be reached is that the desired change to a higher quality composition of public expenditure in EMU member states has not taken place.

Notes 1. Most studies argue that the expansionary effect is generated on a horizon below that of two years after the fiscal adjustment. 2. The view of the European Union institutions and the ECB about the role to be played in the euro zone by fiscal policy is reflected in a number of documents published by those institutions. In the case of the European Commission, see the annual reports ‘Public Finances in EMU’ published by the DirectorateGeneral for Economic and Financial Affairs since 2000. In the case of the ECB, see European Central Bank (2001, 2004a, 2004b, 2006). 3. Article 1.3 Council Regulation (EC) No. 1467/97 of 7 July 1997, as amended by Council Regulation (EC) No. 1056/05 of 27 June 2005 (italics added).

References Aarle, B. van and H. Garretsen, ‘Keynesian, Non-Keynesian or No Effects of Fiscal Policy Changes? The EMU Case’, Journal of Macroeconomics, 25(2) (2003): 213–40. Afonso, A., ‘Non-Keynesian Effects of Fiscal Policy in the EU-15’, Working Paper, Department of Economics, Institute for Economics and Business Administration (ISEG), Technical University of Lisbon, WP 2001/07 (2001). Afonso, A., ‘Expansionary Fiscal Consolidations in Europe’, European Central Bank, Working Paper Series, No. 675 (2006). Afonso, A., W. Ebert, L. Schuknecht and M. Thöne, ‘Quality of Public Finances and Growth’, European Central Bank, Working Paper Series, No. 438 (2005). Aghion, P. and P. Howitt, ‘A Model of Growth through Creative Destruction’, Econometrica, 60 (1992): 323–51. Alesina, A. and R. Perotti, ‘Fiscal Expansions and Adjustments in OECD Countries’, Economic Policy, 21 (1995): 205–40. Alesina, A. and R. Perotti, ‘Fiscal Adjustments in OECD Countries: Composition and Macroeconomic Effects’, IMF Staff Papers, 44, June (1997): 297–329. Alesina, A., S. Ardagna, R. Perotti and F. Schiantarelli, ‘Fiscal Policy, Profits and Investment’, American Economic Review, 92 (2002): 571–89. Alesina, A., R. Perotti and J. Tavares, ‘The Political Economy of Fiscal Adjustments’, Brooking Papers on Economic Activity, Spring (1998): 197–248. Angelopoulos, K., G. Economides and P. Kamman, ‘Tax-spending Policies and Economic Growth: Theoretical Predictions and Evidence from the OECD’, European Journal of Political Economy, 2006, doi: 10.1016/j.ejpoleco. Arestis, P. and M. Sawyer, Re-examining Monetary and Fiscal Policy for the 21st Century (Cheltenham: Edward Elgar, 2004). Aschauer, D., ‘Is Government Spending Productive?’, Journal of Monetary Economics, 23 (1989): 177–200.

106 Fiscal Policy in the European Union Atkinson, P. and P. van den Noord, ‘Managing Public Expenditure: Some Emerging Policy Issues and a Framework for Analysis’, OECD Economics Department, Working Paper, 285 (2001). Baier, S. and G. Glomm, ‘Long-run Growth and Welfare Effects of Public Policies with Distortionary Taxation’, Journal of Economic Dynamics and Control, 25 (2001): 2007–42. Barro, R. J., ‘Government Spending in a Simple Model of Endogenous Growth’, Journal of Political Economy, 98(1) (1990): 103–17. Barro, R. J., ‘Economic Growth in a Cross-section of Countries’, Quarterly Journal of Economics, 106 (1991): 407–44. Barro, R. J. and X. Sala-i-Martin, ‘Public Finance in Models of Economic Growth’, Review of Economic Studies, 59(4) (1992): 645–61. Barro, R. J. and X. Sala-i-Martin, Economic Growth (New York: McGraw Hill, 1995). Bleaney, M., N. Gemmell and R. Kneller, ‘Testing the Endogenous Growth Model: Public Expenditure, Taxation and Growth over the Long Run’, Canadian Journal of Economics, 34(1) (2001): 36–57. Bloom, D., D. Canning and J. Sevilla, ‘The Effect of Health on Economic Growth: Theory and Evidence’, NBER Working Paper Series, 8587 (2001). Briotti, M. G., ‘Fiscal Adjustment between 1991 and 2002: Stylised Facts and Policy Implications’, European Central Bank, Occasional Paper Series, No. 9 (2004). Briotti, M. G., ‘Economic Reactions to Public Finance Consolidation: a Survey of the Literature’, European Central Bank, Occasional Paper Series, No. 38 (2005). Cameron, D. R., ‘The Expansion of the Public Economy: a Comparative Analysis’, American Political Science Review, 72 (1978): 1243–61. Chamley, C., ‘Optimal Taxation of Capital Income Equilibrium with Infinite Lives’, Econometrica, 54(3) (1986): 607–22. Council of the European Union, The Contribution of Public Finances to Growth and Employment: Improving Quality and Sustainability, Report from the Commission and the (ECOFIN) Council to the European Council (Stockholm, 23/24 2001), 6997/01, Brussels, 12 March 2001. Devarajan, S., V. Swaroop and H. F. Zou, ‘The Composition of Public Expenditure and Economic Growth’, Journal of Monetary Economics, 37 (1996): 313–44. Easterly, W. and S. Rebelo, ‘Fiscal Policy and Economic Growth: an Empirical Investigation’, Journal of Monetary Economics, 32 (1993): 417–58. European Central Bank, ‘Fiscal Policies and Economic Growth’, ECB Monthly Bulletin, January (2001): 39–53. European Central Bank, ‘EMU and the Conduct of Fiscal Policies’, ECB Monthly Bulletin, January (2004a): 41–50. European Central Bank, ‘Fiscal Policy Influences on Macroeconomic Stability and Prices’, ECB Monthly Bulletin, April (2004b): 45–57. European Central Bank, ‘The Importance of Public Expenditure Reform for Economic Growth and Stability’, ECB Monthly Bulletin, April (2006): 61–73. European Commission, Directorate-General for Economic and Financial Affairs, ‘Public Finances in EMU 2000’, European Economy, No. 3/2000 (2000). European Commission, Directorate-General for Economic and Financial Affairs, ‘Public Finances in EMU 2001’, European Economy, No. 3/2001 (2001). European Commission, Directorate-General for Economic and Financial Affairs, ‘Public Finances in EMU 2002’, European Economy, No. 3/2002 (2002).

Fiscal Adjustment and Public Expenditures

107

European Commission, Directorate-General for Economic and Financial Affairs, ‘Public Finances in EMU 2003’, European Economy, No. 3/2003 (2003). European Commission, Directorate-General for Economic and Financial Affairs, ‘Public Finances in EMU 2004’, European Economy, No. 3/2004 (2004). European Commission, ‘The Broad Economic Policy Guidelines (for the 2005–8 Period)’, European Economy, No. 4/2005 (2005a). European Commission, Directorate-General for Economic and Financial Affairs, ‘Public Finances in EMU 2005’, European Economy, No. 3/2005 (2005b). European Commission, Directorate-General for Economic and Financial Affairs, ‘Public Finances in EMU 2006’, European Economy, No. 3/2006 (2006). European Commission, Directorate-General for Economic and Financial Affairs, ‘Public Finances in EMU 2007’, European Economy, No. 3/2007 (2007). Florio, M., ‘On Cross-country Comparability of Government Statistics: Public Expenditure Trends in OECD National Accounts’, International Review of Applied Economics, 15(2) (2001): 181–98. Fölster, S. and M. Henrekson, ‘Growth and the Public Sector: a Critic of the Critics’, European Journal of Political Economy, 15 (1998): 3337–58. Gemmel, N. and R. Kneller, ‘The Impact of Fiscal Policy on Long-run Growth’, European Economy: Reports and Studies, 1 (2001): 97–129. Giavazzi, F. and M. Pagano, ‘Can Severe Fiscal Contractions be Expansionary? Tales of Two Small European Countries’, NBER Macroeconomics Annual, 5 (1990): 75–111. Giavazzi, F., T. Japelli and M. Pagano, ‘Searching for Non-Keynesian Effects of Fiscal Policy’, Working Paper, Centro Studi in Economia e Finanza, No. 16 (1999). Giavazzi, F., T. Japelli and M. Pagano, ‘Searching for the Non-linear Effects of Fiscal Policy: Evidence for Industrial and Developing Countries’, NBER Working Paper Series, 7460 (2000). Giudice, G., A. Turrini and J. in’t Veld, ‘Can Fiscal Consolidation be Expansionary in the EU? Ex-post Evidence and Ex-ante Analysis’, Economic Papers, No. 195, European Commission, Directorate-General for Economic and Financial Affairs (2003). Giudice, G., A. Turrini and J. in’t Veld, ‘Non-Keynesian Fiscal Adjustments? A Close Look at Expansionary Fiscal Consolidations in the EU’, Open Economics Review, 18(5) (2007): 613–30. Grossman, G. M. and E. Helpman, Innovation and Growth in the Global Economy (Cambridge, MA: MIT Press, 1991). Gupta, S., B. Clements, E. Baldacci and C. Mulas-Granados, ‘Fiscal Policy, Expenditure Composition and Growth in Low-income Countries’, Journal of International Money and Finance, 24 (2005): 441–63. Gwartney, J., R. Holcombe and R. Lawson, ‘The Scope of Government and the Wealth of Nations’, Cato Journal, 18(2) (1998): 163–90. Hemming, R., M. Kell and S. Mahfouz, ‘The Effectiveness of Fiscal Policy in Stimulating Economic Activity: a Review of the Literature’, IMF, Working Paper, WP/02/208 (2002). Jones, C. I., ‘R&D Models of Economic Growth’, Journal of Political Economy, 103 (1995): 759–84. Jones, L. E., R. E. Manuelli and P. E. Rossi, ‘On the Optimal Taxation of Capital Income’, NBER Working Paper, No. 4525 (1993).

108 Fiscal Policy in the European Union King, R. G. and S. Rebelo, ‘Public Policy and Economic Growth: Developing Neoclassical Implications’, NBER Working Paper, No. 3338 (1990). Kneller, R., M. F. Bleaney and N. Gemmell, ‘Fiscal Policy and Growth: Evidence from OECD Countries’, Journal of Public Economics, 74 (1999): 171–90. Kneller, R., M. F. Bleaney and N. Gemmell, ‘Testing the Endogenous Growth Model: Public Expenditure, Taxation and Growth over the Long Run’, Canadian Journal of Economics, 34 (2001): 36–57. Kocherlatoky, N. and K. M. Yi, ‘Is there Endogenous Long-term Growth? Evidence from the US and the UK’, Journal of Money, Credit and Banking, May (1997): 235–62. Kumar, M. S., D. Leigh and A. Plekhanov, ‘Fiscal Adjustments: Determinants and Macroeconomic Consequences’, IMF Working Paper, WP/07/178 (2007). Lamo, A. and R. Strauch, ‘The Contribution of Public Finances to the European Growth Strategy’, in Banca d’Italia, The Impact of Fiscal Policy (Rome: Banca d’Italia, 2002), 479–519. Lucas, R. E., ‘On the Mechanics of Economic Development’, Journal of Monetary Economics, 22 (1988): 3–42. Lucas, R. E., ‘Why Doesn’t Capital Flow from Rich to Poor Countries?’, American Economic Review, Papers and Proceedings, 80 (1990): 92–6. McDermott, C. J. and R. F. Wescott, ‘An Empirical Analysis of Fiscal Adjustments’, IMF Staff Papers, 43, December (1996): 725–53. Mendoza, E., G. Milessi-Ferreti and P. Asea, ‘On the Effectiveness of Fiscal Policy in Altering Long-run Growth: Harberger’s Superneutrality Conjecture’, Journal of Public Economics, 66(1) (1997): 99–126. Nourzad, F. and M. D. Vrieze, ‘Public Capital Formation and Productivity Growth: Some International Evidence’, Journal of Productivity Analysis, 6 (1995): 283–95. Rebelo, S., ‘Long-run Policy Analysis and Long-run Growth’, Journal of Political Economy, 99 (1991): 500–21. Romer, P. M., ‘Increasing Returns and Long-run Growth’, Journal of Political Economy, 94 (1986): 1002–37. Romer, P. M., ‘Endogenous Technical Change’, Journal of Political Economy, 98 (1990): S71–S102. Romero de Avila, D. and R. Strauch, ‘Public Finances and Long-term Growth in Europe: Evidence from a Panel Data Analysis’, ECB Working Paper, 246 (2003). Sanchez-Robles, B., ‘Infrastructure Investment and Growth: Some Empirical Evidence’, Contemporary Economic Policy, 16 (1998): 98–108. Thöne, M., ‘Public Expenditure for Growth and Sustainable Development (PEGS): Conceptual, Empirical and Quantitative Issues’, summary and main conclusions of a study commissioned by the German Federal Ministry of Finance, Cologne Centre for Public Finance (2003). Zagler, M. and G. Dürnecker, ‘Fiscal Policy and Economic Growth’, Journal of Economic Surveys, 17(3) (2003): 397–418.

5 Public Capital and Economic Growth: a Spurious Empirical Link? Gwenaëlle Poilon

1 Introduction Assessing the factors underlying long-run economic growth is one of the most important issues of applied macroeconomics. And whereas the role of fiscal policy is still in debate among economists, there is general agreement about the key role played by one of its spending items, namely public investment. The essential role played by public investment is recognized officially. Indeed, the view that public investment should be fostered has been officially accepted by European governments since they signed the ‘Lisbon Agenda’ document in 2000. The European Union has set for itself the ambitious goal of becoming within a decade the most competitive and dynamic knowledge-based economy in the world. To achieve this goal, the Presidency conclusions of the Lisbon Council outlined a strategy designed in particular to promote the development, adoption and use of new technologies through increased and more efficient investment in infrastructures. There has been general agreement about this policy among economists and among politicians. Public investment spending seems to have a key role in the Lisbon Strategy because it is seen as an essential growthdriving force. Thus, this item of fiscal spending reconciles economists with politicians and even economists seem to find common ground on this issue. But, why should public investment be protected? First, public investment is a part of the demand side of the economy, affecting short-run economic growth. For instance, we understand that a restrictive policymix in Europe (responsible for an inadequate aggregate demand and so

109

110 Fiscal Policy in the European Union

economic growth below its potential) explains to a large extent the rising unemployment. Furthermore, public investment is a part of the supply side of the economy since it is an essential input in the production process. Lower public investment could restrict the long-term potential of the economy and so lead to higher unemployment. Consequently, winding back public investment can entail a vicious circle which affects both short and long runs. Yet, in most European countries, government investment (in terms of GDP) decreased substantially during the consolidation process in the 1980s and 1990s because cutting public investment spending can be achieved without much resistance from society as there are no strong pressure groups associated with public capital spending (see Tables 5.1 and 5.2). According to Oxley and Martin (1991: 161), this scheme reflects ‘the political reality that it is easier to cut-back or postpone investment spending than it is to cut current expenditures’.

Table 5.1 Government net capital as a percentage of GDP Countries

1970

1980

1990

2000

Austria Belgium Germany France Italy Netherlands

63.89 25.05 54.10 53.54 46.83 80.00

71.16 37.72 59.96 55.03 44.67 80.59

65.98 42.82 53.12 53.02 48.96 68.87

53.03 35.56 47.12 53.48 47.71 56.24

Source: Kamps (2004b) and own calculations.

Table 5.2 Government net capital as a percentage of private non-residential net capital Countries

1970

1980

1990

2000

Austria Belgium Germany France Italy Netherlands

59.23 20.78 36.47 49.27 35.48 57.97

58.45 33.04 47.01 47.55 33.62 57.02

53.02 38.81 39.47 45.95 38.93 52.27

41.95 30.47 36.55 47.91 39.06 47.81

Source: Kamps (2004b) and own calculations.

Public Capital and Economic Growth

111

In this sense, the application of a ‘golden rule of public finance’ (as proposed by Fitoussi and Creel, 2002; Le Cacheux, 2002; Blanchard and Giavazzi, 2003) instead of the actual fiscal rules could resolve this problem. Indeed, such a rule states that, over the cycle, government borrowing should not exceed net government capital formation; consequently current expenditures should be financed by current receipts. Thus, reducing public investment spending could no longer serve as an adjustment variable in respect of the 3 per cent ceiling. Thus, adopting the golden rule would be a means to boost economic growth by promoting public investment. These comments suggest that the importance of public investment is recognized both from a theoretical point of view and also by economists and politicians. So, we might wonder whether this intuition is wellfounded empirically. Indeed, whereas human capital and education, for instance, are theoretically considered as key growth factors, very few articles have found a positive robust relation between growth and human capital when econometric issues are properly taken into account (Gurgand, 2000). We can ask whether the same phenomenon is observed with the public capital issue. The purpose of this chapter is, first, to review the empirical research on the impact of public capital spending on economic growth. The methodological issues and the treatment of econometric problems surrounding public capital estimations are reviewed. We further test on some European countries the specification proposed by Aschauer’s (1989b) seminal article in order to highlight these empirical caveats. Then we empirically evaluate the impact of public capital on economic growth in Europe, making use of dynamic panel estimation techniques which control for the several econometric estimation problems. The chapter is organized as follows. Section 2 presents the literature dealing with the relationship between public capital and growth in a production function approach framework (time series and panel studies). Section 3 contains the empirical outcomes for some European countries (France, Germany and Italy) when the procedure followed by Aschauer is applied to these European countries. Section 4 is dedicated to empirical estimations on a European panel when econometric biases are controlled for. Section 5 concludes.

2 A review of the literature In spite of the theoretical consensus concerning the productive role of public capital, empirical results obtained in the literature are still

112 Fiscal Policy in the European Union

contradictory. This empirical controversy has taken up much room in the economic literature since the 1990s. It is undoubtedly the article by Aschauer (1989a) which explains the wave of literature devoted to this question. Historically, Ratner (1983) was the first one to estimate with chronological series (from 1949 to 1973) a Cobb-Douglas production function augmented by public capital for the United States. He found an elasticity of 5.6 per cent for public capital with a constant return production function. By extending the sample by twelve years and by adding the oil price as an explanatory variable, Ram and Ramsey (1989) obtained a much higher elasticity for public capital (24 per cent). Nevertheless, it was only when Aschauer published his second article in 1989 that the debate on public capital was relaunched. In this article, Aschauer used American data from 1949 to 1985 and tested different specifications. He obtained a public capital elasticity figure of between 39 per cent and 56 per cent. Thus, Aschauer found a high productive contribution for public capital spending; and in some cases public capital obtained a higher return than private capital. According to Aschauer, the decrease in public capital spending in the US in the 1970s and 1980s can explain the productivity slow-down observed in the same period. Several contributions have subsequently confirmed these first results (see Table 5.3). Munnell (1990) found an estimated elasticity for public capital of between 31 per cent and 39 per cent depending on the scale returns. Similar results have been found by Eisner (1991). Sturm and de Haan (1995) also proposed new estimates for Aschauer’s equations and found an elasticity for public capital around 41 per cent. Vijverberg et al. (1997) estimated a production function integrating the stock of intermediary goods and the stock of federal and local capital. Like Ram and Ramsey (1989), they found a high contribution of state public capital (48 per cent) but no significant contribution of federal capital. Table 5.3 shows that a lot of estimations have been performed to test the impact of public capital on growth in particular with American data. But some empirical studies for other OECD countries were also carried out. For instance, Berndt and Hansson (1992) found an elasticity of 69 per cent for public capital in Sweden and Wylie (1996), using Canadian data, obtained results similar to the American ones. But, in general, these results have appeared to be too excessive and have given rise to a lot of methodological criticism. Indeed, Tatom (1991) calculated that public capital marginal productivity would be around 60 per cent and 80 per cent in the article by Aschauer (1989a), that

Public Capital and Economic Growth

113

Table 5.3 Studies based on a Cobb-Douglas production function in level Studies

Data

Public capital elasticity

Ratner (1983) Aschauer (1989a) Ram and Ramsey (1989) Munnell (1990) Aaron (1990) Berndt and Hansson (1991) Finn (1993) Bajo-Rubio and Sosvilla-Rivero (1993) Eisner (1991) Sturm and de Haan (1995) Ai and Cassou (1995) Otto and Voss (1996) Toen-Gout and Jongeling (1994) Dalamagas (1995) Wylie (1996) Lau and Sin (1997) Vijverberg et al. (1997) Balmaseda (1997) Kavanagh (1997)

United States (49–73) United States (49–85) United States (49–85) United States (49–87) United States (52–85) Sweden (60–88) United States (50–89) Spain (64–88)

0.06 0.39 0.24 0.31–0.39 not robust 0.68 0.16 0.19

United States (61–91) United States (49–85) United States (47–89) Australia (66–90) Netherlands (not known) Greece (50–92) Canada (46–91) United States (25–89) United States (58–89) United States (49–83) Ireland (58–90)

0.27 0.41 0.15–0.20 0.38–0.45 0.37 0.53 0.11–0.52 0.11 0.48 0.35 no significance

is to say twice as high as the private capital one. The very high level of estimations may suggest a stationarity bias in the series. Indeed, the first regressions were estimated without testing the stationarity of the series. This can lead to spurious regressions if these tests are not satisfied (Aaron, 1990; Tatom, 1991). Sturm and de Haan (1995) tested the same series as Aschauer and found that the series were neither stationary nor cointegrated. Consequently, first differences must be tested and in this case, Sturm and de Haan have shown that public capital is no longer significant. This absence of robustness with differentiated data has been confirmed by Tatom (1991) and Hulten and Schwab (1991). Only the study by Ford and Poret (1991) has given positive results for public capital in the United States and Germany (no significant results have been obtained for France and the United Kingdom). But their results have been quite implausible and have presented high levels of instability (Table 5.4). Furthermore, taking first differences raises other difficulties too. Indeed, using first differences implies that economic growth one year is only correlated with the growth of production factors in the same

114 Fiscal Policy in the European Union Table 5.4 Studies with first differences Studies

Data

α

Tatom (1991) Hulten and Schwab (1991) Sturm and de Haan (1995) Sturm and de Haan (1995) Ford and Poret (1991) Ford and Poret (1991) Ford and Poret (1991) Ford and Poret (1991)

United States (49–89) United States (49–85) United States (49–85) Netherlands (60–90) United States (57–89) France (67–89) United Kingdom (73–88) Germany (62–89)

no significance no significance no significance 1.16 0.40 no significance no significance 0.81

year; it destroys all possibility of estimating the underlying long-term relation between production and its factors. Moreover, examining the stationarity issue is not enough; one must also check whether the variables are growing together and are converging to their long-term relation, that is to say whether they are cointegrated. Several authors have studied the cointegration in a production function extended to public capital. Whereas Sturm and de Haan (1995), Tatom (1991) and Crowder and Himarios (1997) concluded that public capital and private capital were not cointegrated in the United States and in the Netherlands, Argimon et al. (1993), Bajo-Rubio and Sosvilla-Rivero (1993) and Flores de Frutos et al. (1998) found that these variables were cointegrated in Spain. Coe and Moghadam (1993) reached the same conclusion for France, as did Lau and Sin (1997) for the United States and Otto and Voss (1996) for Australia. Yet, regarding a production function as a long-term relation is not so obvious. Some authors, like Crowder and Himarios (1997), are convinced that the production function cannot represent a cointegration relation. The production function represents a cointegration relation only if the regression is well-specified and if all explaining variables are integrated, such as human capital or education, R&D spending (Nadiri and Mannumeas, 1994) and the degree of trade openness (Coe and Moghadam, 1993). Thus, omitting one or several factors can lead to biased results. For instance, according to Tatom (1991), Aschauer’s (1989a) specification (and that of 1989b) is misspecifying, with the result that the estimated elasticities are too high. In particular, Tatom (1991) refers to the negative effect of oil price on private productivity in the 1970s: for this reason he argues that the oil price variable should be included in the regression. Finally, another issue is the causality bias: is it the decrease in public investment which slows down growth or is it the opposite? (Eisner, 1991;

Public Capital and Economic Growth

115

Gramlich, 1994). According to the Wagner law, it is economic development which explains the rise in public spending. Thus, governments would spend more in public investment during economic upswings and public investment would be a superior good. So at the heart of the causality bias is the issue of public capital endogeneity. This question has been studied by Finn (1993) and by Ai and Cassou (1995) by using the Generalized Moments Method (GMM). This method is based on an instrumental variable procedure and avoids the inverse causality bias. With this technique, they have also found significant results. All these limitations explain why the empirical findings on the productive role of public capital are not clear-cut and must be considered with caution. The same drawbacks can be found when we use panel data but the one advantage of this method is that it allows us to combine information provided by chronological data with that given by international or regional variance of data. The possibility of introducing specific effects (and different specific effects) brings an additional precision to estimates. However, the results may be strongly dependent on the specification form (time and/or individual effects). As far as multinational panels1 are concerned, there are only a few articles dealing with the productive role of public capital in comparison with the specifications based on one country. Aschauer (1989b) found for a sample of G7 countries an output elasticity for public capital of 0.41. Seitz (2001) reported for a sample of thirteen OECD countries estimates between 0.12 and 0.17. But above all, the study by Evans and Karras (1994) must be quoted because their conclusions are representative of the criticisms often addressed to the panel approach.2 They estimated three alternative regressions (with data in first differences) for a panel of seven OECD countries from 1963 to 1988. They insisted on the absence of robustness stemming from the direct estimation of a production function. Public capital is notably significant only when no specific effect is included. The high instability of estimations in Evans and Karras’ study illustrated a strong correlation between these effects and the explaining variables (Hausman, 1978). Countries with a high rate of economic growth are likely to have more public and private capital. So to conclude this section, it can be noted that most studies have found a positive impact for public capital but from an econometric point of view they are also often inconsistent. The next section presents some new estimates for three European countries (France, Germany and Italy). The aim is to examine whether Aschauer’s approach is justifiable from an econometric point view when some biases are corrected. We

116 Fiscal Policy in the European Union

conclude that even when this model is estimated in first differences, the relationship between public capital and productivity is still ambiguous.

3 Aschauer’s specification for some European countries The general aim of this section is to test the impact of public capital on productivity in three different European countries, France, Germany and Italy. For this, we follow the specification used by Aschauer (1989a): the same choice of variables and the same method (OLS). But while safeguarding the spirit of Aschauer’s model, we have decided to correct the econometric biases which could make the estimates inconsistent. In Aschauer’s article, an aggregate Cobb-Douglas production function is used: β γ

Qt = At Kgtα Kpt Lt

(3.1)

where Qt is real aggregate output of the private sector, At is multifactor productivity, Kgt is public capital, Lt is aggregate hours worked by the labour force of the private sector, and Kpt is the aggregate non-military stock of private capital. Dividing both sides of equation (3.1) by Kpt , taking logarithms and assuming constant returns to scale across all inputs, Aschauer has obtained:       Qt Kgt Lt ln = ln (At ) + γ ln +α (3.2) Kpt Kpt Kpt Ln(At ) is proxied by a time trend and a constant. Aschauer also added the capacity utilization rate to control for the business cycle influence. Then the equation to be estimated is:  ln

Qt Kpt



 = c + τt + γ ln

Lt Kpt



 + α ln

Kgt Kpt

 + δ ln (CURt )

(3.3)

where t and CUR are respectively the trend and the capacity utilization rate of the private sector. We will apply this specification to three different European countries, France, Germany and Italy, using annual data over the period 1965– 2002.3 Data on the private production, the hours worked in the private sector and the utilization capacity rate come from the OECD database ‘Economic Outlook’; but capital data stocks have been calculated by Kamps (2004b). These series have been estimated according to the perpetual inventory method from the OECD series of private and public

Public Capital and Economic Growth

117

investments. One of the main advantages of these data is the international comparability of the definition used for the public sector. This data set is to our knowledge the only one which provides long-term series for European countries. A common econometric problem with time series is the presence of unit roots; that is why, before estimating any regression, a stationarity test has to be performed. This test has not been carried out by Aschauer and it can explain why his results are so implausible. In fact, when the stationarity issue is tested on Aschauer’s data, the result is that the variables used by Aschauer are I(1). Moreover, it is also shown that the hypothesis that the series are cointegrated must be rejected (Sturm and de Haan, 1995). Thus, to make consistent estimations, Aschauer should have tested his equation in first differences to turn time series into stationary ones. To avoid this criticism, we have carried out a unit root test (the Augmented Dickey-Fuller test) for our three countries. Table 5.5 shows that series are not stationary in levels (except for the variable CU in France and in Germany), but stationary in first differences. So, equation (3.3) must be re-estimated: either first-differencing is imposed if no long-term relation is present; or an Error Correction Model (ECM) has to be estimated if one or several cointegration relations are found. That is why we examine the cointegration issue with the Johansen procedure (Table 5.6). It appears that France and Germany have no

Table 5.5 Augmented Dickey-Fuller stationarity test Series

France

Germany

Italy

Lags

Statistic

Lags

Statistic

Lags

Statistic

ln(Q/Kp) ln(Kg/Kp) ln(L/Kp) ln(CUR)

1 2 1 3

−2.58 −2.03 −2.49 −2.96∗

2 2 0 1

−1.03 −2.38 −1.53 −3.95∗∗

0 1 2 0

−2.44 −1.59 −2.43 −2.21

ln(Q/Kp) ln(Kg/Kp) ln(L/Kp) ln(CUR)

0 1 1

−4.21∗∗ −3.41∗ −3.73∗∗

1 1 0

−4.80∗∗ −4.06∗∗ −4.79∗∗

0 0 0 0

−5.42∗ −3.03∗ −2.97∗ −4.04∗∗

Notes: The tests are conducted with a constant.1 ∗∗ significant at the 1% level; the critical value is −3.62 when a constant is included. ∗ significant at the 5% level; the critical value is −2.94 when a constant is included. The number of lags is chosen so as to minimize Akaike’s Information criterion. 1 The general conclusion is the same when a trend is added in the stationarity test.

118 Fiscal Policy in the European Union Table 5.6 Johansen test results H0 : r = 0

Trace H0 : r = 1

Statistic H0 : r = 2

H0 : r = 3

Cointegration rank

France Germany Italy

45.20 46.30 70.16

24.27 24.09 45.57

11.48 15.10 24.23

0.15 3.41 6.98

0 0 4

Critical values

47.86

29.80

15.49

3.84

Countries

Table 5.7 Estimation results using first-differences for France, Germany and Italy Dep. variable : ln(Q/K) Constant ln(Kg/Kp) ln(L/Kp) ln(CUR) R2

France

Germany

Italy

0.06 (6.23) (−0.05) (−0.17) 1.24 (8.46) −0.11 (−1.07) 0.74

0.01 (0.71) 0.13 (0.10) 0.42 (3.83) −0.37 (−2.42) 0.50

0.07 (3.08) 0.30 (0.86) 1.07 (3.74) −0.20 (−1.06) 0.49

Note: The regression includes also a time trend but the coefficient obtained is always close to zero; that is why its values are not reported.

long-term relation. Consequently, since series are neither stationary in levels nor cointegrated, equation (3.3) has to be first-differenced. For Italy, the results are more puzzling. Indeed, the Johansen test indicates the presence of four cointegration relations. This would imply that the variables in the Italian model are stationary; yet the unit root test shows clearly that that is not the case. Consequently, the first problem is that two tests give two different results. We have decided to override the cointegration test and to estimate Italy with first-differences too in order to make comparisons with the two other countries. But this shortcoming has to be kept in mind and it will be mentioned in the concluding paragraph of this section. This ambiguous result for Italy may be a sign of mis-specification; this problem could also affect our OLS estimation. Therefore, Table 5.7 reports the estimations for equation (3.3) in first differences. Contrary to what Aschauer obtained, the ratio of public capital to private capital is never significant. However, the validity of this result is

Public Capital and Economic Growth

119

put into question when we look at the coefficient of labour. Indeed, the elasticities for γ are superior to 1 for France and Italy. So it implies a negative elasticity for private capital, which is not realistic. For Germany, we obtain a 0.42 elasticity for γ, which is also too high for a growth rate estimation. Thus, even with first differences, the specification used by Aschauer still presents implausible results. As mentioned in the first section, a misspecification can induce biased estimates. Yet another frequent problem with macroeconomic data is the multicolinearity across variables. Here, the correlation matrix indicates a correlation of 0.75 and 0.76 between (L/Kp) and (Q/Kp) respectively for France and Germany; a 0.85 correlation is obtained for Italy between CUR and (Q/Kp). These results confirm a multicolinearity bias across variables. To be corrected, other variables must be chosen or the variables must be transformed. However, given that our goal in this section was to transpose Aschauer’s specification on European data, we cannot replace the variables, even if it turns out to be necessary. Moreover, as a general rule, the OLS method is not adapted to examine the causality between variables. For instance, it may be that it is not the public capital which explains productivity, but on the contrary that productivity explains public capital. This endogeneity bias can be corrected with an instrumental variable method (like the GMM). Once again, in order to follow Aschauer, we have used the same estimation method and so using instrumental variables has not been tested. Finally, it has to be noted that econometric tests lose power and so reliability when they are used on a restricted number of observations (38 in our case). That is why using panel data can be very useful, even if this method can present biases which must be controlled for. In the next section, estimations on European panel data are proposed to test the impact of public capital on economic activity.4 In this empirical application, all the caveats brought up in the two last sections will be considered.

4 The impact of public capital on economic growth: new robust estimations on European panel data In this section we present the estimates concerning the impact of public capital on economic growth. A panel of six EMU countries (Austria, Belgium, Germany, France, Italy and the Netherlands) has been considered. The data come from the same source as the ones of the previous section. In order to modify

120 Fiscal Policy in the European Union

the specification chosen by Aschauer, which does not seem the more relevant, we have decided to test the impact of public capital on growth, rather than on private productivity, by using a standard growth accounting framework. Taking logarithms, the production function is written as: log (yit ) = log (ait ) + α log (kgit ) + β log (kpit ) + ηt + μi + εit

(4.1)

where yit is output per worker in country i at time t, ait denotes an index of economy-wide productivity, kgit the stock of public capital per worker and kpit the stock of private capital per worker. ηt is the time-specific term while μi is the state-specific term; εit is the transitory error term. Time dummies are included to control for temporal shocks that may be responsible for economic slow-downs that are common to most OECD countries. Country dummies pick up permanent cross-country differences that reflect differences in omitted variables (social and institutional variables for instance). Using per worker variables allows us to reduce the size effect which could bias the estimations. Furthermore, the choice of this country sample is justified by the fact that unobserved panel heterogeneity is curbed at the source by the selection of a homogeneous group of countries in our sample. Panel data at five-year intervals starting in 1970 and ending in 2000 are used. Time spans of just one year are technically feasible but short-term periods are more likely to be influenced by measurement error and business cycle shocks and thus may not be appropriate for the study of long-run growth.5 Before specifying the relevant regression, the stationarity of variables has to be controlled because this issue is essential when time series are used. For panel data, statistical distributions of the null hypothesis test were proposed by Levin et al. (2002). The tests carried out on our sample indicate that series are not stationary.6 If we assume that there is no cointegration relation, regression has to be estimated in first-differences. Such a hypothesis has been made by Tatom (1991), Hulten and Schwab (1991) and Sturm and de Haan (1995). This point is a strong hypothesis but it is tricky to examine the cointegration issue with panel data. Indeed, Table 5.8 shows the national divergences about the number of cointegration relations and so it is difficult to know whether or not, once these countries are stacked together in a panel, the production function represents a long-term relation. So, we have decided to estimate equation (4.1) in first differences (series being I(1)), taking first differences in order to sweep out unobserved individual country effects that are a source of inconsistency in the estimates.

Public Capital and Economic Growth

121

Table 5.8 Johansen test results Countries

H0 : r = 0

Trace H0 : r = 1

Statistic H0 : r = 2

Cointegration rank

Germany Belgium Austria France Italy Netherlands

33.72 28.73 29.77 28.19 44.87 35.69

10.98 9.42 12.48 10.15 15.77 14.36

1.61 1.54 1.23 1.65 2.18 2.22

1 0 0 0 2 1

Critical values

29.80

15.49

3.84

So, equation (4.1) in first differences gives7 : yit = ait + α kgit + β kpit + ηt + μi + εit

(4.2)

The various specifications of this basic equation (4.2) involve different assumptions about the constant term, a, and the term error, ε. Indeed, a lot of factors can affect the explained variable coefficient and yet some of these factors may not be explicitly introduced as explaining variables in equation (4.2). These factors are approximated by residuals structure. In the panel approach, three factors can be envisaged: factors affecting the endogenous variable in a different way according to the period and/or the individual (country, here) considered; factors having the same impact on all individuals, but whose influence depends on the period (time effects); and other factors that reflect structural differences between individuals, i.e. independent of time (individual effect). Time and individual effects can be combined. Individual effects can be fixed (a totally heterogeneous structure) or random (an intermediary specification between no individual effect and fixed effects). Contrary to Holtz-Eakin (1994), who has used first-differencing to eliminate the country-specific effects, taking first differences is in our case justified by the non-stationarity of data. So, there is a priori no reason to suppose that individual effects are null. In a study on American data, Evans and Karras (1994) have excluded the possibility of estimating a first-differences estimation with individual effects. They explain that these effects would entail permanent differences in regional productivity growth and could imply the reallocations of factors to the more successful states. However, in our study, it seems unlikely that such reallocation mechanisms appear at an international level given the low worker mobility across European countries (Hurlin, 1996). The hypothesis of

122 Fiscal Policy in the European Union Table 5.9

OLS results with individual effects

Regression (3.2) Constant (kg ) (kp ) R2

Fixed effect

Random effect

0.06 (8.93) 0.09 (4.42) 0.30 (5.09) 0.84

0.06 (9.42) 0.08 (3.86) 0.33 (7.35) 0.87

Note: values are in brackets.

individual effects in the specification in first differences is confirmed by Fisher tests. However, a correlation between these individual effects and the explaining variables has to be examined. To test this correlation, a Hausman test is performed (Hausman, 1978).8 The Hausman specification test rejects the null hypothesis of independence and thus urges the use of the fixed-effects results. Table 5.9 shows that the estimated elasticities with fixed effects are very close to those obtained with random effects: this seems to prove that the bias linked to a potential correlation is very weak. Thus, the growth rate of public capital has a significant and positive impact on the economic growth rate and this conclusion holds whatever the form of the individual effect. Contrary to Evans and Karras (1994), the introduction of fixed individual effects does not affect the results. With fixed effects (as recommended by the Hausman test), the estimated elasticity for public capital is 9 per cent, which is much lower than the result Aschauer obtained. But it has to be noted that our variables are not in level but expressed in growth rate. However, in spite of these positive results, it must be noted that the elasticity for private capital is too high to be credible in a growth rate specification. The large coefficient obtained for the growth of physical capital suggests that these results are biased upwards. The large coefficient can be explained by the endogeneity bias which is likely to appear with OLS estimates. The OLS method relies on the assumption that all explanatory variables are exogenously determined and thus not correlated with the error term. This assumption is often violated since most of the economic variables interact and feed back on each other when there are changes in the economy. For instance, if there is a correlation between economic growth and private capital, such simultaneity in

Public Capital and Economic Growth

123

the relationship may lead to upward-biased estimates of the coefficients. One of the most likely sources of simultaneity is business cycle effect. In this study, this bias is in part controlled by taking five-year averages of data. But using an instrumental variables method is a more radical way to tackle the simultaneity bias. For this reason, the GMM estimator proposed by Arellano and Bond (1991) is used. This estimator is widely used in short dynamic panels. Additional instruments, whose validity is based on orthogonality between lagged values of the dependent variable and the errors, are also used. The estimable equation is differenced and then all possible lagged values of each endogenous variable are used as instruments for the first difference of that variable. As instruments for the variables that are correlated with the error term, we use the lagged levels of the observed series yit−3 to instrument for yit−1 − yit−2, kgit−2 to instrument for kgit − kgit−1 and kpit−2 to instrument for kpit − kpit−1 . The original estimator of Arellano and Bond (1991) makes use of all possible lags from t − 2. We limit our instruments to the lags at t − 2 or t − 3 because using a large number of instruments is not feasible with a small cross-sectional panel and moreover, it could lead to an over-fitting bias. In addition, by using levels of the dependent variable lagged by one additional period to instrument for the first-differenced term – the first instrument for yit−1 − yit−2 is yit − 3 – we account for the potential presence of measurement errors (Blundell and Bond, 1998). In our estimation we also correct for the heteroskedasticity that may be present in the error structure by following the two-step procedure proposed by Arellano and Blond (1991). Finally, the GMM estimator is consistent if there is no second-order serial correlation in the error term of the first-difference equation. The most common test of the instrument is Sargan’s (1958) test of overidentifying restrictions implying the absence of correlation between the instruments and the error term. This test is carried out and supports the validity of our instruments (no second-order correlation). The problem of endogeneity is not the only one which blurs the estimates. The high level of R2 in the regression (4.2) seems suspicious and may be the sign of a multicolinearity bias. If we look at the correlations matrix, we can see a high correlation between GDP per worker and private capital per worker (0.77); a transformation of variables is necessary to solve this problem. Using the variable ‘share of private capital per worker in GDP per worker’ instead of ‘private capital per worker’ significantly reduces this bias. The same transformation is carried out for the public capital in order to be homogeneous. Table 5.10 reports the results when GMM is used with the transformed variables. The Sargan test confirms that our choice of

124 Fiscal Policy in the European Union Table 5.10

Final results with GMM

(kg ) (kp )

Without fiscal variables

With fiscal variables

0.10 (3.07) 0.15 (3.48)

0.09 (1.88) 0.15 (2.00) −0.03 (−1.13) (0.06) (1.78) 0.44 0.18

public consumption spending fiscal surplus R2 Sargan

0.41 0.27

Notes: The lagged dependent variable is also included in the regression. Kg and kp represent respectively the share in percentage of public capital and private capital in GDP (expressed per worker).

instrumental variables is valid. Once again, the productive role of public capital is confirmed. The outcome is an elasticity of 0.10 for public capital and 0.15 for private capital. These coefficients and the level of R2 (0.41) are much more plausible. To confirm these results, a robustness test can be performed. To do this, we add two fiscal variables in the regression. Indeed, the existence of omitted variables correlated with the regressors leads to inconsistent estimators. Following Kneller et al. (1999), we introduce the fiscal surplus and the public consumption spending. Indeed, Kneller et al. (1999) explain that most growth studies dealing with the link between fiscal policy and growth obtained mixed results by failing to control both sides of the government budget constraint in the regression. As a result, when variables from only one side of the budget constraint are included in the regression, such estimates are likely to be biased since they would be capturing the indirect effect that the omitted element from the other side of the budget has on growth. For this reason, we add public consumption spending and fiscal surplus in our regression. With this last estimation, the productive role of public capital is definitively confirmed (Table 5.10). Indeed, when both fiscal variables are added, the estimated elasticity for public capital is 0.09. These results have the same order of magnitude as the previous one, which tends to confirm the robustness of this conclusion.

Public Capital and Economic Growth

125

5 Conclusion This chapter has investigated the growth effects of public capital in Europe using a time series framework and a panel approach. When the estimations for France, Germany and Italy are carried out following the specification proposed by Aschauer (1989a), implausible results are obtained. It confirms the fact that this type of estimation is mis-specified and has to be modified. As mentioned in the literature, a lot of econometric biases may lead to inconsistent estimates if they are not correctly treated. For this reason, we have decided to test the impact of public capital on a European panel (1965–2002) by controlling for econometric issues (in particular the endogeneity issue is treated by using a GMM approach). Public capital turns out to have a significant productive role and this result seems robust upon the inclusion of new variables in the regression. This result has strong implications in terms of economic policy. It confirms that the Lisbon Strategy is well-grounded since our estimations have shown that public capital is a key factor for economic growth. One way of fostering this type of spending would be the introduction of a ‘golden rule of public finances’. With this kind of rule, public investment spending will be more than simply the adjustment variable to achieve the 3 per cent ceiling imposed by the Stability and Growth Pact (as was the case in the 1980s and 1990s during the Maastricht consolidation process). The research papers on the quality of fiscal adjustments have shown that budgetary consolidations are less likely to last and to be successful when public investment is cut (Alesina and Perotti, 1997). So, even if one considers that fiscal balance should be the top priority for European countries, public investment is not the relevant variable to sacrifice, all the more so when its productive role is kept in mind.

Notes 1. 2. 3. 4.

5. 6. 7. 8.

Only panels composed of developed countries will be mentioned. Even if this study is based on regional data for the United States. The capital data series we use are only available until 2002. The regression proposed by Aschauer seems to be mis-specified, which is why, rather than testing the impact of public capital on productivity, we test its influence on economic growth. A possible alternative may be to use data based on trend measures. Results available upon request. Variables are in logarithms. Results available upon request.

126 Fiscal Policy in the European Union

References Aaron, H., ‘Discussion’, in A. H. Munnell (ed.), Is There a Shortfall in Public Capital Investment? (Boston: Federal Reserve Bank of Boston, 1990). Ai, C. and S. P. Cassou, ‘A Normative Analysis of Public Capital’, Applied Economics, 27 (1995): 1201–9. Alesina, A. and R. Perotti, ‘Fiscal Adjustments in OECD Countries: Composition and Macroeconomic Effects’, IMF Staff Papers, 44(2) (1997): 210–48. Arellano, M. and S. Bond, ‘Some Tests of Specification for Panel Data: Monte Carlo Evidence and an Application to Employment Equations’, Review of Economic Studies, 58 (1991): 277–97. Argimon, I., J. Gonzalez Paramo, M. J. Martin and J. M. Roldan, ‘Productivity and Infrastructure in the Spanish Economy’, Working Paper, Banco de España, Servicio de Estudios, no. 9313 (1993). Aschauer, D. A., ‘Is Public Expenditure Productive?’, Journal of Monetary Economics, 23(2) (1989a): 177–200. Aschauer, D. A., ‘Public Investment and Productivity Growth in the Group of Seven’, Journal of Economic Perspectives, 13 (1989b): 17–25. Bajo-Rubio, O. and S. Sosvilla-Rivero, ‘Does Public Capital Affect Private Sector Performance? An Analysis of the Spanish Case, 1964–88’, Economic Modelling, 10 (1993): 179–84. Balmaseda, M., ‘Production Function Estimates of the Rate of Return on Public Capital’, Working Paper, CEMFI (1997). Berndt, E. R. and B. Hansson, ‘Measuring the Contribution of Public Infrastructure Capital in Sweden’, NBER Working Paper, 3842 (1991). Blanchard, O. J. and F. Giavazzi, ‘Improving the SGP through a Proper Accounting of Public Investment’, European Economic Perspectives Newsletter, CEPR, no.1, February (2003). Blundell, R. W. and S. R. Bond, ‘Initial Conditions and Moment Restrictions in Dynamic Panel Data Models’, Journal of Econometrics, 87 (1998): 115–43. Coe, D. T. and R. Moghadam, ‘Capital and Trade as Engines of Growth in France’, IMF Staff Papers, 40(3) (1993). Crowder, W. J. and D. Himarios, ‘Balanced Growth and Public Capital: an Empirical Analysis’, Applied Economics, 29 (1997): 1045–53. Dalamagas, B., ‘A Reconsideration of the Public Sector’s Contribution to Growth’, Empirical Economics, 20(3) (1995): 385–414. Eisner, E., ‘Infrastructure and Regional Economic Performance: Comment’, Federal Reserve Bank of New England, New England Economic Review, 47 (1991): 47–58. Evans, P. and G. Karras, ‘Are Government Activities Productive? Evidence from a Panel of US States’, Review of Economics and Statistics, 76(1) (1994): 1–11. Finn, M., ‘Is All Government Capital Productive?’, Federal Reserve Bank of Richmond, Economic Quarterly, vol. 76, no. 79(4) (1993): 53–80. Fitoussi, J. P. and J. Creel, How to Reform the ECB, Center for European Reform, London, October (2002). Flores de Frutos, R., M. Gracia-Diez and T. Perez-Amaral, ‘Public Capital Stock and Economic Growth: an Analysis of the Spanish Economy’, Applied Economics, 30 (1998): 985–94.

Public Capital and Economic Growth

127

Ford, R. and P. Poret, ‘Infrastructure and Private-Sector Productivity’, OECD Economic Studies, 17 (1991): 63–89. Gramlich, E. M., ‘Infrastructure Investment: a Review Essay’, Journal of Economic Literature, 32 (1994): 1176–96. Gurgand, M., ‘Capital humain et croissance: la littérature empirique à un tournant?’, Economie Publique, 6 (2000): 71–93. Hausman, J. A., ‘Specification Tests in Econometrics’, Econometrica, 6 (1978): 1251–71. Holtz-Eakin, D., ‘Public-sector Capital and the Productivity Puzzle’, Review of Economics and Statistics, 76 (1994): 12–21. Hulten, C. R. and R. M. Schwab, ‘Public Capital Formation and the Growth of Regional Manufacturing Industries’, National Tax Journal, 44(4) (1991): 121–34. Hurlin, C., ‘La Contribution Productive du Capital Public à la Croissance: Estimation sur un Panel Sectoriel de dix Pays de l’OCDE’, mimeo (1996). Kamps, C., The Dynamic Macroeconomic Effects of Public Capital: Theory and Evidence from OECD Countries (New York: Springer, 2004a). Kamps C., ‘New Estimates of Government Net Capital Stocks for 22 OECD Countries 1960–2001’, IMF Working Paper, no. 67 (2004b). Kavanagh, C., ‘Public Capital and Private Sector Productivity in Ireland 1958–90’, Journal of Economic Studies, 24(1/2) (1997): 72–94. Kneller, R., M. Bleaney and N. Gemmel, ‘Fiscal Policy and Growth: Evidence from OECD Countries’, Journal of Public Economics, 74 (1999): 171–90. Lau, S. H. P. and C. Y. Sin, ‘Public Infrastructure and Economic Growth: Time Series Properties and Evidence’, Economic Record, 73 (1997): 125–35. Le Cacheux, J., ‘A Golden Rule for the Euro Area?’, paper presented at the Fiscal Discipline Workshop, CDC-Ixis and CEPII, Paris, November (2002). Levin, A., C. F. Lin and C. S. J. Chu, ‘Unit Root Test in Panel Data: Asymptotic and Finite Sample Properties’, Journal of Econometrics, 108 (2002): 1–24. Munnell, A. H., ‘Why Has Productivity Growth Declined? Productivity and Public Investment’, New England Economic Review, January–February (1990): 2–22. Nadiri, M. and T. Mannumeas, ‘Infrastructure and Public R&D Investments, and the Growth of Factor Productivity in US Manufacturing Industries’, NBER Working Paper, 4845 (1994). Otto, G. D. and G. M. Voss, ‘Public Capital and Private Production in Australia’, Southern Economic Journal, 62 (1996): 723–38. Oxley, H. and P. Martin (1991), ‘Controlling Government Spending and Deficits: Trends in the 80s and Prospects for the 90s’, OECD Economic Studies, 17 (2001): 145–89. Ram, R. and D. Ramsey, ‘Government Capital and Private Output in the United States’, Economics Letters, 13 (1989): 223–6. Ratner, J. B., ‘Government Capital and the Production Function for US Private Output’, Economics Letters, 13 (1983): 213–17. Sargan, J., ‘The Estimation of Economic Relationships Using Instrumental Variables’, Econometrica, 26 (1958): 395–415. Seitz, H., ‘Infrastructure Investment and Economic Activity: Theoretical Results and International Evidence’, in Deutsche Bundesbank (ed.), Investing Today for the World of Tomorrow (Berlin: Springer, 2001). Sturm, J. E. and J. de Haan, ‘Is Public Expenditure Really Productive? New Evidence for the US and the Netherlands’, Economic Modelling, 12 (1995): 60–72.

128 Fiscal Policy in the European Union Tatom, J. A., ‘Public Capital and Private Sector Performance’, Federal Reserve Bank of St Louis Review, 73 (1991): 3–15. Toen-Gout, M. and M. Jongeling, ‘Investment in Infrastructure and Economic Growth’, OCFEB Research Memorandum, 9404 (1994). Vijverberg, V. P. M., C. P. C. Vijverberg and G. L. Gamble, ‘Public Capital and Private Productivity’, Review of Economics and Statistics, 12 (1997): 60–72. Wylie, P. J., ‘Infrastructure and Canadian Economic Growth 1946–1991’, Canadian Journal of Economics, 29 (1996): 351–5.

6 Has France Become a Deficit Running Country?1 Jérôme Creel, Guy Gilbert and Thierry Madiès

1 Introduction The main purpose of this chapter is to review the French public finances over a long horizon in order to examine whether the common view, largely shared during the latest presidential election, according to which French governments have always been ‘deficit runners’, is right. In fact, we show that French deficits since 1980 have stemmed from: (i) the large gap between long-term interest rates and economic growth in the early 1980s; (ii) high unemployment and low economic growth, most notably during the recession of the early 1990s whose effects have been felt until recently; and (iii) pro-cyclical policies related to the electoral cycle. The first two elements somewhat reduce the general responsibility of French successive governments for the recent increase in public deficits and debts. Conversely, other prominent elements are the direct responsibility of these governments: first, the inability to implement a contra-cyclical fiscal policy after economic growth had resumed. This inability has delayed the implementation of a strong and persistent consolidation of public finances. The second reason is concerned with the delay in adopting more transparent procedures during the budget process which are aimed at increasing the so-called ‘credibility’ of institutions. Of course, credibility has no influence on the deficit per se, but it is also meant to hold creative accounting in check. Moreover, had France been ‘credible’, it would have been less urged by the European authorities to reduce its deficits in the worst situation, i.e. at the moment when economic growth was flat. As far as the longer term is concerned, the situation of French public finances appears mildly worrying. If real economic growth resumes quickly and is strong enough, and the unemployment rate continues 129

130 Fiscal Policy in the European Union

to fall, then deficits and public debt will also naturally fall. Moreover, reforms of the budgetary process associated with structural fiscal reforms will no doubt improve France’s public finances. The chapter is organized as follows. Section 2 presents an overview of public deficits and debt and tackles the problem of sustainability of French fiscal policy over the last three decades. Section 3 identifies the causes of deficits and debt accumulation, focusing on both electoral and business cycles. Section 4 deals with changes taken to improve shortterm fiscal imbalances and shows that recent structural measures have been undertaken in order to enhance the transparency of the budgetary process to meet public deficit and public spending targets.

2 An overview of public deficits and debt over 1970–2005 That France experienced some episodes of expansionary fiscal policies that resulted in a long-lasting worsening of public finances cannot be denied. Figure 6.1 testifies to the fact that unlike in the 1970s and 1980s, the French public deficit has been relatively substantial in comparison with other EU members since the 1990s. The main and most recent break concerning the trend of public imbalances is related to the Maastricht Treaty which required European countries to strongly reduce their deficits to meet the so-called public 6 4 2 0 ⫺2 ⫺4 ⫺6

Belgium Germany Denmark Spain France United Kingdom Ireland Italy Netherlands

⫺8 ⫺10 ⫺12 ⫺14 ⫺16

1980

1982

1984

1986

1988

1990

1992

1994

Figure 6.1 Government budget balance, % of GDP Source: OECD.

1996

1998

2000

2002

2004

Has France Become a Deficit Running Country? 131

finance criteria. In this respect, the French authorities demonstrated their concern for sounder public finances. As can be observed, the highest government deficit was reached in 1993 in France, during the worst economic crisis since the Second World War, and between 1993 and 1997, the fiscal stance was considerably reversed to cope with the criteria. Within the European context, some have argued, most notably at the European Commission and the OECD, that France had stopped relatively early its efforts to cut deficits (see Buti and van den Noord, 2003; Burns and Goglio, 2004); deliberately so, because the fiscal limits embedded in the Stability and Growth Pact (SGP) had been outperformed. Indeed, Figure 6.1 shows that, except in the 1980s, France has generally not been among Europe’s best pupils in terms of its fiscal stance: between 1991 and 1997, France had lower deficits than Italy, Spain and the United Kingdom only; and in 2000, 2001 and 2005, it had lower deficits than Italy and Germany. The French fiscal stance has therefore generally been worse than in the majority of its EU partners. The burden of gross public debt2 was on a downward trend over the 1970s thanks to fiscal surpluses in the 1960s and moderate deficits in the 1970s,3 but the situation has changed dramatically since the early 1980s. The ratio of general government gross public debt to GDP was multiplied by 2.5 from 1980 to 2005 (from 30 per cent to more than 75 per cent of the GDP) (see Figure 6.2). According to the Maastricht definition, the French debt has been above 60 per cent of GDP since 2003. Five main periods of this growth process can be identified over the last two and a half decades: 1. First, following a string of fiscal deficits in a context of decreasing inflation, high interest rates, and sluggish economic growth, this ratio has been rising by 2 percentage points of GDP per year on average from 1981 to 1986. During these years, many reforms were implemented and fuelled the expenditure side of the budget: retirement at the age of 60, nationalizations, etc. 2. Between 1987 and 1991, the resumption of economic growth and slightly lower interest rates allowed for a short period of quasistability: the ratio of gross public debt has been increasing by a 0.5 percentage point of GDP per year. 3. The economic slowdown starting in 1991 and which later became a genuine recession – in 1993 – in a context of rising interest rates rekindled the upward trend in the ratio which was growing by 5 percentage points of GDP per year on average until 1996. Part of this

132 Fiscal Policy in the European Union 80

16

70

14

60

12

50 10 40 8 30 6 20 4

10

2

0 ⫺10

1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005

0

General government gross financial liabilities, % of GDP General government net financial liabilities, as % of GDP GDP nominal growth rate (%, right-scale)

Figure 6.2 Net and gross public debt, France, % of GDP, and nominal GDP growth Source: OECD.

growth was due to a discretionary counter-cyclical policy, as will be discussed later. 4. Between 1997 and spring 2001, the economy experienced a strong recovery with a 3 per cent average real growth rate.4 The trend of gross public debt to GDP was flatter in 1997–8 and then it fell by almost 1 percentage point of GDP per year on average in 1999–2000 so that the ratio of gross public debt amounted to 57.6 per cent (Maastricht definition) at the end of 2000. 5. In 2001, the economic growth rate slowed down again and the ratio of public debt-to-GDP stopped decreasing and even rose again to exceed 60 per cent in 2003. The question which arises naturally is whether French fiscal policy is sustainable or not. Broadly speaking, fiscal policy is considered to be sustainable if the current public debt can be repaid by future fiscal surpluses without any substantial change in the fiscal stance (Blanchard, 1993). Sustainability also ensures that government will not play a Ponzi game against private agents, repaying past loans and interests with new loans at an increasing rate. Using the government present-value budget constraint and assuming a non-Ponzi game, the current value of the debt must equal the discounted sum of expected primary surpluses (assuming

Has France Become a Deficit Running Country? 133

the so-called transversality condition). In other words, the discounted debt must follow a zero mean stationary stochastic process. Most sustainability tests are hence based on the existence of a unit root in the debt process (see Hamilton and Flavin, 1986; Wilcox, 1989). Recent evaluations for France point to mixed results. INSEE (2006) argues that ‘strong sustainability’ must be rejected but it also argues that France has experienced a regime of ‘weak sustainability’. French public debt is found to be non-stationary (hence, ‘strong non-sustainability’) but a 1 percentage-point-of-GDP-increase in public spending is found to produce a 0.24 percentage point of GDP of extra tax receipts (hence, ‘weak sustainability’), a low response indeed in comparison with the USA for instance5 but not a null one. Creel and Le Bihan (2006) adopt different techniques (Augmented Dickey-Fuller, Schmidt-Phillips and Kwiatkowski, Phillips, Schmidt and Shin tests) to different public finance variables. They conclude that public debt is non-stationary but some key determinants like the primary surplus or the primary structural surplus (see below for definitions) are stationary according to the ADF test and the KPSS test. Fève and Hénin (2000), applying Feedback Augmented Dickey-Fuller tests where non-sustainability is characterized by the null hypothesis of both a unit root in the debt process and the nullity of a deficit correction in response to inherited debt, conclude against sustainability in the French case. However, unit root tests suffer from a well-known lack of power due to small sample size. Similar difficulties appear when sustainability tests are performed using cointegration equations. Moreover, sustainability requires an assessment of future, not past, fiscal policies. After qualifying the determinants of past deficits in section 3, section 4 will elaborate on recent improvements in the budgetary process which may enhance French public finances; these improvements may also remove the risks of ‘unsustainability’.

3 Public finances, electoral cycles and business cycles: a French perspective Although the deterioration of French public finances is uncontroversial in historical perspective, it remains to be explained why that situation occurred. At least three explanations can be put forward which are not mutually exclusive. The first one, clearly visible in Figure 6.1, relates the size of government deficits to country size. Countries with the highest government deficits since 2003 have been ‘large’ countries: France, Germany, Italy and the UK. Le Cacheux (2005) and Laurent and Le Cacheux (2006) argue that the European race towards competitiveness

134 Fiscal Policy in the European Union

since the Single European Act and the adaptation of European institutions to this objective have been almost mechanically detrimental to ‘large’ countries that are less open to international trade than ‘smaller’ ones. In this context, ‘large’ countries have had to rely more on domestic policies. Focusing on purely domestic arguments leaves us with two other explanations: on the one hand, French public finances may have been influenced by the political cycle; on the other hand, public finances can be related to the business cycle. In both cases, public debt becomes a crucial factor in explaining the economic and political trends of the government deficit: first, slower economic growth may produce higher deficits, through automatic stabilizers, that have to be financed by higher debt; the dynamics of public debt then produces higher deficits, and a vicious circle begins. Second, public debt levels can be made instrumental in various ways by politicians: either as a bad inheritance to their successors, or as an excuse for implementing a tough fiscal consolidation. 3.1 An analysis of the main components of public deficits The previous section has suggested that high debts and deficits have emerged in France since 1980. Disentangling the different causes for this phenomenon is the subject of this chapter. As a first step, we turn towards the available fiscal indicators and check the extent to which this phenomenon has been discretionary. This first step is a prerequisite to a more thorough study of the political determinants of high debts and deficits. Creel and Sterdyniak (1995) and Gali and Perotti (2003), have defined four independent components of a public deficit, related to interest rates, growth and discretion. Their main innovation consisted of introducing a distinction between ‘cyclical discretion’ and ‘structural discretion’. Creel and Sterdyniak (1995), for instance, argued that government facing a GDP slowdown can decide to change the tax or social system to dampen the slowdown; in so far as this modification is temporary – i.e. neutral over the cycle – this discretionary action retains a cyclical feature. Therefore, a government has two cyclical components: one automatic (the usual ‘automatic stabilizers’ that are operating without changing the tax and social systems) and another one ‘discretionary’ (operating through a temporary change in the tax and/or social systems). Based upon this distinction, the responsibility of the central government in the evolution of deficits can be delimited. Unfortunately, there is no macroeconomic indicator of the discretionary temporary part of fiscal

Has France Become a Deficit Running Country? 135 2 1 0 ⫺1 ⫺2 ⫺3 ⫺4 ⫺5 ⫺6

1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 Government net lending, as a percentage of GDP Cyclically adjusted government net lending, % of potential GDP Government primary balance, as a percentage of GDP Cyclically adjusted government primary balance, % of potential GDP

Figure 6.3 Various fiscal surpluses, France, % of GDP Source: OECD.

deficit. Creel and Sterdyniak (1995) therefore argued that the cyclically adjusted primary general government balance was, among the different indicators of the fiscal stance, the closest one to its discretionary part. Data for this indicator are now computed by the OECD and delivered in the Economic Outlooks. Using OECD data and four fiscal policy indicators, one can draw very different pictures of French public finances since 1972. Figure 6.3 draws on the overall public deficit (government net lending), the primary deficit (which corresponds to the former adjusted to exclude the net interest payments), the OECD cyclically adjusted deficit and the OECD cyclically adjusted primary deficit, all expressed in percentage points of GDP or potential GDP. Over the entire time span, the average public deficit has been close to 2.5 per cent of GDP which may seem high, but the same average for the primary deficit has been 0.5 per cent of GDP. With such a primary deficit, the critical gap, i.e. the difference between the interest rate and GDP growth rate, which would stabilize public debt at 60 per cent of GDP, would be a small 0.8 per cent, i.e. a nominal GDP growth rate superior by 0.8 percentage point to the nominal longrun interest rate.6 Reaching such a critical gap should not be out of reach

136 Fiscal Policy in the European Union 10 8 6 4 2 0 ⫺2

Belgium Germany Denmark Spain France United Kingdom Ireland Italy Netherlands

⫺4 ⫺6 ⫺8 ⫺10

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

Figure 6.4 Cyclically adjusted government primary balance, % of potential output Source: OECD.

for a fast-growing economy. Now, turning towards an indicator of discretionary fiscal stance, it is worth emphasizing that at the beginning of the 1990s, when overall deficits were close to 6 percentage points of GDP, the cyclically adjusted primary deficit was a mere 2 percentage points deficit that finally turned to a surplus until the beginning of the 2000s. Since 2005, the French discretionary fiscal stance has been neutral, and public debt can thus be expected to decrease in the future. These different indicators of the fiscal stance are highly complementary. On average, the burden of interest payments on the overall deficit has amounted to 2 percentage points of GDP and the remaining deficit has recently turned into a surplus. Hence, the situation of public finances in France does not seem out of control; and the 2007 rise in public debt and relaxation of fiscal consolidation may be easily reversed. It remains to understand the mechanisms that produce cyclicality in French public finances. As can be seen from Figure 6.3, the cyclically adjusted government primary balance has fluctuated regularly: surpluses between 1982 and 1989, deficits between 1990 and 1995, surpluses between 1996 and 2001 and, finally, deficits between 2002 and 2005. This cyclical feature appears typical of French policy: nowhere in the EU has a feature like this appeared with such a regular frequency (Figure 6.4). In the following section, we question the relevance of a political cycle in France that would influence the trend of discretionary budget deficits.

Has France Become a Deficit Running Country? 137

3.2 Deficits and the electoral cycle Recent papers have renewed attention on the relationship between fiscal deficits and the electoral cycle in the EU context (Buti and van den Noord, 2003) or for democracies (Brender and Drazen, 2005). Whereas the first concludes that ‘a “genuine” discretionary boost took place in correspondence to political elections’ in EMU countries between 1999 and 2002, the second argues that in ‘established democracies’ (to be opposed to ‘new’ democracies) the electoral cycle is no longer a significant determinant of fiscal deficits. The common perception is that incumbents often adopt opportunistic behaviours before elections – through an expansionary economic policy, for instance – to increase their re-election chances. ‘Political business cycles’ would thus arise: macroeconomic cycles would be induced by the political cycle. One therefore assumes that good macroeconomic conditions prior to the election will help the incumbent to be re-elected. The relationship between politics and public finance has a long history in the political economy literature. To mention only one paper, Roubini and Sachs (1989) focused on government structures (and the so-called ‘war of attrition’) and showed that difficulties of political management in coalition governments, short average tenure of government and several political parties in the coalition government usually produce high fiscal deficits. Empirical evidence on the specific relationship between the electoral cycles and fiscal deficits remains equivocal. Notwithstanding the two above-mentioned papers, Vavouras (1999) found significant evidence that political fiscal cycles, resulting from clientelist politics, resulted in public deficit growth in Greece in the 1980s and 1990s. In contrast, Galli and Padovano (2002) were unable to show that fiscal data in Italy, between 1950 and 1998, had responded to electoral events. To investigate the incidence of the electoral cycle on French fiscal deficits since 1978, a statistical analysis is performed using the cyclically adjusted primary balance in percentage of potential output. The discretionary fiscal stance is then compared over different situations regarding the electoral cycle: election year, pre- and election years, post-election years, and cohabitation years. We have also computed the average discretionary fiscal stance of ‘left-wing’ and ‘right-wing’ governments. The results are reported in Table 6.1. As was already mentioned above, the political economy literature argues that pre-election and election years may witness higher deficits than in other periods. France also witnessed ‘cohabitation years’, i.e. periods during which the head of government did not have the same

138 Fiscal Policy in the European Union Table 6.1 Discretionary fiscal policy over the electoral cycle in France (1978–2005)

Mean*

All years

Election years

Pre- & election years excluding 1997

Election and postelection years excluding 1997

Leftwing govts

Rightwing govts

Cohabitation years

0.00

−0.27

−0.14

−0.41

0.27

−0.36

0.27

∗ Figures are based upon the cyclically adjusted primary balance in percentage points of potential output. Source: OECD, computations by the authors.

political colour as the president. This situation can be labelled an ‘ambiguous coalition’: responsibility for the fiscal stance is unclear; high fiscal deficit may ensue with both authorities (head of government and president) claiming that the fault lies in the other’s activism.7 Finally, the political economy literature usually concludes that ‘left-wing’ governments are more deficit-prone than ‘right-wing’ ones.8 We chose electoral years pertaining to ‘Legislatives’ (parliamentary elections), conforming with Brender and Drazen (2006): France is defined as a parliamentary system because the government and its head are dominant in determining economic policy. Cohabitation is more ambiguous, and we checked for its specific incidence on the outcomes. Election years in France were: 1978, 1981, 1986, 1988, 1993, 1997 and 2002. Over the entire time span (1978–2005), discretionary fiscal policy in France has been neutral. This first outcome confirms the view presented in section 2: on a long horizon, the situation of public finances in France is not worrying. On a shorter horizon, however, the electoral cycle seems to influence the discretionary fiscal stance. During election years in fact, this stance has been expansionary, although the annual fiscal impulse has never been substantial (0.3 percentage point of GDP on average). Moreover, one year after the election, French governments use the benefit of their ‘honeymoon’ with the country to perform higher discretionary deficits than before the election: pre-election and election years witness a smaller fiscal stance than election years only, whereas adding post-election years to election years increases the fiscal impulse by a further 0.1 percentage point of GDP.9

Has France Become a Deficit Running Country? 139

Checking the political colour origin of fiscal deficits in France gives an amazing outcome: contrary to the literature, ‘left-wing’ governments have, on average, performed discretionary fiscal surpluses whereas ‘rightwing’ governments have, on average, implemented discretionary fiscal deficits. However, it is necessary to scrutinize whether this outcome does not result from a mirror effect. Indeed, one reason explaining this outcome may rest on ‘cohabitation years’ and on the fact that the last five-year-long cohabitation was dominated by the left: during cohabitation periods, the fiscal stance is positive (Table 6.1). Contrary to our expectation of a ‘not-my-fault’ attitude between the two most important authorities of the French political system – the president and the head of government – cohabitation constrains fiscal policy. Since 1978, three cohabitation periods have occurred, in 1986–8, 1993–5 and 1997– 2002: except between 1993 and 1995 (a deep recession period), fiscal surpluses were high. In the following, we check if this political outcome resists more scrutiny on the fiscal performance of successive French governments since 1978. Table 6.2 distinguishes between these governments. For each, the cumulative change in the overall public surplus, expressed in percentage points of GDP, is explained by discretionary actions undertaken by the government on the revenue or the expenditure sides. It is also explained by the evolution of interest payments and the automatic stabilizers. Explanations on the computation methodology are given in Box 6.1. A look at Table 6.2 immediately shows that total deficits have been higher in France under left-wing governments than under right-wing ones, except under the Jospin (left-wing) government. Between 1981 and 1985, France underwent its highest fiscal deficit period with an average deficit of 3 per cent of GDP. Between 1988 and 1992, deficits were also high. Two main differences between these two episodes must be mentioned: under the first and second governments of President Mitterrand, bad economic growth performance (for which negative automatic stabilizers testify) and high interest payments explain the bad fiscal performance; under the Rocard and Cresson governments, however, economic growth was tough and interest payments were lowered. Despite different economic situations, it is noticeable that changes in discretionary fiscal expenditures had almost the same size: where earlier left-wing governments could argue that these changes corresponded to a temporary dampening of the economic crisis, governments between 1988 and 1992 could not use that argument. In comparison with the Rocard–Cresson era, public spending was sharply reduced under the Jospin government. The good fiscal

1978–80

1981–85

1986–87

1988–92

1993–94

1995–96

1997–2001

2002–5

Prime Ministers

Barre

Mauroy & Fabius

Chirac

Rocard & Cresson

Balladur

Juppé

Jospin

Raffarin & de Villepin

Party

Right

Left

Right

Left

Right

Right

Left

Right

Cumulative change in public surplus (in % of GDP) Contributions of . . . discretionary changes in compulsory levies discretionary changes in general govt spending interest payments automatic stabilizers (sign -: deficit)

1.5

−3.1

0.9

−2.4

−1.1

1.4

2.5

−1.3

2.8

1.9

0.2

−0.9

1.2

1.3

−0.1

0.2

4.1

6.5

−1.7

5.2

2.0

−0.6

−1.1

4.2

0.2 0.0

1.3 −2.2

−0.1 0.2

0.5 2.0

0.2 −1.4

0.3 −0.6

−0.6 1.6

−0.4 −0.5

Notes: (1) Table 6.2 splits up changes in public surplus-to-GDP ratio over 8 distinct periods according to (a) discretionary changes in tax policy (a rise in compulsory taxes-to-GDP ratio results in increasing the surplus); (b) the gap between general government expenditure growth and the potential growth rate which allows the estimation of discretionary changes in public expenditures (if public spending growth is higher than potential growth, it is then considered as an expansionary deliberate policy); (c) changes in interest payments; (d) the gap between real economic growth rate and potential output growth which allows the estimation of the effects of cyclical fluctuations on the fiscal deficit. (2) Global estimated effects differ from the effective cumulative change in public-to-GDP ratio since computations ignore non-tax revenues and acquisitions and disposals of non-financial assets. Their estimated contributions are reported in Figure 6.5 and correspond either to well-known large nationalization waves (most notably: 1981–5) or equally well-known privatization waves (most notably: 1986–7). Sources: INSEE, OECD, computations by the authors.

140

Table 6.2 Fiscal deficits and the political cycle: contributions to the cumulative change in general government surplus (in percentage points of GDP) over the last three decades

Has France Become a Deficit Running Country? 141

Box 6.1 Computation methodology in Table 6.2 Discretionary changes in compulsory levies are estimated as a rise in the ratio of compulsory levies to actual GDP. A rise (resp. fall) involves a discretionary and contractionary (resp. expansionary) fiscal stance. Discretionary changes in general government spending exclude interest payments and acquisitions and disposals of non-financial assets; they are thus limited to current spending and social transfers. The reason for excluding acquisitions and disposals of non-financial assets is that they consist of structural expenditures which are undertaken in a long-run perspective; limiting ourselves to discretionary changes in current spending and social transfers has to be viewed in the perspective of their possible incidence on the temporary (hence short-term) discretionary part of fiscal deficit.10 Discretionary changes in public spending correspond to the part of their overall changes which are above changes in potential output: maintaining constant the ratio of public spending on potential output is thus a non-discretionary situation. Setting it above or below is discretionary. To compute automatic stabilizers, we used the elasticity of the public surplus to the output gap, 0.5, which stemmed from the estimations of the OECD (Girouard and André, 2005). Fiscal data come from the French national institute (INSEE). Potential output and the output gap come from OECD Economic Outlooks. We checked that these figures were close to INSEE’s.

performance was also explained by lower interest payments and relatively good growth performance. As for left-wing governments, there are exceptions in the average fiscal performance of right-wing governments. Balladur, Raffarin and de Villepin governments underwent overall fiscal deficits. Still like the leftwing case, reasons for bad performance are not the same: under the Balladur government, France coped with a severe recession. Here, it is noticeable that discretionary public spending was only slightly used to dampen the shock, in opposition to left-wing governments of the early 1980s. It must be acknowledged that the European context had changed in the early 1990s, with the start of the transition period before the adoption of the euro: new constraints were limiting the scope for implementing expansionary fiscal policies. Since 2002, hence under the Raffarin and de Villepin governments, France has undergone fiscal deficits again, although the output gap has

142 Fiscal Policy in the European Union

been only weakly negative. The most striking result over this period is the substantial increase in discretionary public spending. This is all the more striking when one compares the fiscal performance of the Juppé government (1995–6) with the situation in the mid-2000s: under similar economic conditions, the Juppé government reduced public spending and increased compulsory levies sharply to cope with the Maastricht criteria. The ‘consolidation fatigue’ which Hughes-Hallett et al. (2003) attribute to EMU countries after the euro was created seems to have been relevant to the French economy since 2002. As a conclusion, distinguishing left-wing and right-wing governments as far as fiscal performance is concerned is not straightforward: nondiscretionary actions, related to economic growth, interest payments and the European context, need to be carefully analysed. Moreover, the relative tendency of all governments to adopt an incremental behaviour (Siné, 2006), where past expenditures are always renewed, has reduced the distinction between political colours and reduced the margins for manoeuvre. This makes the latter outcome of very high discretionary public spending under Raffarin and de Villepin even more surprising.11 Thus far, we have concluded on the expansionary or contractionary nature of successive French governments strongly relying on discretionary changes in compulsory levies and general government spending. One could also check the discrepancy between overall fiscal deficit and automatic stabilizers. Two different types of government can thus be distinguished, and their ‘political colours’ are not crucial. On the one hand, there have been governments which performed overall fiscal deficits beyond what automatic stabilizers required (Mauroy, Fabius, Rocard, Cresson, Raffarin, and de Villepin’s governments). On the other hand, there have been governments which performed overall fiscal deficits (or fiscal surpluses) below (or beyond) what automatic stabilizers required. On our almost 30-year horizon, both types are equitably shared. 3.3 Deficits and the business cycle The focus on automatic stabilizers has led us to adopt the terminology first advocated by Perotti on ‘good’ and ‘bad times’ (Perotti, 1999). Table 6.3 reports French fiscal stances depending on economic conditions: ‘bad times’ (resp. ‘good times’) correspond to a negative (resp. positive) output gap producing a cyclically fiscal deficit (resp. surplus), labelled ‘automatic stabilizers’ in the table. The most striking outcome is the change in fiscal policy occurring over the 1990s decade. Indeed, before 1989, ‘good times’ (resp. ‘bad times’) were associated with an improvement (resp. worsening) in the

Has France Become a Deficit Running Country? 143 Table 6.3 Fiscal deficits and the business cycle: contributions to the cumulative change in general government surplus (in percentage points of GDP) over the last three decades 1978–80 1981–88 1989–92 1993–99 2000–1 2002–5 Good Bad Good Bad Good Bad times times times times times times Cumulative change in public surplus (in % of GDP) Contributions of . . . discretionary changes in compulsory levies discretionary changes in general govt spending interest payments automatic stabilizers (sign -: deficit)

1.5

−2.7

−1.9

2.7

0.1

−1.3

2.8

1.4

−0.2

3.5

−1.2

0.2

4.1

5.0

4.9

0.8

−0.5

4.2

0.2 0.0

1.1 −1.1

0.6 1.1

−0.1 −1.1

0.0 0.7

−0.4 −0.5

Note: see notes (1) and (2) in Table 6.2. Sources: INSEE, OECD, computations by the authors.

fiscal stance. Since 1989 and until 1999, the reverse relationship occurred and fiscal policies were pro-cyclical: before the convergence towards the Maastricht criteria, fiscal policy was over-expansionary; it was overcontractionary during the convergence phase. Although the Raffarin and de Villepin governments were shown to out-perform overall fiscal deficits, their governments have renewed with a more usual trend: fiscal deficits tend to be high when growth is below its potential. It remains to assess the specific incidence of net interest payments on public debt growth. Since part of net interests is linked to the level of the interest rate, it goes within the responsibility of monetary policy. We have taken five-year periods which correspond to key variations in the critical gap (Figure 6.5): the 1981–6 is a period of high variability; 1986–91 is a period of acceleration; 1991–6 and 1996–2001 correspond to periods of decrease, whereas 2001–5 is a period of enhanced stability. After separating net interests and primary expenditures, we have found that the main part of net debt growth was attributable to the former (Table 6.4). As a matter of fact, only one-third of the debt variation between 1981 and 2005 was explained by cumulated primary deficits. Two key features are worth mentioning: first, despite a lower critical gap

144 Fiscal Policy in the European Union

6 5

% of GDP

4 3 2 1 0 1978–1980 1981–1985 1986–1987 1988–1992 1993–1994 1995–1996 1997–2001 2002–2005

⫺1 ⫺2 Figure 6.5 Estimated non-tax revenues and acquisitions and disposals of nonfinancial assets Source: computations by the authors.

Table 6.4 Contributions to the rise of general government net financial liabilities (in percentage points of GDP) 1981–86 1986–91 1991–96 1996–2001 2001–5 1981–2005 Debt variation Cumulated primary deficits Net interests effect

12.3 5.0

6.4 −0.7

23.3 10.5

−5.5 −3.2

9.8 4.3

46.3 15.9

7.3

7.1

12.8

−2.3

5.5

30.4

Source: OECD, computations by the authors.

between 1991 and 1996, net interests kept on pushing debt upwards, testifying to the strong inertia in the dynamics of debt; second, the fall in net debt between 1996 and 2001 was mostly explained by fiscal consolidation rather than by better monetary conditions. It shows that French public authorities are able to curb debt. Still, since 2001, debt growth has resumed and with it primary deficits and interest payments.

4 Policy changes to improve budget balance The deterioration of French public finances was dampened from time to time by discretionary reactions, for instance under the Chirac, Juppé

Has France Become a Deficit Running Country? 145

and Jospin governments when public spending was sharply decreased (Table 6.2). These measures were generally short-lived over the last three decades (section 4.1), but more fundamental or structural reforms have been implemented, some of them under the pressure of European requirements (sections 4.2 and 4.3). 4.1 Tax increases and public spending cuts In a famous seminal paper, Alesina and Perotti (1995) emphasized that permanent improvements in the fiscal balance generally occur through cuts in expenditures and especially via cuts in transfer programmes and compensation of government employees. Conversely, temporary improvements are carried out through tax increases despite the increasing concern about tax competition.12 In this section, we show that short-term improvements in the French fiscal balance have been implemented by both tax increases and, to a lesser extent, expenditure cuts (especially in the period preceding entry into the EMU). The problems arising from increasing spending pressures as the population ages and from the economic and social costs of higher taxes are well known and have been the focus of policy reforms throughout the OECD over the past two decades. However, at first sight and unlike many of its economic partners France has made relatively little progress in preparing the fiscal room to meet these pressures. At the beginning of the 1970s, total general government expenditures in France were equal to 40 percentage points of GDP, 8 points higher than the OECD average at that time but close to the average of current-day EU countries. During the following decades they increased rapidly, and although they recently stabilized on a cyclically adjusted basis at 54 percentage points of GDP, they are still the third highest in the OECD, 16 points higher than the OECD average and 8 points above the EU average. To a certain extent this comparison is influenced by the fact that some countries provide health, education and pension services through the private sector, whereas they are provided by the general government in France.13 However, even after excluding such spending France’s public expenditures share remains one of the highest in the OECD (seventh). As highlighted by Burns and Goglio (2004), the ratcheting up of public spending over the past 30 years suggests that there has been a marked tendency to increase spending in a permanent manner during cyclical peaks, with the result that during downturns the ratio of public expenditures to GDP has risen. It remains to acknowledge that the last surge of public spending in France dates back to the 1974–84 period,14 when primary public

146 Fiscal Policy in the European Union

expenditures were growing at a rate of 5.1 per cent per year in real terms, and GDP was growing at 2.2 per cent. Between 1984 and 1994 and between 1994 and 2004, primary expenditures grew at a rate similar to that of GDP but their distribution between the different levels of government has been modified. Disaggregating general expenditures shows that the central-to-general government expenditures ratio decreased sharply from 50 per cent down to 35 per cent. Due to the gradual application of the decentralization laws of 1982–3 (which transferred some functions from central government to local governments, now more involved than before in spending on education and social interventions), the share of local government expenditures had been rising steadily during the two last decades up to 10 percentage points of GDP (to be compared with 7.9 points in 1980). Last, social expenditures grew by 5 percentage points of GDP between 1980 and 2003; most of this rise has been linked to oldage pensions – they have contributed by more than 2 percentage points of GDP – and to health expenditures which have contributed by almost 1 percentage point of GDP. All in all, the rise in expenditures by local governments and social institutions was met by lower spending by the central government. The most substantial source of spending cuts in France since the 1980s has been wage restraint. A strict management of public-sector wages, especially between 1984 and 1987 and between 1992 and 1993, was implemented. High unemployment rates over the last two decades almost surely helped to impose such restraints; these were also implemented during periods of tight fiscal policy. Moreover, since 2004, public-sector employment has started decreasing because retirees from the public sector have not always been replaced: in 2004, 1,090 public jobs were suppressed, corresponding to 3 per cent of retiring civil servants and 0.1 per cent of public-sector employment. In 2005, more than 7,000 were suppressed (12 per cent of retiring civil servants). This figure was 5,000 in 2006. Conversely, social expenditures and transfers rose sharply as reported above. This rise occurred in spite of a series of measures designed to restrict eligibility and generosity of the various allowances, most notably old-age pensions, unemployment benefits, active labour-market policies (like programmes of subsidized employment contracts for low-skilled workers) and health care reimbursement. Most expenditures cuts have touched subsidized jobs (jobs for young unskilled individuals, or emplois jeunes) which had been promoted by the previous left-wing government; however, those subsidized jobs have generally only changed labels (now contrats jeunes en entreprises, for instance).

Has France Become a Deficit Running Country? 147

On the revenue side, many fiscal instruments have been used to dampen the fiscal deficits from 1980 onwards. They range mainly from taxes on corporate profits to taxes on individual income, with excise taxes and non-tax revenues contributing largely from time to time. The share of general government tax revenues in GDP increased from 35 per cent to 44 per cent between 1970 and 2005, resulting mainly from the growth in social contributions which rose from 12.7 percentage points of GDP in 1970 to 21.0 in 2005. Total revenues from local government also rose sharply over the years as a result of increased financial needs after the implementation of decentralization laws: their share in terms of GDP increased by about 2 percentage points between 1970 and 2005; in 2005, it was equal to 5.5 per cent of GDP. However, their share in general government tax revenues has not changed much between 1970 and 2005, around 10 per cent. Over the whole period under study, the increase in compulsory taxes amounted to 10 percentage points of GDP and concentrated on the financing of social funds, whether through excise duties on tobacco and alcohol or the proceeds of the Contribution Sociale Généralisée (CSG). With deteriorating macroeconomic performance of the French economy during the early 1990s – mainly the rise and persistence of mass unemployment – and the growing imbalance in social security funds, various taxes on individual income were created. The most important one was the CSG: introduced in 1991 and increased in 1993, it is a tax on income from all sources (labour, capital) which aimed at financing social insurance funds.15 Its rate amounts to 7.5 per cent on all incomes (and not only on the sole wages as for social contributions), plus 0.5 per cent for the CRDS (which is a contribution to the reimbursement of the social security debt).16 The proceeds of this tax have increased more than fivefold between 1993 and 2005, from 0.8 to 4.5 points of GDP. This latter figure can be compared with the proceeds of the income tax earned by the central government in 2005: 3.0 percentage points of GDP. As for taxes on goods and services, the decrease in the VAT in 1988, which was undertaken to comply with EC harmonization, was reversed in 1995 after the highest rate rose from 18.6 per cent to 20.6 per cent. This and the resumption of economic growth induced a slight increase in the proceeds of this tax between 1995 and 1997 (of about 0.5 percentage point of GDP). Its proceeds have decreased since then. Taxes on corporate profits were increased twice during the 1990s: in 1995 and 1998. In 1995, the deliberate increase in revenues was of 0.5 billion euros and the total rise in the proceeds of the tax was equal to 1.5 billion euros. In 1998, i.e. on the eve of EMU, the deliberate effort imposed on firms

148 Fiscal Policy in the European Union

amounted to 1 billion euros and the total rise in the proceeds of the tax on corporate profits was equal to 2.6 billion euros. Last, but not least, the reliance on excise duties has never been as huge as in the 1990s. Each year between 1991 and 1993, a new item was being taxed heavier: tobacco in 1991, followed by alcohol in 1992, and oil products in 1993. Since then, deliberate increases in excise duties occurred every year, except in 2002 and 2003. Since 1999, successive governments have also taken measures in order to cut taxes. Under the Jospin government, the local business tax and employers’ social contributions were reduced. As far as households were concerned, they benefited from the decrease in the top-marginal rate of the personal income tax, as well as from the creation of both an earned income tax credit (part of in-work benefit schemes) and the modification of the local income tax for poor households in order to remove the so-called ‘poverty trap’. To boost consumption, this government also reduced the normal VAT rate and awarded a targeted reduced rate for real estate investments. As for the Raffarin and de Villepin governments, they lowered the income tax rate. After three years of constant decrease (from 2002 to 2004), the following Law of Finances enforced a pause in tax decrease. The necessity to curb the public deficit, in order to pass under the 3 per cent of GDP ceiling of the SGP, is the major explanation for this sudden stop. One major challenge for France is tax competition. Indeed, both business statutory and effective tax rates are high in comparison to the EU average (especially when including the new member states) even if they have been on a downward trend for the two last decades (as in other countries): the top marginal rate remains above 60 per cent, compared to 40 per cent in UK, 46 per cent in Italy and the US and 54 per cent in Germany which has recently dramatically cut taxes on income. Several reports have pointed out the necessity of reforming thoroughly the French tax system to meet the tax competition challenge.17 Nevertheless, one step forward was made in the 2006 Law of Finance, where a ‘tax shield’ was introduced and applied for 2007 fiscal revenues: for households, the aggregate direct tax paid to the state and local governments on incomes and wealth should no longer exceed 60 per cent of a taxpayer’s yearly income (50 per cent since 2007 Law of Finance). 4.2 The positive effects of fiscal rules on budget balance: the example of local governments The fiscal health of local governments in France has long been in a better shape than that of the central government18 (see Box 6.2). Two elements

Has France Become a Deficit Running Country? 149

Box 6.2 Deficits and debt of local governments The finances of the local public sector have moved from a permanent deficit over the period from 1980 to the mid-1990s to a permanent surplus ever since, hence reducing local debt. The period 1980–95 coincided with a period of large transfer (the 1982–6 Decentralization Bills) of responsibilities from the central government to the local jurisdictions (36,500 communes (lower-tier), 100 départements (middle-tier), 22 régions (upper-tier)) which led to additional expenditures (mainly capital ones) in sectors which were characterized by an under-provision of public investment. As a consequence, deficits rose and local debt accumulated. In the meantime, the (real) interest rates rose and made the burden of local debt unbearable. The local governments reacted first by restructuring the term-issuance of debt, second by cuts in investment programmes, and then by levying more tax receipts in order to finance the capital expenditures out of their current net savings (Gilbert et al., 2008). Since the mid-1990s and until 2004, local governments have faced fiscal surpluses. The present deficit is likely to vanish in the near future.

can be advocated to explain this feature. First, grants-in-aid received from the central state have greatly increased over the years. But, second, and more importantly, the sustainable behaviour of the local governments comes as the direct consequence of the legal constraints to which they are subject in order to prevent them from fiscal imbalance and bankruptcy. Although no general and permanent rules are explicitly imposed on local governments leading them to adopt a ceiling on expenditures or the capping of the rate of progression in local taxes, all of them are subject to a kind of ‘golden rule’. The latter states that the annual burden of the local debt – repayment of capital and interests – must be financed out of current fiscal surpluses. In other words, borrowings by local governments are earmarked to capital expenditures only; in addition, the debt guarantees provided by the local governments to other public or semipublic bodies are capped in various ways. This ‘rule’ resembles, though at a local level, the national fiscal framework adopted in the UK in 1997 in addition to the Code for Fiscal Stability. The severity of constraints for French local governments is reduced in some special cases: fiscal bail-outs are exceptionally provided to needy local governments, and some negotiations between the local governments as a whole and the central government may lead to additional

150 Fiscal Policy in the European Union

subsidies. Undoubtedly also, the fiscal performance of the local governments has been eased by the sharp increase in the grants received from the central government, and at the expense of the local taxpayers who have faced increasing tax rates. But, on the whole, no formal fiscal bailout can be provided to a local government on a regular basis. The fiscal discipline imposed on the local governments is definitely tighter than the fiscal discipline imposed on the central government (Gilbert and Guengant, 2001). The social security institutions (including unemployment insurance, social insurance, health care, aid to the disabled and the elderly) are also subject to the requirement of balancing their current accounts. Nevertheless, this condition is met through a different process than in the case of local governments because a large part of the social security branch is jointly managed by the ‘social partners’ (mainly representative employers unions and labour unions). Consequently, the debt of the social security institutions is not subject to any specific overall target or ceiling expressed in percentage points of GDP. 4.3 Towards a new governance of public finances In 2001, France started a sound governance reform of its public finances. Since 2006, the public accounts of the central state have been presented and voted by the Parliament in a new way. In the meantime new fiscal institutions have been created which aim at a better compliance of the performance of the French public finances with EMU fiscal requirements. According to an OECD study on the transparency and the ‘best practices’ in terms of public finances within the member countries (OECD, 2000), France was occupying a ‘median’ position as far as fiscal transparency was concerned, but appeared in a less brilliant position with respect to the evaluation of fiscal performance and to the ability to link together the performance and the allocation of financial means among the bureaus. This situation was the legacy of more than 40 years of fiscal conservatism. This immobility came to an end when a reform of the legal framework of the budgetary process was adopted on 1 August 2001 by the Parliament. The purpose of this proposition was to ‘reform the constitutional bylaw since France is worryingly behind with modernization of its public management’ (Migaud, 2001: 37). The parliamentary impulse was in line with initiatives of former ministers of finance since 2000 which tended (i) to enlarge the information available to the Parliament members and in particular the information available to the fiscal commissions members; (ii) to ease the comparison over time of fiscal figures (constant perimeter

Has France Become a Deficit Running Country? 151

of public activities); (iii) to evaluate the net worth of the public administration in a more comprehensive way and (iv) to reinforce the control upon the contracting procedures of the public administrations. The 1 August 2001 constitutional bylaw on Budget Acts was scheduled for gradual implementation in a timeframe running through to 2006. In short, its main objectives were (i) to introduce greater flexibility through presentation of programmes with block appropriations; (ii) to strengthen parliamentary control; (iii) to systematize policy evaluation and to blow performance-based culture in the civil service; and (iv) to make formal documents more transparent and their fairness better guaranteed. The ‘greater flexibility’ of the budget process has permitted a sharp reduction in the waste of resources which ensued from the principle of credit specialization by ‘chapters’: so long as credits were set on an annual basis and could not be reported, bureaus had to spend their remaining credits at the end of each year in order not to lose credit renewal. Until 2001, the means-based logic has been changed to a target-based and performance-based logic. Renewals are no longer automatic, and resources reallocation among different ‘programmes’ within a ‘mission’ is possible and can be initiated by the Parliament. The missions devolved to the state are now regularly defined and are constituted by different ‘programmes’, each of which aggregates several ‘chapters’. Resources reallocation among a programme is also possible: personnel expenditures can be changed into capital expenditures (or vice versa) if it is judged necessary. It is also noteworthy that the change from ‘chapters’ to ‘programmes’, each of which has to be voted by the Parliament, will enhance the transparency of the budget process and will facilitate the control of the Parliament over the whole budget (Chabert, 2004). The fifth and new budgetary principle, fairness,19 should also improve the ability of the Parliament to check the budget. This budgetary principle has thus brought public finance closer to private accounting practices. Last, but not least, the room for manoeuvre of the state as regards the implementation of the Law of Finance has been reduced: during a given year, new credits decided by decrets shall not exceed total credits incorporated in the initial Law of Finance by more than 1 per cent, except in case of ‘emergency and pressing necessity of national interest’; conversely, yearly credit cancellation shall not exceed total credits by more than 1.5 per cent. All these reforms are meant to bring ex ante deficits nearer to ex post public deficits, except in case of a sharp unexpected drop or rise in economic growth. On this latter topic, the European context and the

152 Fiscal Policy in the European Union

enforcement of the SGP are prominent vectors of further reforms in the budget process. Indeed, under the pressure of the European integration process, the necessity of a multi-annual spending target stated in real terms has spread in France. Successive three-year fiscal programmes (Programmes pluriannuels de finances publiques) have been associated with the SGP. These programmes provide quantitative targets and ceilings for the fiscal policy. Even if the objectives stated by these programmes have not been met every year in the recent past, they have the advantage of highlighting the errors of expectations and therefore trigger discussion on the optimal fiscal stance and the optimal way to fulfil this stance. More recently, a national conference on public finances (Conférence Nationale des finances publiques), created in 2006 and gathering the main actors of public finances (central state, local governments, social security branch), aims at surveying the present state of the public finances of the general government, at providing general guidelines for the fiscal year to come in relation to EMU requirements, and the three-year fiscal programmes mentioned above, and at stating a clear sharing of fiscal responsibilities in terms of current deficits, spending and debt between the three tiers of government.

5 Conclusion Determining the extent to which France is able to move quickly on the way to reforms remains an open question. On the one hand, the recent history of public finances has shown that sound reforms have been implemented over a rather short time. Fiscal reforms which started in the mid-1990s were definitely more ambitious than those experienced over the last three decades even if it is too early to estimate their effects. But on the other hand, history shows also how difficult it is to introduce and monitor in due time (fiscal) reforms: for a French observer, the ‘rhythm’ of the recent reforms has been exceptionally rapid as the tension resulting from the reforms of public pensions showed in 2007; but, from an international angle, the speed of reforms in France may still be considered slow. In this chapter, it has been shown that French governments have rarely stopped implementing restrictive policies. Nevertheless, French public finances are in deficit. In fact, three major economic determinants have brought France’s public deficits and debts to their present peaks: high real interest rates, low economic growth and pro-cyclical policies. Although reforming the budget process should have no impact on the first two

Has France Become a Deficit Running Country? 153

determinants, it might considerably improve the transparency of the whole budget. Meanwhile, it should limit the ability of the central government to elude the control of the Parliament. In sum, pro-cyclical policies and their possible corollary – political cycles – should more or less disappear and fiscal policy might gain efficiency.

Notes 1. We are very grateful to Jérôme Adda whose remarks and comments have helped us to substantially improve this chapter. The usual disclaimer applies. This study is part of project ANR – FRAL 022 – Local Rules, financed by Agence Nationale de la Recherche. 2. Some clarifications of the notion of public debt are necessary at this stage. (a) Gross public debt (or gross general government debt) is obtained by aggregating the debt of the three main branches of government (central government, local governments and social security funds) as defined by national accounting. (b) The net debt is equal to the gross debt, net of assets owned by the public sector. (c) The public debt according to Maastricht definition is a gross public debt. It differs from the national accounting definition in that it is (i) a consolidated debt, (ii) put into nominal term (iii) which leaves out some kinds of borrowing such as business credits. On average, the debt according to Maastricht definition is below gross public debt by 15 percentage points of GDP. 3. Over the 1970s, the net public debt was almost equal to zero, and even negative since general government was a net creditor from 1970 to 1975 as well as in 1979 and 1980. 4. Despite a transitory slowdown at the turning point of 1998–9 due to the so-called emergent countries’ crisis. 5. Quintos (1995) found responses that ranged between a 0.6 and 0.8 percentage point of GDP. 6. Some arithmetic of the government budget constraint needs to be introduced here. The flow budget constraint of the general government may be written as: dt = (1 + rt − gt − πt )dt−1 − pst , where d is the stock of public debt in percentage of GDP, r the nominal long-term interest rate, g the GDP growth rate, π the inflation rate, ps the primary surplus in percentage of GDP, and t time proxy. Stability of the debt to GDP ratio requires that: pst = (rt − gt − πt )dt . 7. Under the Jospin government (1997–2002), taxes continued to be lowered; this policy could well be attributed to President Chirac’s insistence that his promise during the presidential election of 1995 had to be honoured. 8. As will be discussed further in the following, this statement has lost most of its relevance (see Siné, 2006). 9. Computations excluded election year 1997: the decision by President Chirac to dissolve Parliament was taken suddenly in early spring. In this context, opportunistic behaviours could not occur. 10. This is the part which is not corrected for in the cyclically adjusted primary balance (see discussion in section 3.1).

154 Fiscal Policy in the European Union 11. It is worth making it clear that present statements are related to the explanation of fiscal deficits. They do not assume efficiency or inefficiency of fiscal policies. In the case of Raffarin and de Villepin’s governments, we argue that deficits have had a substantial discretionary component. They may prove efficient in the future. 12. Garcia and Hénin (1999) added a ‘perverse effect’ to tax increases: higher fiscal margins for manoeuvre can be used to raise public spending. They claimed that increased revenues produced higher public spending in France, Germany and the United States. 13. In France, the health system for instance operates mostly as a public insurance system even though a large part of health care supply is provided by private doctors (médecine libérale). 14. See Creel and Sterdyniak (2006: table 3). 15. The CSG is levied at source on wage income and interest received and was initially not deductible from the personal income tax base. Its rate was 1.1 per cent until 1993; 2.4 per cent until 1996. In 1996 and 1997, a reform of the financing of social security gave rise to an extension of the CSG to financial incomes and the replacement of health contributions by a deductible CSG. 16. It was created in 1996 and was meant to be ‘temporary’. 17. For example, Saint-Etienne and Le Cacheux (2006). 18. A micro-analysis of the accounts confirms this assessment; more than 95 per cent of the ‘communes’ (municipalities) finances are fiscally sustainable. This proportion is even larger for the ‘départements’ and the ‘régions’ (Gilbert and Guengant, 2001). 19. The four other principles are: yearly budgeting, unity (public spending and taxes are presented in a single document), universality (resources are equal to charges) and specialization (formerly by ‘chapter’, now by ‘programme’).

References Alesina, A. and R. Perotti, ‘Fiscal Expansions and Adjustments in OECD Economies’, Economic Policy, 21 October (1995): 207–48. Blanchard, O. J., ‘Suggestions for a New Set of Fiscal Indicators’, in H. A. A Verbon and F. A. A. M. van Winden (eds), The Political Economy of Government Debt (Amsterdam: Elsevier Science Publishers, 1993). Brender, A. and A. Drazen, ‘Political Budget Cycles in New versus Established Democracies’, Journal of Monetary Economics, 52(7) October (2005): 1271–96. Brender, A. and A. Drazen, ‘Political Implications of Fiscal Performance in OECD Countries’, paper presented at the 8th Workshop on Public Finance, Banca d’Italia, Perugia, 30 March–1 April (2006). Burns, A. and A. Goglio, ‘Public Expenditure Management in France’, OECD Economics Department, Working Paper 409, October (2004). Buti, M. and M. P. van den Noord, ‘Discretionary Fiscal Policy and Elections: the Experience of the Early Years of EMU’, OECD Economics Department, Working Paper 351, March (2003). Chabert, G., ‘La loi organique relative aux lois de finances (LOLF): du 1er août 2001’, Regards sur l’actualité, La réforme budgétaire, 297, Paris: La documentation Française, January (2004).

Has France Become a Deficit Running Country? 155 Creel, J. and H. Le Bihan, ‘Using Structural Balance Data to Test the Fiscal Theory of the Price Level: Some International Evidence’, Journal of Macroeconomics, 28(2), June (2006). Creel, J. and H. Sterdyniak, ‘Les déficits publics en Europe’, Revue de l’OFCE, 54, July (1995). Creel, J. and H. Sterdyniak, ‘Faut-il réduire la dette publique?’, Lettre de l’OFCE, 271, January (2006). Fève, P. and P. Y. Hénin, ‘Assessing Effective Sustainability of Fiscal Policy within the G-7’, Oxford Bulletin of Economics and Statistics, 62 (2000). Gali, J. and R. Perotti, ‘Fiscal Policy and Monetary Integration in Europe’, Economic Policy, October (2003). Galli, E. and F. Padovano, ‘A Comparative Test of Alternative Theories of the Determinants of Italian Public Deficits (1950–1998)’, Public Choice, 113 (2002): 37–58. Garcia, S. and P. Y. Hénin, ‘Balancing Budget through Tax Increases or Expenditure Cuts: Is It Neutral?’, Economic Modelling, 16 (1999): 591–612. Gilbert, G. and A. Guengant, ‘The Control of Local Public Deficits and Local Debt’, in B. Dafflon and J. Dafflon (eds), Local Public Finance: Balanced Budget and Borrowing (London: Edward Elgar, 2001). Gilbert, G., A. Guengant and C. Tavéra, ‘Les collectivités locales peuventelles restaurer leur capacité de financement? Les enseignements du modèle macroéconomique APUL’, Economie et Prévision (2008, forthcoming). Girouard, N. and C. André, ‘Measuring Cyclically-adjusted Budget Balances for OECD Countries’, OECD Economics Department, Working Paper 434 (2005). Hamilton, J. and A. Flavin, ‘On the Limitations of Government Borrowing: a Framework for Empirical Testing’, American Economic Review, 76(4) (1986): 808–19. Hughes-Hallett, A., J. Lewis and J. von Hägen, Fiscal Policy in Europe, 1991–2003: an Evidence-based Analysis (London: CEPR, 2003). INSEE, ‘La dette publique en France: la tendance des vingt dernières années est-elle soutenable?’, L’économie française 2004–2005 (Paris, 2006). Laurent, E. and J. Le Cacheux, ‘Integrity and Efficiency in the EU: the Case against the European Economic Constitution’, Center for European Studies, Harvard University, Working Paper Series 130 (2006). Le Cacheux, J., ‘Politiques de croissance en Europe, un problème d’action collective’, Revue économique, 56(3), May (2005). Migaud, D., ‘Moderniser la gestion publique et renforcer le pouvoir budgétaire du parlement’, Revue Française de Finances Publiques, Special issue, 73, January (2001): 37–44. OECD, OECD Best Practices for Budget Transparency, PUMA/SBO(2000)6/FINAL (2000). Perotti, R., ‘Fiscal Policy in Good Times and Bad’, Quarterly Journal of Economics, 114(4) (1999): 1399–1436. Quintos, C. E., ‘Sustainability of the Deficit Process with Structural Shifts’, Journal of Business and Economic Statistics, 13(4), October (1995). Roubini, N. and J. Sachs, ‘Political and Economic Determinants of Budget Deficits in the Industrial Democracies’, European Economic Review, 33 (1989): 903–38.

156 Fiscal Policy in the European Union Saint-Etienne, C. and J. Le Cacheux, ‘Croissance équitable et concurrence fiscale’, Rapport du Conseil d’Analyse Economique (Paris: La Documentation française, 2006). Siné, A., L’ordre budgétaire: L’économie politique des dépenses de l’Etat (Paris: Economica, 2006). Vavouras, I. S., ‘Public Sector Deficits: Their Economic and Policy Determinants in the Case of Greece’, Journal of Policy Modeling, 21(1) (1999): 89–100. Wilcox, D., ‘The Sustainability of Government Deficits: Implications of the Present-value Borrowing Constraint’, Journal of Money, Credit and Banking, 21(3) (1989): 291–306.

7 How to Deal with Economic Divergences in the EMU? Catherine Mathieu and Henri Sterdyniak

1 Introduction Before the launch of the euro, the proponents of European Economic and Monetary Union (EMU) thought that the single currency would lead to a rapid economic convergence among member states. Smaller country specificities would have reduced the need for domestic fiscal policies and the conduct of short-term economic policy could have been handed over mostly to the ECB. Fiscal binding rules would then have been justified. The macroeconomic success of EMU would have helped the convergence of industrial and social policies towards a liberal model (more labour market flexibility and product competition, reduction of the role of the state and public sector and less welfare spending). In return, the successes of these structural policies would have facilitated the coordination of stabilization policies. In fact there have been persistent and sometimes growing divergences between euro area countries in terms of output growth, inflation, unemployment and external balances since 1999. The single currency imposed similar macroeconomic policies in countries in different situations and has widened growth disparities among member states. EMU made winners (Ireland, Spain, Greece) and losers (Germany, Italy, the Netherlands, Portugal). The institutions and the rules set out by the Treaties (the ECB, the Stability and Growth Pact) have been unable to tackle these divergences. European institutions have tried to introduce structural reforms (through the broad economic policy guidelines (BEPGs), the open method of coordination (OMC) or the Lisbon Agenda) but have faced national specificities or people’s opposition. In most economies and especially in the bigger ones, the introduction of the euro did not induce the promised acceleration of growth. In other 157

158 Fiscal Policy in the European Union

countries, the acceleration of growth generated hardly sustainable imbalances. Member states have been unable to set out a common growth strategy. They have not questioned the ECB’s remit and SGP rules. Euro area countries are widely homogeneous in terms of high taxation rates and Bismarckian social protection systems. However, they have been unable or unwilling to maintain this specificity at the EU level. They have hesitated between two strategies: a social-Keynesian one with a strong commitment to maintaining a specific European Social Model (ESM) and a proactive industrial policy; a free market strategy, based on market deregulation and reform of the ESM through public expenditure cuts and a smaller role for the state. European institutions recommended liberal strategies that did not meet people’s expectations, albeit lacking the democratic legitimacy needed to impose such measures. This weakened European construction. The move towards more flexible markets has been questioned (for instance with the non-adoption of the Bolkestein Directive). Some countries were tempted to renationalize industrial policy, while most EU countries opposed the implementation of European social or fiscal policies. This debate takes place at a time when European continental countries especially need to adapt to globalization: should they move towards a liberal or a Scandinavian model? Should this choice be made at European or national levels? Section 2 provides an overview of disparities in terms of output growth, inflation and unemployment in the euro area. The widening of some economic imbalances (current account and government deficits, competitiveness) is highlighted. Section 3 addresses four reasons for persistent and rising disparities: the benefits of the euro area for catching-up countries; the weaknesses of the euro area economic policy framework; the implementation of non-cooperative domestic policies which have induced excessive competition, insufficient coordination and have been detrimental mainly to the larger economies; and the crisis of the continental model. Section 4 discusses strategies aiming at reducing disparities in Europe: increasing market flexibility; moving towards the knowledge society of the Lisbon Agenda; renationalizing economic policies; or introducing a more growth-oriented European economic policy framework.

2 The euro area: disparities and lost illusions GDP growth was relatively satisfactory in the euro area between 1985 and 1991 (+3.1 per cent per year, see Table 7.1). GDP growth decelerated by 1.3 percentage points per year from 1992 to 1998 due to bad

Economic Divergences in the EMU

159

Table 7.1 GDP growth rates and GDP per head 1985–1991

Euro area Belgium Germany Greece Spain France Ireland Italy Netherlands Austria Portugal Finland Denmark Sweden UK US

3.1 2.7 3.5 1.7 3.9 2.6 4.0 2.9 3.6 3.1 5.1 1.8 1.5 1.9 2.6 2.8

1992–1998

1.8 1.8 1.5 1.8 2.3 1.8 7.2 1.3 2.7 2.2 2.4 2.5 2.7 2.7 2.7 3.6

1999–2005

1.9 2.0 1.3 4.3 3.6 2.2 6.5 1.2 1.7 2.0 1.6 2.8 1.9 2.9 2.7 3.0

PPP GDP per head, euro area = 100 1991

2005

100.0 108.7 108.9 67.0 79.2 104.2 78.8 105.3 107.0 113.8 68.6 97.6 106.7 108.2 93.6 131.1

100.0 111.1 101.5 78.9 92.8 102.7 130.6 97.6 117.4 114.8 67.2 108.7 116.2 111.6 110.4 139.7

Source: European Commission.

management of German reunification and to contractionary fiscal policies implemented in the convergence process to meet the Maastricht criteria. The launch of the single currency in 1999 did not enable the area to reach a more satisfactory growth. Since 1991, GDP has grown less rapidly in the euro area than in the UK or in the US (1.8 per cent per year, versus respectively 2.7 and 3.3 per cent). Since 1999, GDP growth has remained strong in Ireland and has accelerated in two countries only – Spain and Greece – although this has led to a rise in current account deficits. Looking at average GDP growth rates in 1999–2005 and 1985–91, the main losers are Germany, Italy, Portugal and the Netherlands. Greece and Spain have been converging towards the area average in terms of GDP per head (in PPP) while Portugal and Italy have been diverging downwards and Ireland upwards: in 14 years (from 1991 to 2005), the GDP per head relative to the euro area rose by 65 per cent in Ireland, 18 per cent in Greece, and 17 per cent in Spain whereas it declined by 1.5 per cent in Portugal. Among the largest economies, the GDP per head relative to the euro area declined by 7 per cent in Germany and Italy, and by 1.5 per cent in France, whereas it rose

160 Fiscal Policy in the European Union

by 18 per cent in the UK. Non-euro area EU countries performed better than euro area ones. Many economists consider that euro area growth is low because potential growth is low and that disparities in domestic GDP growth rates reflect disparities in domestic potential growth (see for instance, Benalal et al. 2006). According to the European Commission and OECD estimates, the euro area potential growth rate was 2.0 per cent between 1998 and 2005, hence very close to observed growth. From that point of view, if demand had been more robust, the outcome would have been higher inflation only. Inflation has remained almost stable in the euro area since 1996, which would mean that demand was roughly equal to potential output, either by chance or thanks to an appropriate monetary policy. Output could have been increased only through structural measures: higher productivity growth (due to capital accumulation, higher R&D or education spending), higher labour supply (through immigration, longer working time and lower female and older people inactivity rates), or lower equilibrium unemployment rate (through increased labour market flexibility). In our view, output growth is not determined by but has an impact on productivity gains, capital accumulation, participation rates, equilibrium unemployment rates and even population (through immigration). What does ‘potential growth’ mean for countries suffering from mass unemployment and low participation rates? When unemployment rates are high, older people and female participation rates decrease, either without or with policy; companies have no incentive to raise labour productivity. Thus contractionary fiscal policies, high real interest rates or low competitiveness may induce low output growth for several years. This will not reflect low potential growth that would materialize independently of observed growth. From 1998 to 2005, unemployment rates decreased slightly in the majority of euro area countries but rose in Germany, Portugal, and also in the Netherlands and Austria, although remaining at a low level (see Table 7.2). Unemployment rates fell rapidly in four countries: Ireland and Finland, thanks to robust GDP growth, Italy thanks to low labour productivity growth, and in Spain thanks to both growth and low labour productivity growth. However, in 2005, eight euro area countries remained in a mass unemployment situation. These countries account for 90 per cent of the area and they could have tried to introduce policies to support growth and employment, especially as inflation was moderate (2.2 per cent) and the current account was in balance. But their priority was instead to implement structural reforms.

Economic Divergences in the EMU

161

Table 7.2 Unemployment and employment rates Unemployment rate, %

Euro area Belgium Germany Greece Spain France Ireland Italy Netherlands Austria Portugal Finland Denmark Sweden UK US

Employment rate, % (full time equivalent)

Labour productivity growth rate, %

1998

2005

2005

1999–2005

10.1 9.3 8.8 10.9 15.0 11.1 7.5 11.3 3.8 4.5 5.1 11.4 4.9 8.2 6.1 4.5

8.6 8.4 9.5 9.8 9.2 9.5 4.3 7.7 4.7 5.2 7.6 8.4 4.8 7.8 4.7 5.1

59.8 57.1 60.4 59.2 60.9 59.7 64.6 55.5 60.9 63.7 65.6 65.3 69.4 68.0 65.4 67.0

1.0 1.2 1.6 3.5 0.3 1.1 2.8 0.4 1.3 1.5 0.8 1.5 1.6 1.9 1.7 2.2

Source: OECD.

In terms of employment rates, Italy stands clearly below the other euro area countries. Belgium comes next. The other countries can be split into two groups: medium rate countries (Germany, Greece, Spain, France and the Netherlands) and high rate countries (Ireland, Austria, Portugal, Finland and, outside the area, Sweden, Denmark and the UK). High rate countries include Scandinavian and liberal countries and also Austria. Medium rate countries would have to increase their employment rates by almost 10 per cent to reach the UK level (18 per cent for Italy). This will be a challenge for economic policies in the years to come, especially in the prospect of financing higher pension expenditures. However, a crucial issue remains: should demand be increased first in order to increase labour demand, hoping that labour supply will follow? Structural reforms like the abolition of early retirement or a better control of unemployment allowances would be implemented only ex post and only in countries where labour demand is clearly above supply. Such a strategy would raise a risk of inflationary pressures. Or should countries where unemployment is high start with structural reforms which would raise labour supply and employment? Such a strategy would provoke

162 Fiscal Policy in the European Union Table 7.3 Inflation and real interest rates

Euro area Belgium Germany Greece Spain France Ireland Italy Netherlands Austria Portugal Finland Denmark Sweden UK US

Price level

Inflation (GDP deflator)

Real interest rate less GDP growth rate

2005

1999–2005

1992–1998

1999–2005

100.0 101.1 104.9 81.4 88.6 103.8 116.3 97.3 103.5 101.9 81.5 109.5 129.7 112.1 105.5 93.9

1.8 1.7 0.7 3.6 3.9 1.4 3.9 2.5 2.7 1.5 3.4 1.2 2.3 1.4 2.5 2.2

2.5 1.6 1.6 6.7 2.1 2.9 −3.5 3.9 0.9 1.3 1.6 1.3 2.5 1.7 3.7 −0.1

−0.6 −0.6 1.1 −3.3 −4.4 −0.5 −7.3 −0.6 −1.4 −0.4 −1.7 −0.9 −0.9 −0.8 0.3 −2.4

Source: European Commission.

the risk of raising unemployment and increasing poverty among the unemployed. Many countries suffering from high unemployment rates have chosen to give firms incentives to reduce labour productivity growth and to hire unskilled workers (Spain, Italy, Belgium and France). Other countries (Germany) tried instead to increase employment through competitiveness gains, especially through higher productivity growth in the industry. The European employment strategy and the Lisbon Agenda failed to impose a common strategy. A good functioning of the monetary union requires avoiding disparities in terms of price levels. Different price levels will generate competitiveness differentials and will need to be corrected later through output growth differentials. In practice, inflation differentials have remained substantial in the euro area (see Table 7.3). Countries running higher inflation were mainly catching-up ones, with higher output growth and low initial price levels, due to the Balassa-Samuelson effect (Greece, Ireland, Spain, Portugal). However, Italy and the Netherlands also had relatively high inflation rates. The Dutch economy ran above capacity for several years and inflation was increased by several rises in

Economic Divergences in the EMU

163

indirect taxes. Inflation was not due to excessive demand levels induced by excessive public deficits in any of these countries. Even when accounting for the Balassa-Samuelson effect, which may explain 1 percentage point of inflation in Greece, 0.7 in Portugal and 0.5 in Spain (for a discussion, see ECB, 2003), prices seem to have risen too rapidly in these three countries and this has led to price competitiveness losses. Inflation has been extremely low in Germany, which has prevented other countries from restoring their price competitiveness. In 2006, inflation disparities remained large in the euro area: inflation was 0.9 per cent in the three countries running the lower inflation, and 3.2 per cent in the countries running the higher inflation. Wage and price formation processes have not converged yet. Another difficulty in a monetary union is that catching-up countries have structurally higher output growth and inflation rates than more ‘mature’ countries. Thus it is difficult to run a single monetary policy even in the absence of asymmetric shocks. With a single nominal interest rate, euro area countries have had different real interest rates corrected for growth (see Table 7.3). The single monetary policy was contractionary for Germany and expansionary for Ireland, Greece and Spain where companies and households had a strong incentive to borrow and invest, which boosted domestic GDP growth and inflation. The euro area interest rate was 3.1 per cent on average from 1999 to 2005. A Taylor rule based on an inflation target of 2 per cent and output gaps estimated by the OECD, would have suggested average interest rates of 1.75 per cent for Germany, 3.05 per cent for France, 3.8 per cent for Italy, 4 per cent for the euro area and 8.05 per cent for Spain. However, the 2 per cent inflation target may be judged too low. The euro area needs higher GDP growth and this may result in transitory inflationary pressures on some markets. The OECD potential output estimates are very low: the euro area output gap was estimated to be nil in 1999, when the unemployment rate was 9.2 per cent and nil also in 2002, when the unemployment rate was 8.3 per cent. The ECB is less concerned with GDP growth than the Fed. Since 1999, the monetary stance has been clearly more expansionary than suggested by a Taylor rule in the US; more accommodative in the euro area too, although to a smaller extent and restrictive in the UK (see Table 7.4). The wage share in GDP decreased at the euro area level from 2.1 points between 1999 and 2005. Real wages increased by a mere 0.35 per cent per year in Germany, 0.5 per cent in Austria and Italy, 0.6 per cent in Belgium, while they rose by 1.3 per cent in France, 1.5 per cent in the Netherlands and 2.5 per cent in the UK. Increasing company profitability and price

164 Fiscal Policy in the European Union Table 7.4 Taylor rules and effective central bank rates

Euro area Taylor based rate Actual rate UK Taylor based rate Actual rate US Taylor based rate Actual rate

1999

2000

2001

2002

2003

2004

2005

2006

2.5 3.1

4.8 4.5

5.3 4.3

4.55 3.3

3.55 2.3

3.75 2.1

3.5 2.2

3.7 3.1

3.2 5.6

3.05 6.2

3.45 5.0

3.3 4.1

3.4 3.7

3.5 4.6

4.0 4.8

4.05 5.0

4.6 5.4

7.15 6.5

5.2 3.8

2.75 1.8

3.75 1.2

4.8 1.6

6.35 3.5

6.15 5.2

Note: The rule is defined as: i = 2 + 0.5(2 − π) + 0.5 gap, where i is: central bank’s interest rate, π: inflation, gap: output gap as estimated by the OECD. Source: Authors’ calculations.

competitiveness through downwards pressure on wages became a major strategy in several countries (Germany, Austria). It was the only tool available for countries which could neither depreciate their currency nor cut their interest rate nor use fiscal policy once the 3 per cent of GDP limit for government deficits had been breached. Moreover, firms could threaten to relocate their production abroad in order not to raise wages. Last, fixed exchange rates (at least in the euro area) ensured that the effect of wage moderation would not be cancelled by exchange rate appreciation. This strategy helped exports but put a drag on private consumption in the countries where it was implemented consequently dampening demand in the whole euro area. No attempt was made by the member states, the trade unions or the European Commission to harmonize wage growth. In this non-cooperative game, Germany, Austria and the Netherlands succeeded in supporting their GDP growth through a positive contribution of net exports (by around a 0.5 percentage point of GDP per year). In contrast, Spain and France suffered from a negative external contribution (0.7 percentage points of GDP per year). In terms of domestic and external demand, euro area countries can be divided into four groups: the ‘winners’ (Ireland, Spain, Greece), where both domestic and external demand is strong; the ‘bad guys’ (Germany, Austria, Netherlands), where a weak domestic demand is offset by strong export demand gains; the ‘losers’ (Italy, Portugal) suffering from both low domestic and external demand; and the ‘victims’ (France, Belgium, Finland) where a weak external demand partly offsets a satisfactory domestic demand.

Economic Divergences in the EMU

165

The euro area as a whole won competitiveness from 1996 to 2001 thanks to the fall in the euro vis-à-vis the US dollar. A weak euro together with the NITC bubble accompanied strong GDP growth (3 per cent per year from 1997 to 2000) and employment (8.7 per cent in five years). This shows that the European economy can grow rapidly if demand is robust. Competitiveness gains were more than cancelled by the euro’s appreciation vis-à-vis the US dollar and Asian currencies from 2001 and 2004. The euro area needs a weaker exchange rate in the light of high unemployment. The euro area has been able to have a low exchange rate only when US domestic demand was strong, because the US then also had an interest in a high dollar. But the euro vis-à-vis the dollar is always high when US domestic demand is relatively weak. The euro area suffers from a less active monetary policy than in the US. Last, the euro area suffers from exchange rate policies in Asian countries, where exchange rates are kept low to support a fragile GDP growth ( Japan), to support export growth (China, new industrial economies) and to accumulate foreign currency reserves. From 1988 to 1999, some European countries had succeeded in depreciating in their currency in real terms vis-à-vis the Deutschmark: Finland, Italy, Spain and outside the future euro area, Sweden. Germany, Austria and Portugal joined the euro area at too high exchange rates which induced substantial current account deficits. Since 1999, Austria and Germany have succeeded in restoring their competitiveness through wage moderation policies. Italy seems unable to maintain its competitiveness in the absence of exchange rate devaluation. Italy and Portugal have been more affected than the average area by the emergence of China. Fixed exchange rates and rigid inflation rates induce persistent exchange rates misalignment periods. In the euro area countries can no longer devalue their currency to restore their competitiveness. Wage moderation policies are the only tool left but take a long time to play and are painful, since they depress demand both at home and in the area. Wage moderation policies would be all the more difficult to implement in euro area countries since they are already implemented in Germany, where domestic inflation is very low which makes it harder for partner countries to gain competitiveness against Germany. Non-coordinated policies have increased imbalances within the euro area: in 2005, a few countries ran substantial current account surpluses: the Netherlands (6.6 per cent of GDP) and Germany (4.2 per cent), whereas some others ran large deficits: Portugal (−9.3 per cent of GDP), Spain (−7.4 per cent) and Greece (−7.9 per cent) (see Table 7.5). The 160

166 Fiscal Policy in the European Union Table 7.5 External positions and savings ratios

Euro area Belgium Germany Greece Spain France Ireland Italy Netherlands Austria Portugal Finland Denmark Sweden UK US

Competitiveness*

Current account balance, % of GDP

National savings rate

Households’ net savings ratio

1998

2005

1998

2005

2005

2005

108 104 101 89 108 101 92 114 105 95 97 127 101 110 92 94

100 96 109 77 95 100 107 95 96 114 73 122 89 132 86 108

0.8 5.2 −0.8 −3.0 −1.2 2.6 0.8 −3.0 3.2 −3.1 −7.1 5.6 −0.9 3.9 −0.4 −2.4

0.0 2.5 4.2 −7.9 −7.4 −1.7 −2.6 −1.1 6.6 1.2 −9.3 5.2 2.9 6.0 −2.2 −6.4

20.6 23.7 22.1 15.8 16.9 19.9 25.5 20.5 24.0 22.7 15.7 19.0 23.8 22.9 15.4 14.5

n.a. 6.2 10.7 n.a. 5.4 11.6 n.a. 9.5 5.7 9.5 4.9 n.a. −5.8 n.a. 0.0 −0.4

* 1988 = 100; on the basis of unit labour costs in the manufacturing sector. A rise means competitiveness gains. Source: OECD.

billion euro surplus of Germany and the Netherlands finances the 145 billion euro deficit of Mediterranean countries. Do these divergences in current accounts reflect an equilibrium process (oldest countries’ savings being invested in younger and more profitable countries) or a disequilibrium one (European savings being spoiled in non-profitable investment, such as housing, in Southern countries)? This situation cannot be considered as optimal since real interest rates corrected for output growth differ across the area. Deficits can widen because they are not financed by financial markets but by transfers within the European banking system and hence can hardly be visible. Foreign direct investments (FDI) cover only a small part of these deficits: Portugal receives a small amount of net FDI (1 per cent of GDP in 2005), but net FDIs are negative for Spain (−1.4 per cent of GDP) or Greece (−0.4 per cent). National savings rates are very low in Greece, Spain and Portugal which is unusual for countries growing at a rapid rate. Output growth is strong in Greece, Spain, the UK and in the US too, where both national and households’ savings rates are very low. In contrast, Belgium, Germany, Austria and France suffer from too high

Economic Divergences in the EMU

167

Table 7.6 Fiscal policies As a percentage of GDP

Euro area Belgium Germany Greece Spain France Ireland Italy Netherlands Austria Portugal Finland Denmark Sweden UK US

Government balance, 2005

Fiscal impulse* 2000/2004

Gross public debt, Maastricht definition, 2005

Net public debt, 2005

−2.4 0.0 −3.2 −5.1 1.1 −2.9 1.1 −4.3 −0.3 1.3 −6.0 2.5 4.6 2.8 −3.4 −3.7

1.2 0.6 1.1 7.1 −0.4 1.5 2.8 2.4 0.0 −2.7 1.3 4.9 −0.4 3.1 5.6 6.0

71 93 68 109 43 67 28 106 53 63 64 41 36 50 44 65

55 86 58 96 31 44 10 99 38 42 45 −60 7 −12 41 46

* The fiscal impulse is the change in the primary cyclically adjusted government balance. A positive figure reflects an expansionary policy. Source: OECD.

savings rates. Low savings rates seem necessary to have high GDP growth and low public debt. Virtue is dangerous in Europe, since the weakness of domestic demand resulting from a high savings ratio cannot be offset by low interest rates or substantial government deficits. In the last recession (2000–4), GDP growth was hardly supported by fiscal policies in the euro area: the fiscal impulse was 1.2 percentage point of GDP only, as compared to 5.6 in the UK and 6 in the US (see Table 7.6). Except for Greece and Finland, euro area countries implemented close to neutral fiscal policies. These years of low growth and rising fiscal deficits generated tensions between European authorities and national governments. In 2005, five euro area countries and seven non-euro area countries were under an excessive deficit procedure (EDP) (Mathieu and Sterdyniak, 2006). The SGP was a corset for fiscal policies in these years. The objective of bringing debts to below 60 per cent of GDP was not fulfilled: government debts still stand at around 100 per cent of GDP while the French and German debts have risen above 60 per cent of GDP. However, public finances were more sustainable in the euro area than in the US, the UK or Japan in 2006 (see Table 7.7). Public finances have

168 Fiscal Policy in the European Union Table 7.7 Public finance sustainability in 2006 As a percentage of GDP

US Japan Germany France Italy Portugal Greece UK

Structural deficit

3.7 5.3 2.1 2.1 3.6 2.8 3.5 3.1

Output gap

0.6 0.5 −1.7 −1.7 −1.3 −4.1 1.2 −0.8

Limit for govt. deficit sustainability* Strict

Soft

3.0 1.0 1.7 1.8 1.9 2.3 3.3 2.0

3.3 3.0 2.7 2.8 4.3 4.0 7.0 2.5

Note: * Government deficit is considered sustainable if the structural deficit is lower than desired public debt times the sum of potential output growth and targeted inflation. The limit for government balance is calculated under two criteria: a strict one, where the level of desired debt is either the current observed level or 50 per cent of GDP (for countries where debt is above this limit); a softer criterion where the level of desired debt also corresponds to the observed level or 50 per cent of GDP (for countries where debt is below this limit). Moreover, under the softer criterion, the structural balance is assumed to include discretionary measures for an amount in terms of GDP corresponding to 25 per cent of the output gap, if the latter is negative. Public finances are unsustainable if the public deficit exceeds the limit, even under the more favourable option: this is the case for countries in italics. Source: OECD, own calculations.

deteriorated in the euro area because of a persistent negative output gap. Fiscal consolidation is not urgently needed in euro area economies. European institutions put too strong a weight on public deficits relative to growth issues. The majority of euro area countries, except Spain and Ireland, have high public expenditure levels, standing at above 50 per cent of GDP. At the euro area level, the share of primary structural public spending was the same in 1990 and 2006. The share has risen in several countries: Portugal (by 10 percentage points), Belgium (3.5) and France (2.5) and outside the euro area, the UK (by 3 percentage points). The share has fallen in other countries: the Netherlands (by 5 percentage points), Spain (4), Ireland (2.5), and also in Sweden (5). The European Commission has failed to impose public expenditure cuts in the European Union and countries have hardly converged. High levels of public expenditure require high levels of taxation. But, as the Scandinavian example shows, high tax-to-GDP ratios are consistent with high employment rates. The expression ‘European Social Model’ is generally used to refer to an original economic and social framework in EU countries but European

Economic Divergences in the EMU

169

social protection systems differ in many respects. They are generally characterized according to four systems (Esping-Andersen, 1990), even if this classification raises many issues: • The Scandinavian model, with a very high level of social expenditure

based on citizenship, funded through taxation, a high female employment rate, low social inequalities and strong cooperation between social partners (Finland, the Netherlands and also Denmark and Sweden). • The continental model, with a high level of social expenditure, based mostly on activity, financed by social contributions, with a high level of labour protection (Germany, France, Belgium, Austria). • The Mediterranean model with an intermediate level of social expenditure based on activity, financed by social contributions, a low level of family and unemployment benefits offset by family solidarity, low female participation rates, and a high level of labour protection (Italy, Greece, Spain, Portugal). • The liberal model with a low level of social expenditure, based on citizenship, targeting the poorer, and with a low level of labour protection (Ireland and also the UK). Replacement ratios derived from unemployment benefits are high in most euro area countries, with the exception of Mediterranean ones (Spain, Greece, Italy), where unemployment rates are not particularly low. Employment protection is strong in continental and Mediterranean countries, as compared to liberal and some Scandinavian countries (according to the OECD employment protection legislation (EPL) indicator; see Table 7.8). Table 7.9 summarizes the economic situation of EU-15 countries using a global index of imbalances based on growth, inflation and unemployment rates and on public and external balances. Some countries have no major macroeconomic problem (Denmark, Sweden, the Netherlands, Finland, Austria and Ireland). Some countries suffer mainly from insufficient GDP growth (Germany, France, Italy, Belgium and Portugal). Some countries have mainly current account deficits problems (Spain, UK, Greece and Portugal). The smaller countries seem to have performed better than the bigger ones, older countries better than catching-up ones, outside the euro area better than euro area countries. The Scandinavian model exhibits the best performance; the Mediterranean model the poorest one (see Table 7.10).

170 Table 7.8 Employment protection indicators

Belgium Germany Greece Spain France Ireland Italy Netherlands Austria Portugal Finland Sweden UK Denmark Japan US

Social protection public expenditures, % of GDP, 2005

Net replacement rate, 2004

EPL, 2003

30.1 30.2 23.8 22.1 33.4 20.1 20.9 24.8 29.7 27.8 29.4 36.7 26.3 34.3 21.1 n.a.

66 75 33 52 71 71 6 79 73 72 75 77 66 77 66 29

2.5 2.5 2.9 3.1 2.9 1.3 2.4 2.3 2.2 3.5 2.1 2.6 1.1 1.8 1.8 0.7

Source: OECD.

Table 7.9 Economic performance disparities in the EU-15 Index* Germany France Italy Spain Netherlands Belgium Austria Greece Portugal Ireland Finland Denmark Sweden UK US

Continental, Big, Old Continental, Big, Old Mediterranean, Big, Old Mediterranean, Medium, Catching-up Scandinavian, Medium, Old Continental, Small, Old Continental, Small, Old Mediterranean, Small, Catching-up Continental, Small, Catching-up Liberal, Small, Catching-up Scandinavian, Small, Old Scandinavian, Small, Old Scandinavian, Small, Old Liberal, Big, Old Liberal, Big, Old

−3.0 −4.0 −5.3 −3.3 4.2 −0.2 3.8 −6.3 −7.2 5.1 4.3 5.5 5.1 1.4 −1.8

The index is calculated as: 2 g-i-(ur-5) + 0.5(sb + eb), using 1999–2005 data for GDP growth (g) and IPCH inflation (i); 2005 data for unemployment rate (ur), public (sb) and current account (eb) balances. The higher the index, the better the country’s economic performance. Source: European Commission, authors’ estimates.

Economic Divergences in the EMU

171

Table 7.10 Disequilibrium index* −2.5 −1.1 0.9

Big country In the euro area Old

Small country Outside the euro area Catching up

1.3 2.6 −3.0

Performance according to the Social Model Scandinavian

4.8

Continental

−0.9

Mediterr.

−5.5

Liberal

1.6

* Unweighted averages of the respective values in Table 7.9. Source: Authors’ estimates.

Until recently, European authorities have been focusing on public finance imbalances. But the weakness of GDP growth in Italy and Germany, the persistence of high unemployment rates in several euro area countries, and competitiveness losses in Southern economies as reflected in rising current account imbalances are more worrying for the euro area as a whole.

3 How to explain disparities in the euro area? 3.1 The advantages of being a catching-up country Catching-up countries have benefited from significant falls in their nominal interest rates from 1992 to 1999. Real interest rates adjusted from growth fell by 10 points in Greece, 6.5 in Spain and 4 in Ireland (see Table 7.3). At the same time, the exchange rate risk disappeared. Households’ and company investment was boosted by interest rates. This raised output and inflation, lowering further real interest rates adjusted for growth (Deroose et al., 2004). Domestic output cannot be stabilized through the common monetary policy. Losses in competitiveness are the only stabilizing factor but take time to play out. The rise in external deficits can be easily financed by the European banking system. Growth in catching-up countries has not decelerated despite rising external deficits because domestic demand robustness more than offset competitiveness losses. The persistence of external deficits induces a progressive deterioration of households’ and firms’ debt ratios, which could initiate a financial crisis. But this risk is minor due to the low level of real interest rates. The rise in foreign borrowing does not generate financial market prudence, because borrowing is in euros and hence there is no exchange risk. The relatively small size of catching-up countries

172 Fiscal Policy in the European Union

means that domestic inflation, credit expansion and external deficits do not generate tensions at the euro area level. Catching-up countries have therefore a strong advantage in being in the euro area. If they wanted to restore their competitiveness, they would need to run restrictive fiscal policies – but European authorities cannot request such a policy from them because robust GDP growth allows for a fiscal surplus – like in Spain – or restrictive wage policies, which are difficult to implement under robust output growth. Restrictive fiscal and wage policies would negatively affect peer countries so the latter do not exert pressure for such policies to be implemented. Hence imbalances can grow and persist for a long time. The bubble burst in Portugal in 2001, but has not yet done so in Spain or Greece. The two latter countries still seemed to be in a favourable situation in 2006.

3.2 Weaknesses in the euro area economic policy framework The euro area economic policy framework is based on three pillars. First, the single monetary policy aims at price stability. Second, domestic fiscal policies are under the surveillance of European procedures requesting medium-term budgetary positions in balance, allowing only economic stabilizers to play and no discretionary policy. Third, a European strategy of structural and liberal reforms is expected to raise medium-term growth, although this strategy is hardly implemented at the domestic level (Fitoussi and Le Cacheux, 2004; Mathieu and Sterdyniak, 2003, 2006). No common short-term stabilization policy is implemented. Specific country situations are not taken into account, in the absence of domestic inflation or external deficits criteria, GDP growth or employment targets and under single public finance targets. Last, there is no consensus in the EU on a macroeconomic strategy: some countries would favour a growth strategy supported by demand, while the Commission and other countries favour a strategy based on structural reforms. In contrast, a global macroeconomic strategy should set out ambitious growth targets for each country, should keep interest rates low, should try to maintain a relatively low exchange rate level and should let each country implement the fiscal policy needed to reach the desired level of output. Contractionary fiscal policies should be requested only in countries running excessive inflation (accounting for the Balassa-Samuelson effect), excessive external deficits or credit growth. Countries running excessive external deficits should implement wage moderation policies in order to restore their competitiveness, but this should be accompanied by interest rate cuts or wage increases or

Economic Divergences in the EMU

173

more expansionary fiscal policies in partner countries. In this framework, excessive inflation would signal specific supply problems in countries having robust demand. These countries could then fight these tensions without fearing insufficient demand. The needed reforms would become clear once tensions have actually emerged: insufficient capital stock, shortage of skilled labour in specific sectors. Experience shows that imbalances resulting from excessive labour demand can easily be tackled either through raising skills, labour market participation or immigration. It is over-supply which is difficult to tackle. Structural reforms are much more difficult to implement when demand is too low. When there are few job offers, it is difficult to find incentives to bring inactive people to work; it is difficult to train the unemployed for jobs that do not exist; investment weakness reduces labour productivity growth; it is difficult to reorganize the productive structure when laid-off people cannot easily find a new job; and structural reforms increase economic uncertainty, and therefore the savings rate, which dampens demand further. Public deficits must then support demand and this increases uncertainty. Output growth is then constrained by the weaknesses of both demand and supply. But the choice of binding economic policies in the EU is not a technical error. It is a political choice made by EU leading classes and technocracy. Monetary policy is already in the hands of unelected technocrats while fiscal policies remain in the hands of democratically elected governments but are constrained within narrow limits. The next step is to paralyze fiscal policy. EU governments should not repeat the mistakes of the 1970s when their desire to maintain full employment led to a sharp drop in profits. Some liberal economists even suggest going beyond: Wyplosz (2002), for instance, proposed establishing a fiscal policy committee of independent experts in each member state. This committee would be given the power of setting the government balance. The objective is to deprive people of any control over economic policies.

3.3 Non-cooperative national policies In the absence of economic policy coordination, countries use the tools at their disposal. Fiscal policy remains effective but in an area with many countries, each country may be reluctant to use the fiscal tool: the positive effects on output will be shared with trade partners and in the end will be relatively weak for the initiating country. Fiscal policy is relatively effective in large countries and restrictive fiscal policy is particularly costly. Large countries may also use their political weight to oppose

174 Fiscal Policy in the European Union

the Commission’s requests on reducing their domestic fiscal autonomy. Fiscal policy is less powerful in smaller countries. These countries also have less political weight in face of the Commission. In contrast, smaller countries can be tempted to improve their competitiveness through wage moderation, because the negative impact on domestic demand will be more than offset by external demand gains. Such strategies are more painful for larger countries. Besides, concerted wage negotiations between social partners are more common in smaller than in larger countries, which facilitates the implementation of wage moderation policies. Smaller economies may also more easily introduce tax competition policies because the gains from attracting foreign companies will exceed the revenue losses on the national tax base. If countries act independently, the outcome will be a non-optimal Nash equilibrium with too restrictive wage and fiscal policies. There is too much competition and not enough cooperation in the area. This coordination default is harmful for the area as a whole, more for larger economies than for smaller ones (Le Cacheux, 2005). Non-cooperative strategies can be implemented and be successful at the national level, especially in smaller countries. The bigger euro area economies initially opposed such strategies but had to follow later on, like Germany since 2001, which depresses demand in the euro area and strengthens the search for competitiveness gains. In practice, the larger countries have run relatively expansionary fiscal policies (Germany, France, Italy), and also Portugal and Greece. From 1998 to 2005, restrictive wage policies have been implemented in large countries (Germany), medium-size countries (Spain), and smaller economies (Austria, Greece). Two smaller economies (Finland, Austria) have cut their tax-to-GDP ratios by a larger extent. Large countries may be tempted to modify their labour market functioning, either by centralizing wage negotiations, or by promoting wage and labour flexibility. But there is no evidence that labour market flexibility alone can keep an economy close to full employment, in the absence of active macroeconomic policies (see the US example), or without benefiting from competitiveness gains, like in the smaller EU economies. Large countries (Germany, France and Italy) account for 65 per cent of the euro area population (71 per cent with Greece and Portugal). They could thus try to impose their strategy. However, they have never firmly taken a common position. For instance, in the SGP reform debate in 2005, they have not insisted that capital expenditures should be deducted from public deficits in the assessment of the 3 per cent of GDP threshold. Is this a sacrifice in favour of European construction? Or is

Economic Divergences in the EMU

175

this a strategic choice of domestic authorities in order to implement at home unpopular structural reforms in the name of Europe? 3.4 The crisis of the European Social Model Tensions are exacerbated by the difficulties faced by the ‘European Social Model’ in a global world. The continental Europe model appears less efficient vis-à-vis globalization than the Anglo-Saxon model or the Scandinavian one (Sapir, 2005; Aiginger, 2006). Continental European countries traditionally shared specificities in terms of the role of the state in the economy, industrial policies, relationships between firms, banks and the state, firms’ financing, and financial markets. These specificities have been implicitly given up in the context of financial globalization and European construction, but a new coherent framework remains to be settled. Globalization places continental European workers in competition with NMS and Asian workers. What strategy does Europe wish to implement in face of industrial job losses? Can Europe choose a two-pillar strategy: on the one hand subsidizing higher education and R&D in order to help the expansion of innovative and high value-added sectors; on the other hand, subsidizing a lower-productivity sector of services to people? Can this be achieved without a dramatic rise in inequalities? Does Europe wish the winners of globalization to compensate for the losers’ losses (but this requires that the winners agree or are constrained to pay, more national solidarity or more tax harmonization, two strategies which Europe does currently oppose)? Could the Scandinavian strategy, combining efforts on innovation and on requalification and social support of the unemployed, be applied in more open, heterogeneous and large countries? The continental European model must be adapted, but this raises conflicts of interest between social classes. This implies economic and social choices. European authorities do not help in the process, for instance when they restrict state aid to firms, impose reforms expected to foster competition, or prohibit fights against tax avoidance. European dominant classes have not tried to protect the European Social Model. They took the opportunity of globalization and of the single market to impose structural reforms in Europe, in particular public and welfare expenditure cuts and labour market flexibility. In their view, active macroeconomic policies cannot support output growth: people should understand that the only choice is to accept a liberal functioning of the economy or see capital flow towards more friendly skies. This is the famous TINA (There is no alternative strategy) of Margaret Thatcher.

176 Fiscal Policy in the European Union

Without any macroeconomic and social long-run coherent strategy, continental countries are the losers of the European construction in a global world.

4 What policy answers? The euro area as a whole suffers from insufficient growth and seems unable to cut mass unemployment. This is true especially for five countries accounting for 72 per cent of the area: Germany, Italy, France, Belgium and Portugal. Two countries (Greece, Spain) run large external deficits. The euro area economic strategy should thus aim at increasing the output level and at reducing imbalances while accounting for disparities. 4.1 A market-oriented view For many economists and for the Commission (for instance, Deroose et al., 2004; Tilford, 2006), increasing flexibility in all markets will reduce disparities in Europe. Instead of trying to improve the economic policy framework, the mainstream view refers to a Walrasian myth: if each economy was fully flexible (prices, wages, workers), there would be no need for economic policy and therefore no coordination problem. But the US example shows that this is an illusion: economic policy is needed even in a flexible economy. More flexible labour markets are not the panacea. Wage flexibility is not the answer to demand or supply shocks because it increases uncertainty and demand weakness. International labour mobility should not be an objective in Europe as a tool to reduce unemployment. The EMU should allow each country to grow, without having to ask people to emigrate (or even leave their region) to find a job. Under liberal strategies, the European Social Model necessarily moves towards the Anglo-Saxon one. In the absence of tax harmonization, member states will have no choice but to cut strongly public expenditure and redistribution. If firms are allowed to earn profits in Germany and pay (reduced) taxes in Switzerland, who will pay for public infrastructure in Germany? If firms can chose to pay for workers’ insurance in a private insurance company, who will pay for the poor? Continental European countries are supposed to have no choice but accept rising inequalities. 4.2 The Lisbon Strategy The European Commission does not recommend a purely liberal strategy, but a mix of sound macroeconomic policies, higher market flexibility,

Economic Divergences in the EMU

177

social protection reforms, public support for innovation and a move towards a knowledge society. This strategy is embedded in the Lisbon Agenda and raises implementation and content issues. The Lisbon Strategy was from the beginning a technocratic project, without democratic debate, without mobilization of European opinion, or involvement of civil society and social partners. The strategy did not account for country differences, necessary trade-offs between objectives, and differences of view among social forces. The majority of its objectives are related to research, innovation, and higher education and have little impact for the majority of people. Short-term issues have been neglected. Last, the Lisbon Agenda raises the political issue of who should manage the reform process: the Commission, national governments, national parliaments, social partners, or public opinion? 4.3 A sound macroeconomic strategy? We will consider here the latest version of the Lisbon Agenda, adopted in July 2005: the 24 integrated guidelines for growth and jobs (2005–8). Six macroeconomic guidelines recall the need for sound macroeconomic policies to support growth. Guidelines 1 and 6 repeat that countries must have medium-term budgetary positions in balance. Countries running deficits should cut their structural deficits by at least a 0.5 percentage point per year, whatever the cyclical situation. The link between the single monetary policy and national fiscal policies is not considered. Guideline 2 asks member states to address the issue of ageing populations by reducing public debt (but population ageing involves a rise in the savings rate, therefore the demand for public bonds), to reform their pensions and health systems (but how?) and finally to raise employment rates. But the growth strategy needed to reach these aims is not organized. Guideline 3 requests public expenditure to be reallocated towards research, infrastructure and education. However, the needed cuts in current expenditure are not specified. Guideline 4 requests that member states introduce structural reforms to facilitate the implementation of sound macroeconomic policies. But one could prefer that coordinated expansionary macroeconomic policies are implemented first in order to facilitate the implementation of structural reforms. 4.4 Microeconomic strategy: competition and innovation . . . The core objective is to raise productivity in Europe. EU-15 GDP per capita has remained at 72 per cent of US GDP per capita since 1973. This is more primarily due to differences in employment rate, unemployment rate and annual worked hours rather than to productivity per head.

178 Fiscal Policy in the European Union

But labour productivity per hour growth has kept on decelerating in the EU-15 (1.0 per cent per year) since 1995 whereas it has accelerated in the US (2.2 per cent per year). Innovations reducing the need for labour are hardly welcome in a mass unemployment situation. The lessons of the burst of the NITC bubble and the collapse in equity prices have not yet been drawn. Growth through innovation (Guideline 8), ICT (Guideline 9), and development of financial markets remains the dogma. Guideline 10 recommends strengthening the ‘competitive advantages of the industrial base’; but this would require a major change in the competition policy of the Commission, which aims at reducing state aid. Guideline 11 considers environmental issues, but contradictions between growth and environment are not highlighted. Guideline 12 recommends deepening the internal market. Here too, contradictions are not accounted for: should the energy sector and collective transportation system be privatized without considering long-term and regional planning issues? The questions raised by the Bolkestein Directive are not addressed: how should companies with different social standards compete? Guideline 13 calls for open and competitive markets; asks for the reduction of state aid which distorts competition, while recognizing the need for addressing market deficiencies, helping research, innovation and education. Guideline 14 calls for reducing regulations as if they were necessarily harmful. For example, should consumer protection be given up? Guideline 15 asks for fostering entrepreneurship, for instance by ‘a tax system that rewards success’, which questions tax progressiveness. 4.5 What employment strategy? The myth of flexibility The general objective remains to increase labour force availability and quality. No suggestion is made on how to increase job offers. Guideline 17 reaffirms ambitious objectives for employment rates (in 2010, 70 per cent for overall population, 60 per cent for women, 50 per cent for older workers). Guideline 18 suggests increasing labour demand by lowering youth unemployment, and giving women and older people incentives to work. Guideline 19 recommends increasing work incentives. But the call for modernizing social protection systems is worrying, if the point is to reduce early retirement, cut pensions or unemployment benefits while job opportunities for older workers are not there yet. Guideline 20 proposes to remove obstacles to labour mobility, but sensitive issues are not addressed: how to prevent workers from Central and Eastern European countries from exerting downward pressure on wages in the West? Guideline 21 recommends increasing flexibility

Economic Divergences in the EMU

179

(albeit reconciling it with job security), anticipating better future changes and facilitating transitions. But what strategy should be implemented? Should it be Anglo-Saxon or Scandinavian? Guideline 22 proposes to ensure that wages grow in line with productivity and that non-wage costs are cut, especially for the low-paid. Wages already tend to grow less rapidly than labour productivity in the euro area: the wage share in value-added dropped from 67.4 per cent in 2000 to 66.2 per cent in 2005. Social contributions cuts cannot imply benefits cuts (what would be the advantage for workers and jobs to cut health contributions if this meant they would have to pay for private insurance?). Other resources should be found. 4.6 The integrated guidelines forget several things The integrated guidelines forget that Europe suffers from insufficient demand which partly results from the European framework. They do not try to make the SGP consistent with the Lisbon Agenda, accounting for investment expenditure in the assessment of domestic fiscal policies. Monetary policy and more especially exchange rate policies are not considered. Can the euro area remain competitive after a 40 per cent rise in the euro vis-à-vis the dollar? The guidelines do not consider social Europe. How to reconcile freedom of movement and establishment with domestic tax autonomy? How to avoid a ‘race-to-the bottom’ tax competition? How to avoid a rising gap between the winners of globalization refusing to contribute to national solidarity and the poor? The guidelines do not consider industrial policy. Are competition policy and cuts in state aid sufficient? Should European champions and innovative sectors be supported? What are the policy answers in the face of delocalization? 4.7 Less Europe? The renationalization of economic policies The renationalization strategy would consist in giving again national governments room for manoeuvre to implement specific economic strategies. Each country could thus take care of their specific imbalances in their own way. The European economic policy framework would be re-examined to leave more autonomy for member states. The latter would be allowed to implement domestic industrial policies, to protect their tax revenues through measures fighting tax avoidance, to choose their social model, etc. Governments (and the people) would be clearly responsible for their choices. However, euro area member states would continue to share common monetary policy, interest rate and exchange rate. A code of good conduct would still need to be defined in order to avoid excessive inflation

180 Fiscal Policy in the European Union

rates or external deficits. The countries would thus not be able to choose their policy mix. It would be difficult to renationalize the external trade policy. The prospect of future tax or social harmonization would vanish. Renationalization would increase the risk of implementation of noncooperative strategies and of social and tax dumping. The countries would have no choice but to move towards the liberal model. Europe would give up the ambition of offering a specific model. 4.8 More Europe? European countries have no other choice but to design a new economic policy framework, at least for the euro area, if not for the EU-27. This framework must, at the same time, match democratic principles and account for diverse national situations. A first option would be for economic policy to be decided at the area level. But Europe is not a nation: political, economic and social lives are not unified. Institutions, taxation and social protection systems differ. Economic circumstances themselves remain different. Some convergence would be necessary, but there are three models in Europe: liberal, Scandinavian and continental. Towards which model should Europe converge? There is no consensus on economic policies, reforms, strategies or institutions in the euro area. Choices can only result from a democratic process. But European peoples would probably not agree on an economic, political and social unification decided at the European level. It is difficult to imagine a single framework able to manage all the different national situations. Giving more power to EU institutions (the Council, the Commission or the Eurogroup) on national economic policies would go against democracy. For instance, decisions concerning Belgium would be taken by a committee made up of EU technocrats, other governments’ representatives and only one person elected by Belgian people. It is difficult to see how Belgian interests and specificities would be taken into account. The decisions of this committee would be similar to those currently recommended by the Council and the Commission, for example in the BEPGs and in the assessments of the Stability Programmes: restrictive fiscal policies to cut government spending and deficits. This would go against democracy (EU citizens would no longer have the right to choose the level of public expenditure in their countries) and economic stabilization (restricting fiscal policy to automatic stabilizers is too binding; the objective of zero deficit as advocated in the Stability and Growth Pact has no economic justification; in the long run it would lead to a zero public debt, which is not optimal). Surprisingly Carlo Panico and

Economic Divergences in the EMU

181

Marta Vázquez Suárez, in this book, advocate giving a Euro Area Fiscal Agency the right to set national fiscal policies. This agency would gather European Commission, ECB and national government representatives. What would be its democratic legitimacy? The technical analysis would be run by the Economic and Financial Committee (experts belonging to the European Commission, the ECB and the member states’ ministries for finance and economics), exactly as today. The idea of independent fiscal committees generally supported by orthodox mainstream economists receives here unexpected support from Panico and Vázquez Suárez, who seem to think that such an agency advised by such committees could undertake a Keynesian policy. How can this be called ‘a post-Keynesian scheme’? European governance will need to be based on national policy coordination. The euro area seems more appropriate for such coordination than the EU-27 because it covers more homogeneous economies. But this requires that member states agree to share a common ‘European Social Model’ that will need to be defined, protected, and able to evolve. However, the economic policy to be implemented remains a difficult issue (see also, Huffschmid, 2006) and will have to combine Keynes, Colbert and social democracy. In a mass-unemployment situation, the euro area needs a growthoriented policy. It would be desirable to set up actual economic policy coordination in the framework of the Eurogroup, with which the ECB would have to communicate. This coordination should not focus on public finance balances, but should aim at achieving a 3 per cent annual growth target. The process will have to account for the needs for growth in each country and also for their competitiveness, external balance and inflation rate. Intra-zone disparities should be accounted for: catchingup countries could run higher inflation, high savings rate countries could be entitled to higher public deficits, some countries should be asked to raise taxes in order to avoid over-accumulating private debt, a depreciation of the euro could be preferred to national competitive policies . . . This is not an easy task. Europe should try to design a specific model of European firms, caring about jobs, regional activity and sustainable growth. Companies have a social role to play. They should care about not only their shareholders but also their employees and customers. This means that member states should maintain a relatively high level of company taxation to give companies incentives to build homogeneous infrastructures, subsidize firms locating their production in areas in difficulty, supporting economic sectors in difficulty and subsidizing R&D. Member states should have an

182 Fiscal Policy in the European Union

active industrial policy, aiming both at developing large European companies and innovating small and medium enterprises (SMEs), research centres and company networks. European authorities should plan their future productive activities and industrial employment in Europe, reduce the weight of competition policy and promote a European industrial policy in the framework of the Lisbon Strategy. Community rules need to be modified, so that domestic aid can support specific sectors. The objective should be to maintain the European Social Model, characterized by a significant level of transfers, public expenditure and thus of taxation. Tax and social competition should be avoided. Some harmonization will be needed to prevent unfair competition, through the introduction of minimum tax rates (corporate income, household wealth and income) and minima benefits (minimum income, minimum pension replacement ratio) and by tough measures against tax havens at a worldwide level. The European Social Model will have to rely on its comparative advantages (free education and health for all, good quality public infrastructure, efficient social security benefits) to remain competitive in a global world. A stronger GDP growth would lower unemployment rates and would allow the introduction of a ‘flexisecurity system’ in continental countries, with an adequate support for the unemployed (vocational training, training for new jobs). The improvement of the European economic framework is not only a technical issue; it requires a major change in the economic policy thinking, a new alliance between social classes concerned about full employment and social cohesion, and a willingness to depart from financial markets’ and multinational companies’ points of view. A prerequisite would be that member states’ populations agree on a European model, but we are far from there.

References Aiginger, K., ‘The Ability of Different Types of Socio-economic Models to Adapt to New Challenges’, Wifo, June (2006). Benalal, N., J. L. Diaz del Hoyo, B. Pierluigi and N. Vidalis, ‘Output Growth Differentials across the Euro Area Countries: Some Stylised Facts’, ECB Occasional Papers, 45, May (2006). Deroose, S., S. Langedijk and W. Roeger, ‘Reviewing Adjustment Dynamics in EMU: From Overheating to Overcooling’, European Commission, Economics Papers, 198 (2004). Esping-Andersen, G., The Three Worlds of Welfare Capitalism (Princeton: Princeton University Press, 1990). European Central Bank, ‘Inflation Differentials in the Euro Area: Potential Causes and Policy Implications’, September (2003).

Economic Divergences in the EMU

183

Fitoussi, J. P. and J. Le Cacheux, Rapport sur l’Etat de l’Union Européenne (chapter 2), (Fayard, 2004). Huffschmid, J., ‘Economic Policy for the European Social Model’, in E. Hein, A. Heise and A. Truger (eds), European Economic Policies: Alternatives to Orthodox Analysis and Policy Concepts (Marburg: Metropolis-Verlag, 2006), 283–301. Le Cacheux, J., ‘Politiques de croissance en Europe, un problème d’action collective’, Revue Économique, May (2005). Mathieu, C. and H. Sterdyniak, ‘Réformer le Pacte de stabilité: l’état du débat’, Revue de l’OFCE, 84, January (2003). English version: ‘Reforming the Stability and Growth Pact’, Document de travail de l’OFCE, May (2003). Mathieu, C. and H. Sterdyniak, ‘A European Fiscal Framework Designed for Stability or Growth?’, in E. Hein, A. Heise and A. Truger (eds), European Economic Policies: Alternatives to Orthodox Analysis and Policy Concepts (Marburg: Metropolis-Verlag, 2006), 41–67. Sapir, A., ‘Globalisation and the Reform of European Social Models’, Bruegel Policy Brief, November (2005). Tilford, S., ‘Will the Eurozone Crack?’, CER, September (2006). Wyplosz, C., ‘Fiscal Discipline in EMU: Rules or Institutions?’, mimeo, Meeting of the Group of Economic Advisers of the EC, 16 April (2002).

8 A Scheme to Coordinate Monetary and Fiscal Policies in the Euro Area Carlo Panico and Marta Vázquez Suárez

1 Introduction This chapter deals with the problems of coordination between monetary and fiscal policies in the Euro area. It examines how the existing institutions handle these problems and presents a proposal to reorganize them. The need for policy coordination in the European Monetary Union (EMU) was recognized by the European Council in a meeting held in Luxembourg in 1997. The Council stated that ‘the move to a single currency will require closer community surveillance and coordination of economic policies among euro area member states’. This acknowledgement has not led to a satisfactory institutional organization. According to Von Hagen and Mundschenk (2003: 279), the present arrangements lead to inefficient policy outcomes on account of the non-cooperative attitude that they induce in the monetary and fiscal authorities of the area. They also claim that coordination is needed only in the case of cyclical (short-run) policy, since ‘in the long run monetary policy can achieve price stability without interfering with fiscal policy’ (Von Hagen and Mundschenk, 2003: 293). The term ‘coordination’ has different meanings in this literature. According to some (Pisani-Ferry, 2002; Von Hagen and Mundschenk, 2003; Wyplosz, 1999, 2002), it refers to the set of arrangements and activities aiming at the identification of a unified framework for monetary and fiscal policies and the introduction of commitments on policy decisions at national and super-national level. For others (Buti et al., 2002; Fatàs and Mihov, 2003; Beestma and Debrun, 2005; Calmfors, 2005), ‘coordination must be understood as an agreement to enforce fiscal discipline among members of EMU to avoid any spill-over caused by irresponsible policies’ (Fatàs and Mihov, 2003: 126). 184

Coordinating Monetary and Fiscal Policies 185

The dissatisfaction with the present organization has led those using a broad definition of coordination to call for academic research, since ‘solutions are urgently needed’ (Wyplosz, 1999: 2) and ‘institutional building is necessary’ (Wyplosz, 1999: 23). They have also provided suggestions to improve on the existing institutional arrangements (Wyplosz, 1999, 2002; Casella, 2001; Pisani-Ferry, 2002; Von Hagen and Mundschenk, 2003). The authors accepting a narrow definition of coordination have instead made proposals to improve the economic content and the enforcement of the rules imposing fiscal discipline in order to avoid a return to fiscal policy discretion (for a survey see Buti et al., 2002; Franco et al., 2005; Calmfors, 2005). With respect to the large amount of publications on this issue, the analyses proposed by the post-Keynesian literature appear little developed. Hein and Truger (2006) provide some indications as to how to reform the Stability and Growth Pact (SGP). Other writings refer to the issue without dealing with it in detail. Angeriz and Arestis (2007) argue in favour of coordination between monetary and fiscal policies in a paper dealing with the conduct of monetary policy in the UK. Arestis and Sawyer (2005) claim that the rules currently governing EMU make it difficult to implement the coordination of monetary and fiscal policies.1 The aim of this chapter is to propose a post-Keynesian perspective of these problems. It points out, contrary to what is often stated by the literature, that there is a need for coordination between monetary and fiscal policies when both cyclical (short-run) and structural (long-run) problems are dealt with. Then, it assesses how coordination is carried out under the existing institutional arrangements and proposes a new scheme, which intends to make them work effectively. The chapter is organized as followed. Section 2 deals with the arguments in favour of coordination in both short- and long-run analysis. Sections 3 and 4 describe the existing institutional arrangements and procedures. Section 5 summarizes some proposals for reform present in the literature. Section 6 proposes a new scheme.

2 Arguments in favour of coordination A single monetary policy and a single currency were introduced in the euro area in 1999 and in 2002 respectively. A set of new monetary institutions – the European System of Central Banks (ESCB), the Eurosystem and the European Central Bank (ECB) – was created to manage it. As Wyplosz (1999: 1) points out, the architecture of these institutions is a heritage of the rational expectations revolution applied at the

186 Fiscal Policy in the European Union

policy-making level. It is based on the principles of neutrality of money, inter-temporal inconsistency and the superior performance of independent central banks. The conduct of monetary policy, as Arestis and Sawyer (2006) clarify, follows the lines of what is known as the ‘New Consensus’ in macroeconomics (NCM). The model underlying the NCM is again related to the rational expectations revolution and is based on the following elements: • Say’s Law holds, so that effective demand does not play any role in

the long-run determination of the level of economic activity. • The long-run level of activity is supply-determined and corresponds

to the non-accelerating inflation rate of unemployment (NAIRU). • The potential growth of the economy is fully exploited. • Money is neutral, in the sense that it does not affect the long-run level

of activity. • In cyclical (short-run) analysis, the level of activity fluctuates around

the NAIRU. • Monetary policy, operating through the interest rate, is the key ele-

•

• • •

ment of short-run policy, which deals primarily with inflation, but also takes care of the short-run output gaps. The money supply is an endogenous variable, in the sense that its amount depends on the demand for money at the level of the interest rate stabilized by the monetary policy. Monetary policy should be operated by an independent central bank.2 The independence of the central bank enhances its credibility and improves the effectiveness of monetary policy. Transparency is essential to the working of monetary policy: it allows the authorities to operate in a discretionary way while avoiding the problems of time inconsistency.

Say’s Law and the neutrality of money are the elements that mainly differentiate the ‘New Consensus’ from the Keynesian theoretical positions. They both belong to long-run analyses. In Keynesian analysis Say’s Law does not hold, effective demand plays a role in determining the long-run level of economic activity, the potential growth of the economy is not always fully exploited, money is not neutral, and both fiscal and monetary policy, operated through the interest rate, can affect the long-run level of production. Keynes introduced these views in a typed paper from which he appeared to have lectured in November 1932 (see Panico, 1988). There, he clearly showed his intention to abandon the traditional neoclassical

Coordinating Monetary and Fiscal Policies 187

standpoint and to propose a new theoretical approach, which he named a ‘monetary theory of production’ to underline the fact that the longperiod equilibrium level of economic activity cannot be determined independently of monetary considerations: The root of the objection which I find to the theory under discussion, if it is propounded as a long-period theory, lies in the fact that . . . it cannot be held that the position towards which the economic system is tending . . . is entirely independent of the policy of the monetary authority . . . On my view, there is no unique long-period position of equilibrium, equally valid regardless of the character of the policy of the monetary authority. On the contrary there are a number of such positions corresponding to different policies. Moreover there is no reason to suppose that positions of long-period equilibrium have an inherent tendency or likelihood to be positions of optimum (i.e. full employment) output. (Keynes, 1979: 54–5) These positions, which Keynes further elaborated in The General Theory, were subsequently developed by Harrod’s seminal article ‘An Essay in Dynamic Theory’ (1939), which conceived of modern growth theory as a Keynesian theory. It worked out the views that the economic system does not tend necessarily to full employment and that the rate of growth may be affected by three sources of autonomous demand, coming from the government sector, the private sector, in the form of autonomous investment, and the foreign sector.3 Some years later, Kaldor’s 1958 Memorandum to the Radcliffe Commission considered government policies necessary to pursue stability and growth. His opinion was that, although both policies must constantly be used in a coordinated way, monetary policy is the appropriate tool against the fluctuations of the economy, while fiscal policy is the appropriate tool to use in the pursuit of the long-range objective of sustained growth. In the Memorandum Kaldor used the post-Keynesian theory of growth and distribution to examine the role of government policies.4 Yet, he did not present a formalized analysis, which was worked out later by the debate on the role of the government sector in the post-Keynesian theory of growth and distribution5 and by the writings on the same topic coming from the Kaleckian tradition.6 The need for coordination between monetary and fiscal policies was widely recognized during the post-war Keynesian era. In 1956,

188 Fiscal Policy in the European Union

intervening in a debate between the Treasury and the Federal Reserve, Samuelson underlined it by writing: There is no legitimate clash between Treasury and Central Bank policy: they must be unified or co-ordinated on the basis of the over-all stabilisation needs of the economy, and it is unthinkable that these two great agencies could ever be divorced in functions or permitted to work at cross purposes. (In particular it is nonsense to believe, as many proponents of monetary policy used to argue, that fiscal policy has for its goal the stabilisation of employment and reduction of unemployment, while monetary policy has for its goal the stabilisation of prices. In comparison with fiscal policy, monetary policy has no differential effectiveness on prices rather than on output) . . . I have already asserted that the Treasury and Central Bank have to be co-ordinated in the interests of national stability, so I am little interested in the division of labour between them. (Samuelson, 1956: 14–15) In the subsequent years several lines of argument were presented to criticize the discretionary use of fiscal policy to stabilize the economy. In an attempt to restate the case for fiscal policy Arestis and Sawyer (2006) have recently recalled the limits of these arguments. One of them assumes that the economy permanently fluctuates around a supply-side equilibrium position, corresponding to the natural rate of unemployment (NRU) or to the non-accelerating inflation rate of unemployment (NAIRU) and representing the potential output of the economy. The empirical identification of this position is difficult because the estimation of the rate of structural unemployment is influenced by the rate of unemployment that has actually prevailed in the market (see Blanchard and Katz, 1997; Blinder, 1997; Galbraith, 1997; Gordon, 1997; Staiger et al., 1997; Stiglitz, 1997). As a consequence, the natural rate of unemployment, which should represent a stable reference point for the actual rate, tends instead to follow it: the estimated value of the former depends on the prevailing levels of the latter. Dealing with this problem, Solow has expressed serious doubts on the applicability of the natural rate hypothesis in the USA: As this is written, the unemployment rate has been below the once canonical 6 percent for five years, but the GDP deflator has decelerated during that interval, and hourly compensation, while accelerating slowly, is much more viscous than the 6-percent-Nairu story would

Coordinating Monetary and Fiscal Policies 189

have led a reasonable person to expect. (Reconciling the course of European unemployment with the standard model requires even more in the way of late-Ptolemaic epicycles.) Of course, the Nairu story can always be saved by agreeing that whenever the rate of inflation is falling, the current unemployment rate must be above the Nairu. Yes, but unless the Nairu changes only very slowly or very rarely or is well predicted by the background model, the story has been saved by emptying it of content. (Solow, 2000: 157)

3 Institutional arrangements The arrangements regarding the coordination between monetary and fiscal policies in the EU and the EMU are complex. Many institutions participate in the process and several procedures have been set up to pursue this objective. The European Commission (EC) occupies a central position in the coordination process. It participates in the meetings of all other bodies, sets the agenda for the Council of the Ministers and proposes the procedures regarding the financial sustainability of the member states. The Council of the Ministers of the European Union (CMEU), composed of the ministers and the staff of the public administration of all countries of the Union, takes all formal resolutions on policies. The Council of Economics and Finance Ministers (ECOFIN) is the section of the Council of the Ministers dealing with government deficits, spending and taxation. In Luxembourg in 1997, the European Council declared that the ECOFIN is the centre of coordination of the economic policies of the member states. The European Central Bank (ECB) participates in the meetings of the ECOFIN to favour the dialogue between these institutions. At the same time, the President of the ECOFIN participates in the ECB Governing Council.7 He can submit motions to this body, but has no right to vote. The Eurogroup was set up in 1997 to allow the euro countries to discuss their policy issues without the participation of the other member states of the Union. It meets every month, before the meetings of the ECOFIN. Participation in its meetings is limited to the Ministers of Economics and Finance of the euro countries, to one member of the European Commission and to the President of the ECB, each accompanied by another person. The Eurogroup has no formal decision-making authority, which remains with the ECOFIN. Its role is limited to assessing

190 Fiscal Policy in the European Union

the economic situation of the area and discussing its major policy issues. The lack of decision-making authority prevents the Eurogroup from playing the same role that the Eurosystem plays with respect to the ESCB. The Economic and Financial Committee (EFC) was set up in 1999 to provide the ECOFIN with analysis and advice on economic and financial issues. It is composed of experts belonging to the European Commission, to the member states and to the ECB. The EFC prepares the analytical basis upon which the decisions of the Council of the Ministers are taken. It favours dialogue between the Council of the Ministers and the ECB on issues like the economic policy orientation of the EU, the monitoring of the fiscal policies pursued by the member states as they are described by the stability and convergence programmes that the states present each year, and the positions of the European Union on international economic and financial problems. The EFC also deals with other technical issues, like questions related to the printing of money, institutional reforms and the international economic relations of the EMU and the EU. The analyses of the EFC are used for the meetings of the ECOFIN and the Eurogroup. The ECB participates in this activity by contributing to the evaluation of the economic perspectives of the euro area, of the exchange rate, of the general orientation of the national fiscal policies and of their financial sustainability. The Economic Policy Committee (EPC) was set up in 1974 and has a composition similar to that of the EFC. Unlike the latter, which deals with macroeconomic and financial policies, the EPC deals with the structural reforms of the member states. It meets annually for analysing economic and social reforms related to the ageing of population and to the working of the markets for goods, services, capital and labour, and for discussing the possibility of introducing some forms of coordination in the policies of the different European countries on these issues. This description of the institutions participating in the process of policy coordination underlines that the arrangements chosen by the European Union are characterized by the following elements: • The analyses upon which the coordinating activity is carried out are

elaborated by the EFC and the EPC with the participation of the ECB. • The European Commission leads the processes, setting the agenda

of the Council of the Ministers and proposing and developing the coordinating procedures. • The Council of the Ministers (and the ECOFIN for the economic policy) is the formal decision-making authority.

Coordinating Monetary and Fiscal Policies 191 • The Eurogroup deals with economic policy within the euro area, but

has no formal decision-making authority. It does not play, with respect to the ECOFIN, the same role that the Eurosystem plays in relation to the ESCB. Thus, in spite of the fact that the single currency creates a broader need for cooperation, within the euro area ‘a framework for cooperative policy making among the ECB and the national governments has not yet been developed’ (Von Hagen and Mundschenk, 2003: 293). The absence of a place where the actors can define a common stance on monetary and fiscal policies and enter binding commitments on their policy choices, makes the institutional organization of the euro area more oriented towards the exchange of information than towards the identification of lines of action on coordinating activities. This deficiency, as Von Hagen and Mundschenk (2003: 280 and 293–4) conclude, leads the actors to non-cooperative attitudes. The ECB, feeling uncertain about the fiscal choices of the national governments, tends to defend its freedom of action to be able to counteract, if necessary, their deficit bias. The national governments, feeling unable to persuade the ECB to pay more attention to the specific conditions of their economy and uncertain about the fiscal behaviour of the other governments, are induced to be sceptical on cooperation, prepared to counteract any restrictive stance of monetary policy and unwilling to accept further reductions of their fiscal sovereignty. The result, as some econometric analyses reveal (see Wyplosz, 1999), is that monetary and fiscal policies in the euro area are not conducted as complementary, but as strategic substitutes.

4 Procedures Coordination between monetary and fiscal policies does not work through delegation to community institutions, as in the case of monetary policy and the administration of the single market. It is based on a narrow approach, which only tries to challenge the national policies that are expected to affect in a negative way the performance of the area (Von Hagen and Mundschenk, 2003: 288). The methods used for the harmonization of policies mainly rely on ‘soft enforcement’, i.e. on the ability to persuade the actors to follow proper behaviours through monitoring, dialogue, exchange of information, warnings and peer pressure. Only in some cases do they try to establish ‘common rules of behaviour’ that the national governments have to follow.

192 Fiscal Policy in the European Union

The main procedures of economic policy coordination in the EU can be listed as follows: • The Broad Economic Policy Guidelines (BEPGs). • The process of multilateral surveillance. • The process of ‘early warnings’. • The process of peer pressure. • The ‘Macroeconomic Dialogue’, known as the Cologne Process. • The Excessive Deficit Procedure (EDP). • The Stability and Growth Pact (SGP).

The BEPGs are the core of the coordination process in the EU. They define the general orientation of economic policy in the Union on the basis of the analyses developed by the EFC, with the participation of the ECB. The European Commission proposes the BEPGs to the European Council and to the ECOFIN, which take formal decisions on their content by unanimity vote. The BEPGs form the reference framework for the multilateral surveillance, which monitors the economic situation and the state of policy implementation in the Union and in the member states. The EFC develops these analyses by moving from the ‘stability programmes’ that the euro countries submit each year.8 The European Commission proposes the results of the monitoring to the European Council and to the ECOFIN for formal approval. Some of these results are conveyed to the member states and used to give ‘early warnings’ and recommendations. In some cases they set in motion a process of peer review, aiming at persuading the national governments to adopt fiscal policies that do not affect in a negative way the performance of the area. The tendency to use methods of ‘soft enforcement’ led the European Council to introduce in Cologne in 1999 the Macroeconomic Dialogue. The Council decided that an informal meeting should be held every six months in order to clarify how to deal with the European Agreement on Employment (which develops the coordination process on unemployment set up by the ‘Luxembourg process’) and how to solve the problem regarding the efficient working of the goods and capitals markets (as a way of keeping the Cardiff process at work). The meeting sees the participation of the representatives of the member states, the ECB, the European Commission, the central bank of a country that has not adopted the euro, and the representatives of the unions of the workers and the entrepreneurs. The Dialogue is carried out by inducing collaborative behaviours among the actors involved and without imposing specific lines of action on the participants.

Coordinating Monetary and Fiscal Policies 193

The procedures envisaged to enhance policy coordination also try to establish ‘common rules of behaviour’, as in the case of the Excessive Deficit Procedure (EDP) and the Stability and Growth Pact (SGP). These procedures, based on the results of the multilateral surveillance, stipulate: • the prohibition of ‘bailing out’ the governments that cannot meet

their debt obligations; • the prohibition that the gross government debt exceed 60 per cent of

GDP, unless it is diminishing and approaching the reference level ‘at a satisfactory pace’; • the prohibition of exceeding a 3 per cent ratio between the actual budget deficit and the GDP unless ‘either the ratio has declined substantially and continuously and reached a level close to the reference value or, alternatively, the excess of the reference value is only exceptional and temporary and remains close to the reference value’; • the governments’ adoption of a medium-term (or cyclically adjusted) objective of a budget position ‘close to balance or in surplus’. The procedures are set in motion by a report of the European Commission, clarifying the existence of the violation and of any other factors that can affect the decisions regarding the subsequent actions to be taken. The EFC examines the report and formulates an opinion on it. The Commission then decides whether to send the ECOFIN a statement regarding the existence of the violation and a recommendation suggesting the appropriate policy actions to correct it. The ECOFIN takes a formal decision on the existence of the violation and gives the country a recommendation on the correcting measures with two deadlines regarding the adoption of the measures and the correction of the violation. If the member state fails to observe the deadlines, the ECOFIN can give notice to it and can ask the country to make a non-interest bearing deposit with the Commission, whose amount is related to the extent of the violation. If the ECOFIN decides that after two years the violation has not been corrected, it can request the conversion of the deposit into a fine. From the start of the monetary union in 1999 there have been several violations of what the SGP stipulates. Portugal (2001, 2005 and 2006), Netherlands (2003), Greece for the whole period, France (2002, 2003, 2004 and 2006), Germany from 2002 to 2005, and Italy from 2003 to 2006 have exceeded the 3 per cent limit for the ratio deficit:GDP. The high number of violations testifies to the rigidity of the form of coordination chosen and the need to make it more flexible and in conformity to the needs of stabilization and growth of the national economies.

194 Fiscal Policy in the European Union

The application of the procedures too has proved difficult. In the case of the Netherlands (2003) the procedures were applied as prescribed. In all other cases they have not been enforced in the stipulated way. The cases of France and Germany are the most relevant. In November 2003 the ECOFIN halted the application of the established procedures against these two countries. This decision induced the Commission to take legal steps in the European Court of Justice. In 2004, the Court annulled the ECOFIN’s decision, but stated that the steps of the EDP are not automatic: even if the Treaty and the SGP envisage them, they are subject to ECOFIN’s decisions based on a qualified majority. The problems that emerged in the management of these cases led to a reform of these procedures in March 2005. The reform, as Calmfors (2005: 58–70) points out, watered down the economic content and the enforcement of these procedures. It rightly introduced ‘greater tolerance of deficits in severe downturns, the differentiation of medium-term fiscal objectives in order to take heterogeneity among countries in both debt levels and potential growth rates into account, the attempts at increasing the emphasis on debt developments in general, the plans on better assessment of implicit debt and long-run sustainability, and the “commitments” to avoid procyclical fiscal policy’ (Calmfors, 2005: 69). Yet, the way in which the reform specifies the meaning of a ‘severe economic downturn’ allowing the overtaking of the 3 per cent ratio between government deficit and GDP and of ‘the other relevant factors’ that enter in the evaluation of whether a deficit is excessive has made the content of the rules less transparent.9 Moreover, the new reform refers to both cyclical and structural factors, such as the Lisbon Agenda, R&D, innovation, debt sustainability and public investment, thus contradicting the view that rules and coordination are only needed in relation to cyclical problems because the long-run level of activity always exploits fully the growth potentials of the economy. Finally, the changes introduced by the reform in the deadlines set to the member states for correcting the violations substantially reduce the strictness of the enforcement and ends up by representing a movement away from the use of ‘common rules of behaviour’ and towards a return to unconstrained discretion in national fiscal policies (Calmfors, 2005: 63–6). Instead of strengthening the coordination of economic policies by endowing the procedures with stricter enforcement and greater flexibility with respect to the specific needs of the individual economies, the reform has reduced the level of coordination by making less transparent the economic content of the rules, weakening the enforcement and favouring unconstrained national discretion.

Coordinating Monetary and Fiscal Policies 195

To sum up, the existing arrangements regarding the coordination between monetary and fiscal policies in the euro area appear unsatisfactory in several respects: • The frequent violations of what the Maastricht Treaty and the SGP

stipulate testify to the rigidity of the forms of coordination chosen and the necessity to make them more flexible with respect to the needs for growth and stabilization of the national economies. • The 2005 reform of the SGP has rightly recognized the need for greater tolerance of deficits in the presence of cyclical downturns and the necessity to take into account the heterogeneity of countries on their debt position, international competitiveness and potential growth. Yet, instead of strengthening the coordination between monetary and fiscal policies by endowing the procedures with stricter enforcement and greater flexibility with respect to the specific needs of the national economies, the reform has reduced the level of coordination by making less transparent the economic content of the EDP and the SGP, weakening the enforcement of these procedures and favouring unconstrained national discretion. Under these conditions there is a serious risk that in future downturns fiscal policies in the Union will go out of control due to contagious effects with high deficits spreading from one country to another. • The existing institutions and procedures are more oriented towards the exchange of information than towards the identification of lines of action on coordinating activity. • In spite of the fact that the single currency creates a broader need for policy coordination, the euro area is not endowed with a specific institution where the actors can identify a common stance on monetary and fiscal policies and enter binding commitments on policy choices. This deficiency enhances non-cooperative attitudes among the actors, with the ECB giving priority to the defence of its freedom of action and the national governments being unwilling to accept further reductions of their fiscal sovereignty. The creation of a specific institution where common objectives can be clearly defined and set as the basis of subsequent policy choices appears a crucial step to improve the current situation.

5 Recent literature on coordination The need to change the existing arrangements is testified by the high number of reform proposals recently presented. The main weaknesses

196 Fiscal Policy in the European Union

of these arrangements have been found in the difficulty of identifying a common stance for economic policies, in the weak enforcement of the rules of the SGP, in their lack of transparency and flexibility, in the fact that the rules do not differentiate among the conditions of the economies, in the limited attention paid to financial sustainability, to the introduction of structural reforms, to the avoidance of pro-cyclical policies, and to the pursuit of policies favouring innovation and the formation of physical and human capital. The differences in the proposals reflect the diverse views on these problems and the meanings attributed to the term ‘coordination’. For some authors (Alesina et al., 2001; Buti et al., 2002; Calmfors and Corsetti, 2003; Fatàs and Mihov, 2003; Beestma and Debrun, 2005; Franco et al., 2005), the introduction of rules imposing fiscal discipline on national governments and the independence of the central bank generate satisfactory policy outcomes. In this case, coordination is understood as an agreement to enforce fiscal discipline. No other form of coordination is needed. Like Canzoneri and Diba (1999) and Alesina et al. (2001), Franco et al. (2005) claim that the maintenance of the ‘cyclically adjusted budget’ close to balance and the independence of the central bank represent an effective form of coordination. They claim that the working of automatic stabilizers in line with tax smoothing theory contributes to the achievement of an optimal fiscal stance both at national and at euro area level. Several other authors, moving along similar lines, have proposed changes in the economic content and in the enforcement of the rules of the SGP. Their proposals try to achieve a satisfactory balance between the need to impose clear and transparent constraints on national fiscal policies in order to protect the independence of the ECB and the need to introduce sufficient flexibility in the rules of the SGP in order to allow national governments to use counter-cyclical fiscal policies effectively. Fatàs and Mihov (2003: 118–19, 129) notice that the use of a ‘cyclically adjusted budget’ as the reference value of the SGP contributes to achieving greater flexibility. Yet, the process of calculation of this variable introduces arbitrary elements that reduce the transparency of the rules and favour political manipulation. They propose avoiding the reference to simple numerical rules and taking into account national differences, claiming that ‘as long as restrictions are not a simple numerical rule, but are built in as checks and balances in the budgetary process, they can prevent policy-makers from implementing large and frequent politically motivated discretionary changes in fiscal policy’ (Fatàs and Mihov,

Coordinating Monetary and Fiscal Policies 197

2003: 118). The meaning of ‘checks and balances in the budgetary processes’, however, is not clear. Fatàs and Mihov first recognize that this solution may somehow resemble that of Wyplosz (2002), which calls for the introduction of Fiscal Policy Committees (see below), then they claim that the surveillance of the European institutions may be able to constrain the formation of the national budgetary processes, as shown by the large reductions in the government deficits in the 1990s after the adoption of the Maastricht criteria. Blanchard and Giavazzi (2004), after stating that the SGP is increasingly held responsible for the inability of the euro area economy to sustain demand and maintain growth, have proposed a rule that distinguishes between current and investment expenditure, claiming that the latter must be excluded from the restrictions imposed by the SGP. Fiscal discipline has to achieve a balanced budget between government revenues and current expenditure, including depreciation cost and capital maintenance, but excluding investment expenditure. The application of this ‘golden rule’ would add flexibility to the process, favour those government expenditures that increase future revenues and prevent the negative effects of the SGP on the growth potential of the economies. Calmfors and Corsetti (2003) have argued against the ‘golden rule’, which, according to them, makes the rules of SGP less transparent.10 They have claimed that ‘there is no theoretical reason why a golden rule should apply to physical capital investment, but not to investment in human capital or to other expenditure increases or tax reforms that will generate future revenues. But extending a golden rule in this way makes it impossible to operate: the political-economy risks that fiscal laxity could then always be justified would simply be unmanageable’ (Calmfors and Corsetti, 2003: 7). They have proposed conditioning the deficit ceilings to the debt levels of each country, presenting a table in which higher deficit ratios, numerically specified, are associated with lower ranges of debt ratios. As an alternative to this proposal, Calmfors and Corsetti (2003), like Buti et al. (2002), have suggested maintaining the 3 per cent deficit ratio, but allowing countries with low debt ratios to use extra-budgetary ‘rainy-day funds’, which can be established by channelling government surpluses into them during upswings in order to use them for stabilization purposes in downswings. This rule would enhance the use of counter-cyclical policies and favour the countries with low debt ratios. Finally, Calmfors and Corsetti (2003), like Calmfors (2005), have underlined the importance of having an effective enforcement of the SGP and have proposed transferring the decisions against the states violating

198 Fiscal Policy in the European Union

the rules from the political level of the ECOFIN to the judicial level of the European Court of Justice. Another rule proposed to reform the SGP is that envisaged by Buiter and Grafe (2004), named the ‘Permanent Balance Rule’, which identifies a tax rate that can guarantee a sustainable long-term path for the government debt. The tax rate is calculated by introducing the condition that government debt must not be greater than the present value of the future primary surpluses. For Buiter and Grafe (2004: 77) this rule has several advantages. It allows discretion within the limits set by the envisaged solvency condition; it takes into account the inflation and growth rates foreseen for each country; and it amends and extends the golden rule by considering within the solvency condition all future sources of revenues affected by public investment. Yet, this proposal too introduces arbitrary elements when determining the future value of primary surpluses. A second group of reform proposals has been advanced by some authors (Wyplosz, 1999, 2002; Casella, 2001; Pisani-Ferry, 2002; Von Hagen and Mundschenk, 2003; Hein and Truger, 2006) who consider it necessary to modify the institutional organization. For them, the term ‘coordination’ refers to the set of arrangements and activities aiming at the identification of a unified framework for monetary and fiscal policies and the introduction of commitments on policy decisions at national and super-national level. The main difficulty faced by these authors is to propose changes that are acceptable to reluctant and non-cooperative actors. This preoccupation affects their proposals, introducing into them some puzzling elements. Wyplosz (2002) compares the organization of monetary and fiscal policy in recent years. He argues that ‘the crucial change that has rehabilitated monetary policy has been the move from rule design to institutional reform’ (Wyplosz, 2002: 5). In the 1970s the conduct of monetary policy was restricted by ‘rules’ (e.g. that on the rate of growth of the money supply), which proved difficult to implement. In the subsequent years, the tendency has been to replace these ‘rules’ with delegation to institutions endowed with independence and a specific objective to achieve. Fiscal policy, Wyplosz (2002: 5) says, is following similar lines with some delay. It is currently dominated by ‘rules’, which are difficult to implement, and there are already discussions to replace the ‘rules’ with delegation to newly created institutions. The mechanism of delegation is based on a principal-agent relationship over the achievement of a specific objective. Price stability is that chosen in monetary policy. Fiscal policy, Wyplosz says, has to fulfil two main tasks. The first, which is concerned with the structure of the tax

Coordinating Monetary and Fiscal Policies 199

system and the composition of expenditure, directly redistributes benefits among groups of citizens. The immediate political implications of this task make it impossible to delegate it to an agent (see Wyplosz, 2002: 8). The second is concerned with the dimension of the budget balance. It generates macroeconomic effects, which vary in intensity but are not fundamentally different from those generated by monetary policy decisions.11 Like the latter, decisions concerning this task can be delegated to an agent, provided that it is endowed with technical independence and has been assigned the achievement of a specified objective. For Wyplosz (2002: 9) this second task of fiscal policy should be conducted by giving a mandate to ensure ‘debt sustainability’ over the appropriate horizon to a new institution, named the Fiscal Policy Committee (FPC). This institution, which is similar to the Monetary Policy Committee (MPC), founded to conduct monetary policy, can provide competent judgement and generate better results than any conceivable rule: ‘Competent and dedicated policymakers are better able than quantitative ceilings and rules to exercise good judgement and deliver the adequate mix of restraint and flexibility. To do so, however, they must be shielded from temptation and pressures that are part of political life’ (Wyplosz, 2002: 14). For Wyplosz (2002: 9), the Fiscal Policy Committees should be set up at national level, should decide the annual deficit figure in percentage of planned GDP, and should have no power over the composition of the government budget. The committees should approve the budget bill before it becomes law and, in the event of abrupt changes in economic conditions, they should ask the Parliament to revise it. According to Wyplosz, the FPCs can also enhance coordination among national fiscal policies and between fiscal and monetary policies at the Euro area level: It is fair to conclude that formal coordination of fiscal policies is unlikely to be agreed upon in the foreseeable future. Informal exchanges, on the other side, are highly desirable as they may help avoid the most grievous mistakes. Such an informal approach would be much easier to organize among independent, non-political FPCs than it currently is among Finance Ministers. The coordination between national governments and the ECB has not been satisfactory so far, largely because the ECB insists on keeping governments at arm’s length. Here again, it can be expected that the ECB will find it less threatening to entertain informal contacts with like-minded, independent FPCs. (Wyplosz, 2002: 13)

200 Fiscal Policy in the European Union

Wyplosz’s proposal specifies the institutional changes to be made and the tasks to be performed. Yet, by avoiding interventions at the euro area level, it plays down super-national coordination and the relevance of a policy mix. Moreover, his proposal does not clarify why political bodies, which are reluctant to lose further control over fiscal policy decisions, should prefer doing it in favour of a technocratic national entity rather than in favour of a super-national entity, for instance the Eurogroup, in which they are directly represented. Pisani-Ferry (2002) underlines the need to strengthen coordination at the euro area level and to identify the right policy mix in the conduct of monetary and fiscal policies. His main proposal is ‘the transformation of the Eurogroup into a collective executive body with the ability to make decisions by qualified majority voting’ (Pisani-Ferry, 2002: 4). His suggestions regarding the way in which this change should be done raise two doubts. The first regards the need that decisions taken at the euro area level are not subsequently contradicted by the choices made at national level. Pisani-Ferry (2002: 4) proposes that the Eurogroup should formulate and adopt an economic policy charter for all its members and that ‘the interaction between European procedures and national budgetary procedures should be both streamlined and reinforced in order to make sure that domestic policy decisions are consistent with commitments made in Brussels’. This claim is, however, opposed by the subsequent statement that this code of conduct should not be binding, but should represent a common understanding on policy principles, a solution that preserves the current situation in which, as said above, the lack of binding constraints drives the national governments and the ECB towards non-cooperative behaviours. The second doubt regards the coordination between monetary and fiscal policies and the forms and the degrees of independence that the central bank should be given in this context. Pisani-Ferry (2002) rightly recalls the problems coming from an excessive degree of independence attributed to the central bank, like that on goals and priorities enjoyed by the ECB. He claims that ‘an ECB that would be perceived as acting in isolation, without the backing of its shareholders, would lack legitimacy. An ECB that would be perceived as the only policy player of the EMU system would risk being held responsible for whatever does not work in the Eurozone – which could only result in a clash with public opinion’ (Pisani-Ferry, 2002: 9). Later, however, he questions whether the code of conduct adopted by the Eurogroup should include elements of the ECB strategy and, instead of noticing that the technical independence of the central bank is not undermined by the identification of a

Coordinating Monetary and Fiscal Policies 201

satisfactory policy mix for the Euro area, he claims that ‘the ECB would certainly be reluctant to participate in an exercise that might jeopardise its independence’ (Pisani-Ferry, 2002: 13). This conclusion fails to distinguish between different forms of central bank independence and to appreciate that it is the ‘goals’ and ‘priority’ independence enjoyed by the ECB that makes it come to be perceived as the only policy player of the EMU system. The identification of a policy mix does not undermine the ‘technical’ independence that the ECB enjoys, like all other central banks. The lack of clarity on these topics reduces the strength of this proposal, which, again, tends to preserve the current conditions that drive the national governments and the ECB to adopt non-cooperative behaviours. It overlooks the suggestion of Von Hagen and Mundschenk (2003), recalled above in Section 4, that to reform the institutional organization of policy coordination in the euro area it is necessary to modify the incentives that affect the attitude of the national governments and the ECB.

6 A new scheme The analysis presented in the previous sections clarifies the limits of the existing arrangements of policy coordination in the euro area and suggests what should be changed to avoid future problems and a loss of legitimacy for the European institutions. Moving from these conclusions we present in what follows a proposal to reform these arrangements. The first objective to achieve is the transformation of the Eurogroup from a body where exchange of information and reciprocal understanding are realized to one where • the economic situations of the euro area and of the economies

belonging to it are analysed; • the fiscal and financial conditions prevailing in these countries are

commonly verified; • the preferences of the central banks and of the national governments

over monetary and fiscal policy decisions are made explicit; • a common stance on these policies is identified; • and the constraints to be put on these decisions at decentralized lev-

els are commonly decided and binding commitments on them are credibly assumed by the central bank and the national governments. The Eurogroup thus has to be transformed into a Euro Area Fiscal Agency. The Agency must combine political and technical abilities and must have

202 Fiscal Policy in the European Union

formal decision authority on matters regarding the coordination of monetary and fiscal policies. Its governing board has to be composed of the representatives of the European Commission, the ECB and the national governments, in the same way as the Eurogroup. Its president can be nominated in the same way as that of the Eurogroup.12 He or she would set the agenda of the monthly meetings, in which the current situation is monitored, the targets and requirements of monetary and fiscal policies are identified, and formal decisions on all matters regarding policy coordination are taken. To perform its functions the Agency must be endowed with qualified technical staffs, appointed by the national governments. They must elaborate, in collaboration with the EFC and the ECB, the analytical basis upon which discussions are carried out at both technical and political level and decisions are taken by the governing body of the Agency. A new reform of the rules regarding the economic content and the enforcement of the EDP and the SGP must accompany the transformation of the Eurogroup. The new arrangements have to provide incentives for all actors to enter binding commitments on coordination of monetary and fiscal policies and to make these commitments credible. The ECB and the national governments are the agents to which these incentives must be offered. Those regarding the ECB must guarantee the maintenance of its technical independence and the possibility of considering credible the commitments on policy decisions entered into by the national governments. The reform thus has to defend the ‘technical’ independence of the central bank and must be able to persuade the national governments to have a cooperative attitude in order to induce the ECB to conduct monetary policy in a way that is more attentive to the needs of the different economies of the area. On the other side, the national governments must be persuaded to have a cooperative attitude by granting them the maintenance of fiscal sovereignty and by giving the impression that the ECB, through the participation and the involvement in the decision process of the Agency, will be more attentive to the specific needs of their economy and that they are reducing their uncertainty about the fiscal behaviour of the other governments. Further incentives to cooperate can be found in the opportunity to overtake, under specific and authorized circumstances, the 3 per cent ratio between deficit and GDP, in the possibility of funding the activities authorized in excess of the 3 per cent ratio through the Agency, and in the fact that, if the national governments do not comply with the prescriptions of the new reform, they will certainly undergo the sanctions it envisages.

Coordinating Monetary and Fiscal Policies 203

To achieve these results the new reform of the EDP and the SGP may maintain the no bailing out condition, the 60 per cent ratio between government debt and GDP and the 3 per cent ratio between actual deficit and GDP. In this case, it must eliminate the lack of transparency on the economic content of the rules introduced by the 2005 reform by avoiding reference to a ‘cyclically adjusted’ budget to be kept close to balance. The new reform must also stipulate that a country can be authorized to overtake the 3 per cent ratio for cyclical and structural reasons. During a recession, the Agency can identify whether and by how much the national policies can exceed this ratio, thus avoiding the situation in which the national governments, by breaking the stipulated rules during downturns, withdraw legitimacy from the European institutions. Moreover, when the Agency identifies that an economy has specific structural problems, like those related to high government debt and loss of international competitiveness, the national government can be given a permit to deduct from its expenditure funds which may be used to introduce measures, agreed with the Agency, aiming at the solution of these problems. As a further incentive the Agency, with the assistance of the ECB, can raise funds in the financial markets for the solution of these structural problems and make them available to the national governments. To strengthen the legitimacy of the European institutions the new reform, as proposed by Calmfors (2005: 81–2) must ask the national parliaments to hold public discussions on the economic, financial and fiscal situation of their country with experts appointed by the European Commission. Finally, as yet another way to strengthen the incentives to cooperate, the new reform must also make the enforcement of the EDP and the SGP strict and effective. The several steps of the enforcement have to be clearly specified and must be applied automatically by an authority such as the European Court of Justice. The procedures should be set in motion by the European Commission, which asks the Agency, with the assistance of the EFC and the ECB, to prepare a report on the existence of the violation. The report is sent to the President of the Agency, who asks its governing body to take a formal decision on it. If the existence of the violation is recognized, the Agency sends the country a recommendation on the correcting measures with two deadlines regarding the adoption of measures and the correction of the violation, as was prescribed by the original SGP. At the same time, the Agency officially informs the European Court of Justice that an excessive deficit procedure has been opened. The European Commission reports to the Court on whether the member state has respected the deadlines and, in the case of failure,

204 Fiscal Policy in the European Union

asks the Court to give notice to the country and to make a non-interest bearing deposit with the Commission, whose amount is related to the extent of the violation. If, after two years, the violation has not been corrected, the European Court of Justice asks for the conversion of the deposit into a fine. The transformation of the Eurogroup into an Agency taking formal decisions on policy coordination in the euro area, combined with a new reform of the EDP and the SGP, which introduces transparent rules on the economic content of these procedures and a set of incentives able to persuade the relevant actors to cooperate, including a stricter enforcement of these rules, can lead to satisfactory outcomes of the economic policies carried out at centralized and decentralized levels and can enhance the stability and the growth potentials of the economies belonging to this area.

7 Conclusions The current organization of policy coordination in the euro area is unsatisfactory and there are concerns that, after the rules of the SGP have lost effectiveness with the changes introduced in 2005, irresponsible behaviours will dominate during the next cyclical downturn, damaging the economy and the legitimacy of the Union. Many reform proposals have been presented recently to deal with these problems. A first group has focused on the economic content and the enforcement of the rules of the SGP, looking for a satisfactory balance between transparency and flexibility. Yet, this balance is difficult to find. A second group of proposals has focused instead on the institutional organization of policy coordination in order to induce the national governments and the ECB to have a cooperative attitude and to identify a common stance on monetary and fiscal policies, to which they commit themselves. Those who have developed the second group of proposals have adopted a broad definition of coordination. They believe that ‘rule design’ is not sufficient to achieve satisfactory policy outcomes and that, as has happened in the organization of monetary policy, there is a tendency, generated by the need to find suitable solutions to these problems, to replace ‘rule design’ with the delegation to new institutions, endowed with technical independence and a clearly specified objective, in the organization of fiscal policies too. Following this line of argument, we have presented a proposal to reform the current organization of policy coordination in the euro area, whose central element is the transformation of the Eurogroup into a Euro

Coordinating Monetary and Fiscal Policies 205

Area Fiscal Agency, an institution able to combine political and technical abilities and endowed with formal decision-making authority on matters regarding policy coordination. The proposal has taken in several ideas presented by the existing literature and has tried, at the same time, to maintain the links with the post-Keynesian tradition, in which, owing to the close interaction between monetary and real variables, coordination between monetary and fiscal policy is crucial when dealing with both cyclical and structural problems.

Notes 1. According to Arestis and Sawyer (2005), a strict interpretation of the requirement that national governments (and hence the fiscal authorities) should not exert any influence on the ECB (and hence the monetary authorities) would rule out any attempt at coordination of monetary and fiscal policies. 2. The independence of a central bank may include its ability to decide how to use the instruments of monetary policy (‘instrument’ or ‘technical’ independence), the power to decide the level of the objective variables (‘goal’ independence), the power to decide the relative importance of different objectives, i.e. the weights to be attributed in the social loss function to the output gap and to inflation (‘priority’ independence), and other forms of independence (see Panico and Rizza, 2004). Most central banks enjoy ‘technical’ independence. The ECB also enjoys ‘priority’ and ‘goal’ independence. 3. See Commendatore et al. (2003) and Panico (2003). 4. The Memorandum referred to the determination of the rate of profit proposed in Kaldor (1955–6). 5. Steedman (1972) provided the first formal presentation of the role of the government sector in this theory. Later on, Fleck and Domenghino (1987), Pasinetti (1989) and others analysed this issue in several respects (for a review of this debate, see Panico, 1997). 6. See You and Dutt (1996), Lavoie (2000) and Dos Santos and Zezza (2004). 7. When the Presidency of the CMEU falls on a non-euro state, the President’s participation in the ECB Governing Council is represented by the chairman of the Eurogroup. 8. The non-euro countries present instead a ‘convergence programme’. 9. In the original regulations of the EDP and the SGP, a fall of 2 per cent of the GDP automatically allowed the overtaking of the 3 per cent ratio, whereas a fall of 0.75 per cent could represent an exceptional case allowing the overtaking if the ECOFIN recognized it. The 2005 reform allows overtaking if there is ‘a protracted period of very low growth relative to potential growth’. The ambiguity of the new regulation is clear. As to the fact that the declaration of ‘excessive deficit’ has to ‘take into account all other relevant factors’, while the previous regulation did not specify which factors could be considered, the new reform recalls factors like ‘potential growth, prevailing cyclical conditions, the implementation of policy in the context of the Lisbon agenda

206 Fiscal Policy in the European Union and policy to foster R&D and innovation’ as well as ‘fiscal consolidation efforts in good times, debt sustainability, public investment and the overall quality of public finances’. Moreover, attention has to be paid to ‘any other factors which, in the opinion of the Member State concerned, are relevant in order to comprehensively assess in qualitative terms the excess over the reference value’. See Calmfors (2005: 61). 10. See also Buti et al. (2002) and Calmfors (2005). 11. From a Keynesian perspective, changes in the interest rates and in the demand coming from the government sector generate redistributive effects too. These effects, however, operate through complex market mechanisms related to the interaction between aggregate demand and supply. 12. Following Pisani-Ferry (2002: 4) it can be proposed to appoint the President for a fixed and not too short a period.

References Alesina, A., O. Blanchard, J. Gali, F. Giavazzi and H. Uhlig, ‘Defining a Macroeconomic Framework for the Euro Area’, Monitoring the European Central Bank, 3 (2001), CEPR, London. Angeriz, A. and P. Arestis, ‘Monetary Policy in UK’, Cambridge Journal of Economics, 31(6) (2007): 863–84. Arestis, P. and M. Sawyer, ‘What is Wrong with the Euro Area Monetary Model?’, in G. Fontana and R. Realfonso (eds), The Monetary Theory of Production: Tradition and Perspectives (Basingstoke: Palgrave Macmillan, 2005). Arestis, P. and M. Sawyer, ‘The Case for Fiscal Policy’, in P. Arestis, J. Ferreiro and F. Serrano (eds), Financial Developments in National and International Markets (Basingstoke: Palgrave Macmillan, 2006). Beestma, R. and X. Debrun, ‘Implementing the Stability and Growth Pact: Enforcement and Procedural Flexibility’, ECB Working Paper, 433, January (2005). Blanchard, O. J. and F. Giavazzi, ‘Improving the SGP through a Proper Accounting of Public Investment’, CEPR Discussion Paper Series, 4220 (2004). Blanchard, O. J. and L. F. Katz, ‘What Do We Know and Do Not Know about the Natural Rate of Unemployment?’, Journal of Economic Perspectives, 11(1), Winter (1997): 51–72. Blinder, A. S., ‘Distinguished Lecture on Economics in Government: What Central Bankers Could Learn from Academics – and Vice Versa’, Journal of Economic Perspectives, 11(2), Spring (1997): 3–20. Buiter, W. H. and C. Grafe, ‘Suggestions for Enhancing Fiscal Sustainability and Macroeconomic Stability in an Enlarged European Union’, Economics of Transition, 12 (1) (2004): 67–102. Buti, M., S. Eijffinger and D. Franco, ‘Revisiting the Stability and Growth Pact: Grand Design or Internal Adjustment’, mimeo (2002), available at: http://www. cepii.fr/anglaisgraph/communications/pdf/2002/211102/franco.pdf Calmfors, L., ‘What Remains of the Stability Pact and What Next?’, Swedish Institute for European Policy Studies, Report No. 8, November (2005). Calmfors, L. and G. Corsetti, ‘How to Reform Europe’s Fiscal Policy Framework’, mimeo (2003), available at: http://rincewind.iies.su.se/∼lcalmfor/ wejjan16calmfors.pdf

Coordinating Monetary and Fiscal Policies 207 Canzoneri, M. and B. Diba, ‘The Stability and Growth Pact: a Delicate Balance or an Albatross?’, Empirica, 26 (1999): 241–58. Casella, A., ‘Tradable Deficit Permits’, in A. Brunila, M. Buti and D. Franco (eds), The Stability and Growth Pact: the Architecture of Fiscal Policy in EMU (Basingstoke: Palgrave Macmillan, 2001). Commendatore, P., S. D’Acunto, C. Panico and A. Pinto, ‘Keynesian Theories of Growth’, in N. Salvadori (ed.), The Theories of Economic Growth: a Classical Perspective (Aldershot: Edward Elgar, 2003), 103–38. Dos Santos, C. H. and G. Zezza, ‘A Post-Keynesian Stock-flow Consistent Macroeconomic Growth Model: Preliminary Results’, The Levy Economic Institute of Bard College, Working Papers, 402 (2004). Fatàs, A. and I. Mihov, ‘On Constraining Fiscal Policy Discretion in EMU’, Oxford Review of Economic Policy, 19(1) (2003): 112–31. Fleck, F. H. and C. M. Domenghino, ‘Cambridge (UK) versus Cambridge (Mass.): a Keynesian Solution to the “Pasinetti Paradox”’, Journal of Post-Keynesian Economics, 10(1), Fall (1987): 22–36. Franco, D., F. Balassone and M. Francese, ‘Fiscal Policy in Europe: the Role of Fiscal Rules’, mimeo (2005), available at: http://www.ief. es/investigacion/ Recursos/Seminarios/EconomiaPublica/2004_11Febrero_1.pdf Galbraith, J. K., ‘Time to Ditch the NAIRU’, Journal of Economic Perspectives, 11(1), Winter (1997): 93–108. Gordon, R. J., ‘The Time-varying NAIRU and its Implications for Economic Policy’, Journal of Economic Perspectives, 11(1), Winter (1997): 11–32. Harrod, R. F., ‘An Essay in Dynamic Theory’, Economic Journal, 49, March (1939): 14–33. Hein, E. and A. Truger, ‘Fiscal Policy and Macroeconomic Performance in the Euro Area: Lesson for the Future?’, IMK Working Paper, 7 (2006). Kaldor, N., ‘Alternative Theories of Distribution’, Review of Economic Studies, 23(2) (1955–6): 83–100. Keynes, J. M., The General Theory and After: a Supplement, in D. E. Moggridge (ed.), Collected Writings of J. M. Keynes, vol. XXIX (London: Macmillan, 1979). Lavoie, M., ‘Government Deficits in Simple Kaleckian Models’, in H. Bougrine (ed.), The Economics of Public Spending: Debts, Deficits and Economic Performance (Aldershot: Edward Elgar, 2000). Panico, C., Interest and Profit in the Theories of Value and Distribution (London: Macmillan, 1988). Panico, C., ‘Government Deficits in the Post-Keynesian Theories of Growth and Distribution’, Contributions to Political Economy, 16 (1997): 61–86. Panico, C., ‘Old and New Growth Theories: What Role for Aggregate Demand?’, in N. Salvadori (ed.), Old and New Growth Theories: an Assessment (Aldershot: Edward Elgar, 2003), 53–66. Panico, C. and M. O. Rizza, ‘Central Bank Independence and Democracy: a Historical Perspective’, in R. Arena and N. Salvadori (eds), Money, Credit and the Role of the State: Essays in Honour of Augusto Graziani (London: Ashgate, 2004). Pasinetti, L. L., ‘Government Deficit Spending is Not Incompatible with the Cambridge Theorem of the Rate of Profit: a Reply to Fleck and Domenghino’, Journal of Post-Keynesian Economics, 11(4), Summer (1989): 640–7.

208 Fiscal Policy in the European Union Pisani-Ferry, J., ‘Fiscal Discipline and Policy Coordination in the Eurozone: Assessment and Proposals’, mimeo (2002), available at: http://www.pisaniferry.net/base/re02-gea-discipline-vmai.pdf Samuelson, P. A., ‘Recent American Monetary Controversy’, Three Banks Review, 29, March (1956): 3–21. Solow, R. M., ‘Towards a Macroeconomics of the Medium Run’, Journal of Economic Perspectives, 14(1), Winter (2000): 151–8. Staiger, D., J. H. Stock and M. W. Watson, ‘The NAIRU, Unemployment and Monetary Policy’, Journal of Economic Perspectives, 11(1), Winter (1997): 33–49. Steedman, I., ‘The State and the Outcome of the Pasinetti Process’, Economic Journal, 82(328), December (1972): 1387–95. Stiglitz, J., ‘Reflections on the Natural Rate Hypothesis’, Journal of Economic Perspectives, 11(1), Winter (1997): 3–10. Von Hagen, J. and S. Mundschenk, ‘Fiscal and Monetary Policy Coordination in EMU’, International Journal of Financial Economics, 8 (2003): 279–95. Wyplosz, C., ‘Economic Policy Coordination in EMU: Strategies and Institutions’, mimeo (1999), available at: http://hei.unige.ch/∼wyplosz/cw_bonn_final.pdf Wyplosz, C., ‘Fiscal Discipline in EMU: Rules or Institutions?’, mimeo (2002), available at: http://hei.unige.ch/∼wyplosz/gea_0402.pdf You, J. I. and A. K. Dutt, ‘Government Debt, Income Distribution and Growth’, Cambridge Journal of Economics, 20 (1996): 335–51.

Index NB: Page numbers in bold refer to figures and tables

see also under Stability and Growth Pact (SGP)

Aarle, B. van 77, 86 Aaron, H. 113 active fiscal policy 25, 66, 86 adjustment models, expansionary fiscal 90 Afonso, A. 31, 86, 89 Afonso, O. 7, 13–14, 20 Aghion, P. 88 Ai, C. 115 Alesina, A. 26, 31, 86, 125, 145, 196 Allington, N. 14 Alves, R. H. 7, 11, 13–14, 20–1 ‘ambiguous coalition’ 138 Amsterdam Council 26, 49 André, C. 141 Angelopoulos, K. 88–9 Angeriz, A. 185 Anglo–Saxon model 175, 176, 179 anti-Keynesian behaviour 32–3 Ardagna, S. 31 Arellano, M. 123 Arestis, P. 13, 29, 53, 85 coordination and 185, 186, 188 Argimon, I. 114 Aschauer, D. A. 88–9 Aschauer’s specification 116–19 public capital and 111, 112–15, 120, 122, 125 Asian currencies 165 Asian workers 175 Atkinson, P. 89 Augmented Dickey-Fuller test 117, 117, 133 Australia 30, 33, 114 Austria 43, 56, 90, 96–7, 119 economic divergences and 160–1, 163–6, 169, 174 automatic stabilizers 134, 196 fiscal deficit and 142

‘bad times’, French fiscal stance and 142 Bahl, R. 8 Baier, S. 88 Bajo-Rubio, O. 114 balanced/surplus budget rule 27, 39, 43–4, 43 Balassa-Samuelson effect 162–3, 172 Balladur, Prime Minister Édouard 141 Barro, R. J. 34, 88–9 Barroso, J. M. 17 Barry, F. 31 Bartolini, L. 27 Bean, C. 53 Beestma, R. 184, 196 Begg, I. 13 Belgium 30, 119 economic divergences and 161–4, 166, 168–9, 176, 180 public expenditures and 90, 92, 96–7, 104 Benalal, N. 160 Berndt, E. R. 112 Bertola, G. 31 binding rules, fiscal 157 Blanchard, O. J. 31, 111, 132, 188, 197 Bleaney, M. 89 Blinder, A. S. 53, 188 Bloom, D. 89 Blundell, R. W. 123 Bolkestein Directive 158, 178 Bond, S. R. 123 Brender, A. 137–8 Briotti, M. G. 31, 86 209

210 Index Broad Economic Policy Guidelines (BEPGs) 192 economic divergences and 157, 180 public expenditures and 84, 86, 87 Brück, T. 26–7 Brunila, A. 27 budget balance 71–2, 74: France 148–50 balanced/surplus rule 27, 39, 43–4, 43 central 10 ‘close to balance/surplus’ 27 community 15, 17–18 cyclical component 39 cyclically adjusted 196, 203 deficit rule 40, 46–8, 46 full-employment 49n ‘hard constraints’ 16 policy changes, France 144–52 process, France 150, 151 structural balance target 40–1, 46–8, 47 Budget Acts (France), constitutional bylaw on 150–1 Buiter, W. H. 12, 14, 198 Burns, A. 131, 145 business cycles 25, 35, 58 effects 123 France 133–44, 138, 143 Buti, M. 13–14, 26–7, 131, 137 coordination and 184–5, 196–7 Calmfors, L. 53, 184–5, 194, 196–7, 203 Cameron, D. R. 89 Canada 30, 112 Canzoneri, M. 196 Cardiff process 192 Carton, B. 26 Casella, A. 13, 198 Cassou, S. P. 115 catching-up countries 163, 171–2 causality bias, public investment and 114–15 central bank independence 200–1, 205n rates 164

see also European Central Bank (ECB) Chabert, G. 151 Chamley, C. 88 China 165 Chirac, Jacques 144 Cobb-Douglas production function 112, 113, 116 Code for Fiscal Stability (UK) 149 Coe, D. T. 114 ‘cohabitation years’, France 137–9 cointegration, public and private capital 114 Colbert, J.-B. 181 Collignon, S. 13 Cologne Process (‘Macroeconomic Dialogue’) 192 Common Agricultural Policy (CAP) 14, 15 common monetary policy 61, 63, 65, 172, 185 ‘common rules of behaviour’ 193 competence sharing 9 competitiveness 6, 9 economic divergences and 164–5, 171–2, 174, 177–8 stabilization and 70, 71, 73–4, 74, 80n compliance conditions 41 Conférence Nationale des finances publiques 152 consolidation effects 28–34 ‘fatigue’, France 142 successful 30–4 Constitutional Treaty, proposed 21 continental Europe social protection model 169, 175, 180 contrats jeunes en entreprises 146 Contribution Sociale Généralisée (CSG) 147 coordination 176, 184–206 arguments in favour of 185–9 institutional arrangements 189–91 literature on 195–201 new scheme 201–4 overview 4–5 procedures 191–5 Corsetti, G. 196–7

Index Council of Economics and Finance Ministers (ECOFIN) 87, 189–91, 192–4, 198 Council of Ministers of the European Union (CMEU) 189–90 CRDS (Contribution pour le Reimboursement de la Dette Sociale) 147 credit specialization, principle of 151 Creel, J. 13, 26, 111, 133, 134–5 Cresson, Prime Minister Édith 139, 142 Crowder, W. J. 114 currency areas, theory of optimal 10 ‘cyclical discretion’ 134 cyclical position, single monetary policy and 63 cyclically adjusted budget 196, 203 primary deficit 136 De Grauwe, P. 13 de Haan, J. 112–14, 117, 120 de Villepin, Prime Minister Dominique 141–2, 143, 148 Debrun, X. 184, 196 ‘debt sustainability’ 199 decentralization 8–9 Decentralization Bills (France) 149 deficit evolution, fiscal 91 ‘deficit runners’ see France delegation 198 demand shock, restrictive 70–4, 71, 74 ‘democratic deficit’, EU and 20, 21 Denmark 30–4, 161, 169 Deroose, S. 171, 176 Devarajan, S. 88–9 Devereux, M. B. 31 Diba, B. 196 disequilibrium index 171 process 166 Doménech, R. 61 Drazen, A. 31, 137–8 Dürnecker, G. 89 ‘early warnings’ 192 Easterly, W. 89

211

economic crisis, in France (1993) 131 economic divergences 157–82 euro area 158–71: disparities 171–6 overview 4 policy and 176–82 Economic and Financial Committee (EFC) 190, 192–3, 202–3 economic growth 6 public capital impact 119–24 ‘Economic Outlook’ (OECD) 116, 135, 141 economic performance, EU-15 170 economic policies binding 173 renationalization of 179–80 Economic Policy Committee (EPC) 190 EDP (Excessive Deficit Procedure) 12, 167, 192–5, 202–4 education 111 Eisner, E. 112, 114 electoral cycle, France 133–44, 138 deficits and 137–42 emplois jeunes 146 employment 161 protection indicators 170 rates 161, 161 strategy 162, 178–9 employment protection legislation (EPL) indicator (OECD) 169 EMU see European Economic and Monetary Union (EMU) endogenous growth theory 88–9, 92 Enrich, P. D. 8–9 environmental factors 178 ‘federalism’ 8 equalization, fiscal 11 equilibrium process 166 Error Correction Model (ECM) 117 ‘escape clauses’, SGP and 18 Esping-Andersen, G. 169 ‘Essay in Dynamic Theory’ (Harrod) 187 estimation automatic stabilizers 37–8, 41, 42, 43, 45–7 two-step 37–41

212 Index euro area 158–71 explaining disparities 171–6 weaknesses 172–3 see also coordination Euro Area Fiscal Agency 181 Eurogroup as 201–4, 205 Eurogroup (euro zone) 18, 105n, 180, 189–91, 200 as Euro Area Fiscal Agency 201–4, 205 European Agreement on Employment 192 European Central Bank (ECB) 18, 26, 68, 79n, 86 coordination and 185, 188–92, 195–6, 199–204 economic divergences and 157–8, 163, 181 stabilization and 53, 54, 56, 60–3, 69, 70, 76 European Commission (EC) 13, 131, 189, 190 coordination and 174, 192–4, 202–3 economic divergences and 160, 164, 168, 176–8, 180–1 fiscal federalism and 17–18 public expenditures and 86–90 SGP and 26, 28, 38, 78n European Community (EC), harmonization 147 European Constitution 19 European Council 11, 12–13, 55, 87, 180, 184, 192 European Court of Justice 194, 198, 203–4 European Economic and Monetary Union (EMU) 119 economic divergences in 157–82 France and 137, 142, 145, 147, 150, 152 public expenditures and 84–5, 85–90, 90–101, 102, 105 SGP and 25–6, 28, 33–8 passim, 43–4, 46, 48 stabilization and 53–6, 58, 60, 62, 64–6, 70, 73, 75 see also coordination ‘European federation’ 6

‘European scheme of fiscal transfers’ 19 European Social Model (ESM) 158, 168, 181–2 crisis of 175–6 European System of Central Banks (ESCB) 78n, 185, 190–1 European Union (EU) 6, 27, 158, 168, 189, 190 EU-15 169, 170, 177 EU-27 180–1 Eurosystem 185, 190 Evans, P. 115, 121–2 Excessive Deficit Procedure (EDP) 12, 167, 192–5, 202–4 exchange rates 164–5 real 56, 64, 65, 71–2 expansionary fiscal contractions theory 28, 30–1 expenditure, ‘golden rule’ 197–8 Fabius, Prime Minister Laurent 142 fairness, of budget process in France 151 Fatàs, A. 184, 196–7 Favarque, E. 28, 38 Federal Reserve (USA) 163, 188 ‘Federation of Nation States’, EU as 7, 9 Feldstein, M. 10 Fève, P. 133 Figueras, D. 12 Finland economic divergences and 160–1, 164–5, 167, 169, 174 public expenditures and 90, 92, 96 SGP and 33, 35, 37, 44–5, 47 Finn, M. 115 fiscal cycles, theory of 26 fiscal federalism 6–21 instruments of 10 overview 2 progress 15–20 rules 7 theory of 7–11: evolution of 16–19 fiscal policies active 25, 66, 86

Index alternative rules 69, 73 criticisms of 2, 25 current 12–15 domestic 172 EMU 167 perspectives 2007–13 15 principles of national 85 rules 66 Fiscal Policy Committees (FPCs), proposed 197, 199 fiscal theory of price level (FTPL) 26 Fisher test 122 Fitoussi, J. P. 111, 172 Flatters, F. 9 Flavin, A. 133 flexibility 178–9 Flores de Frutos, R. 114 Florio, M. 90 Fölster, S. 89 Fontana, G. 30 Ford, R. 113 foreign direct investments (FDI) 166 France 13, 90, 92, 129–54 budget policy changes 144–52 economic divergences and 159, 161–4, 166–9, 174, 176 electoral/business cycles 133–44, 138, 143 fiscal surpluses 135 political cycles 140 public capital and 111, 113–15, 116–17, 118, 119 public deficits/debt 4, 130–3, 130, 149 SGP and 44–5, 45, 47 violations 193–4 Franco, D. 185, 196 free market strategy 158 full-employment budget 49n functions, sharing of 8 G7 countries 115 Galbraith, J. K. 188 Galí, J. 27, 77, 134 Galli, E. 137 Garretsen, H. 33, 86 GDP government budget balance and 130

213

government net capital 110 growth rates 159 rate of growth 58 trend and 59 Gemmell, N. 88–9 General Theory and After: a Supplement (Keynes) 187 Generalized Moments Method (GMM) 115, 119, 123 estimator 123 Germany 13, 32, 131, 133, 148 economic divergences and 157, 159–67, 169, 171, 174, 176 public capital and 111, 113, 115, 116–17, 118, 119 public expenditures and 90, 92, 96–7 reunification 159 SGP and 41, 43–4, 45, 47: violations 193–4 stabilization and 54, 56, 58, 60, 62–5, 63, 69 Giavazzi, F. 30–3, 86, 111, 197 Gilbert, G. 149–50 Girouard, N. 141 Giudice, G. 86 globalization 175, 179 Glomm, G. 88 ‘GNP resource’ 14 Goglio, A. 131, 145 ‘golden rule’ current/investment expenditure 197–8 French local government 149 ‘good times’, French fiscal stance and 142 Goodhart, C. 19 Gordon, R. J. 188 governance reform, France 150–2 government liabilities, France 144 government net capital 110 Grafe, C. 198 Gramlich, E. M. 115 Greece 56, 58, 137 economic divergences and 157, 159, 161–7, 169, 171–6 passim public expenditures and 90, 92, 96–7, 104

214 Index Greece (Continued) SGP and 35, 37, 41, 43, 45, 47: violations 193 gross public debt 153n Grossman, G. M. 88 growth GDP rate of 58, 59 key factors 111 rates 58 theory, endogenous 88 Guengant, A. 150 guidelines, integrated 179 Gupta, S. 88, 90 Gurgand, M. 111 Gwartney, J. 89 Hamilton, J. 133 Hammond, G. 19 Hansson, B. 112 ‘hard budget constraints’ 16 Harmonized Competitiveness Indicators 80n Harrod, R. F. 187 Hausman, J. A. 115, 122 health expenditures 146 Hein, E. 18 Helpman, E. 88 Hemming, R. 30, 34, 86 Hénin, P. Y. 133 Henrekson, M. 89 Heun, E. 185, 198 Himarios, D. 114 Hjelm, G. 33 Holtz-Eakin, D. 121 Howitt, P. 88 Huffschmid, J. 181 Hughes-Hallett, A. 142 Hulten, C. R. 113, 120 human capital 111 Hurlin, C. 121 income sharing 11, 19 inflation economic divergences and 162, 162, 163, 165, 173 stabilization and 56, 57, 58, 64–5, 67, 70–3 Inman, R. 11

innovation 177–8 INSEE (National Institute for Statistics and Economic Studies) 133, 141 institutional arrangements 189–91 instruments of fiscal federalism 10 of fiscal policy 8 variables method 123–4 integrated guidelines 179 Interest Rate Misalignment (IRM) 62 interest rates 64, 162 euro area 163 real 63–4, 69, 71–2 inter-jurisdictional competition 8 INTERLINK model 27 International Monetary Fund (IMF) 63, 65, 79n Ireland 56, 58 economic divergences and 157, 159–64, 168–9, 171 SGP and 30–4, 35, 37, 41 stabilization and 90, 92, 96, 101 Italianer, A. 10, 19 Italy 56, 58, 131, 133, 137, 148 economic divergences and 157, 159–65, 169, 171, 174, 176 public capital and 111, 115, 116, 118–19, 118 public expenditures and 90, 96, 101 SGP and 35, 43, 45, 47: violations 193 Japan 33, 165, 167 Johansen test 118, 118, 121 Jones, C. I. 88 Jones, L. E. 88 Jospin, Prime Minister Lionel 139, 145, 148 Juppé, Prime Minister Alain 142, 144 Kaldor, N. 187 Kamps, C. 116 Karras, G. 115, 121–2 Katz, L. F. 188 Kenen, P. B. 10 Keynes, J. M. 13, 28, 158, 181 Keynesian analysis 186–7

Index Keynesian effects 33, 86–7 contrary 29–30, 32–3 King, D. 10 King, R. G. 88 Kneller, R. 88–9, 124 Kocherlatoky, N. 89 Kopits, G. 14 Kumar, M. S. 86 Kwiatowski, Phillips, Schmidt and Shin (KPSS) test 133 labour markets demand 173 flexible 176 Lamfalussy, A. 17 Lamo, A. 89 Lau, S. H. P. 114 Laurent, E. 133 Laws of Finance (France) 148, 151 Le Bihan, H. 133 Le Cacheux, J. 111, 133, 172, 174 ‘left-wing’ governments, fiscal stance of 137–9, 141 legitimacy 6 Levin, A. 120 liberal reforms 172 liberal social protection model 169, 180 Linn, J. 8 Lisbon Agenda 15, 55, 88, 157–8, 162, 177, 179, 194 Lisbon Council 109 Lisbon Strategy 84, 87, 109, 125, 176–7, 182 Lisbon Treaty 21 local governments, France 146, 148–50, 149 long-term equilibrium 61 Lucas, R. E. 88 Luxembourg public expenditures and 90, 92, 96–7, 104 SGP and 28, 35, 38 ‘Luxembourg process’ 192 Maastricht Treaty 1, 11, 12, 17, 48, 78n consolidation process 125

215

criteria 142, 143, 153, 197 EDP and 194–5 fiscal federalism and 15 fiscal rules of 26, 84, 85 public debt and 130–2, 153n MacDougall, D. 17 ‘Macroeconomic Dialogue’ (Cologne Process) 192 macroeconomic factors applied 109 cycles 137 performance, France 147 stabilization 10, 12, 15, 54 strategy 172, 177–8 Mannumeas, T. 114 market economy 11 market-oriented view 176 ‘market-preserving federalism’ 8 Martin, P. 110 Martinez-Mongay, C. 27 Mathieu, C. 167, 172 Mauroy, Prime Minister Pierre 142 McCombie, J. 14 McDermott, C. J. 86 McKinnon, R. 8, 11, 16 Mediterranean social protection model 169 Mélitz, J. 19 Mendoza, E. 89 Migaud, D. 150 Mihov, I. 184, 196–7 Milesi-Ferretti, G. M. 26 Mitterrand, President 139 Moghadam, R. 114 Monetary Policy Committee 199 monetary policy, single 61, 63, 65, 172, 185 monetary rules 66 monetary union model 66–9 Mongelli, F. P. 64 More Active Fiscal Rule (MAFR) 68–9, 71–4, 75–6, 77 multinational panels 115 Mundell, R. A. 10 Mundschenk, S. 184–5, 191, 198, 201 Munnell, A. H. 112 Musgrave, R. 11 Myrdal, G. 26

216 Index Nadiri, M. 114 NAIRU (non-accelerating inflation rate of unemployment) 29, 186, 188–9 national fiscal policies, principles of 85 National Institute for Statistics and Economic Studies (INSEE) 133, 141 national policies, non-cooperative 173–5 natural rate of unemployment (NRU) 188 neo-Ricardian framework 28 net public debt 153n Netherlands 44–5, 63 economic divergences and 157, 159–66, 168–9 public capital and 114, 119 public expenditures and 90, 92, 96, 101 SGP violations and 193–4 New Consensus 29–30, 53, 78n, 85, 186 in Macroeconomics (NCM), elements of 29, 186 new industrial economies 165 new member states (NMS) workers 175 Niechoj, T. 18 NITC bubble 165, 178 non-accelerating inflation rate of unemployment (NAIRU) 29, 186, 188–9 non-cooperative national policies 173–5 non-Keynesian effects, fiscal policy and 86 non-monetary policies, coordination of 16 non-optimal Nash equilibrium 174 Nourzad, F. 89 Oates, W. 7–9, 11 OECD (Organization for Economic Cooperation and Development) economic divergences and 160, 163, 169

fiscal consolidation and 32 France and 131, 135, 141, 145, 150 public capital and 112, 115, 116, 120 SGP and 28, 30, 33–4, 35 Ohlin, B. G. 26 open method of coordination (OMC) 157 optimal currency areas theory 10 Optimal Monetary Area theory 64 ordinary least squares method 122, 122 Otto, G. D. 114 output gap 67–9, 70–1, 73–4 growth 166, 173 over-identifying restrictions, test of (Sargan) 123 Oxley, H. 110 Padovano, F. 137 Pagano, M. 30–3, 86 Palacio-Vera, A. 30 Panico, C. 186 participation, citizens and EU 6, 11 Pauly, M. 10 ‘Permanent Balance Rule’ 198 Perotti, R. 31, 86, 125, 134, 142, 145 Philips-Perron unit root test 38 Pisani-Ferry, J. coordination and 184–5, 198, 200–1 fiscal federalism and 10, 13, 18–19 policy mix 200 political factors business cycles 137 cycles, France 40, 134 weight, of EU 6 Pomzi game 132 Poret, R. 113 Portugal economic divergences and 157, 159–66, 168–9, 172, 174, 176 public expenditures and 90, 92, 96–7, 101, 104 SGP and 35, 41, 43, 45–7: violations 193 stabilization and 56, 58, 63

Index post-Keynesian perspectives 185 theory of growth and distribution 187 poverty 162 ‘trap’ 148 primary cyclical balance 35–7 France 136 output gap and 36 vs output gap 37 primary general government balance, cyclically adjusted 135–7 Prodi, Romano 13 ‘productive’ expenditures 87–90, 97, 101 Programmes pluriannuels de finances publiques 152 public capital 109–25 Aschauer’s specification 116–19 economic growth impact 119–24: methodological criticism 112 elasticity for 112, 115 literature review 111–16 overview 3–4 studies 113, 114 public deficits/debt France 4, 130–3, 130, 149: main components 134–6 Maastricht definition 153n public expenditures 29, 84–105 cuts 145–8, 168 EMU and quality of 90–101 EMU theoretical bases 85–90 evolution of 102 fiscal deficit evolution and 91 overview 3 primary 145–6, 168 productive/unproductive 87–90, 103, 104 size and type of 93, 95, 96, 98–101 taxation and 91 public finance sustainability 168 ‘Public Finances in EMU 2001’ (European Commission) 87 public investment 92 see also public capital public services, general 97

217

public spending see public expenditures Radcliffe Commission, Memorandum to (Kaldor) 187 Raffarin, Prime Minister Jean-Pierre 141–2, 143, 148 ‘rainy-day funds’ 197 Ram, R. 112 Ramsey, D. 112 Ratner, J. B. 112 Rebelo, S. 88 redistribution function 9–10 reforms, structural and liberal 172–3 renationalization of economic policies 179–80 restrictive demand shock 70–4, 71, 74 revenues, non-tax, France 144, 147 Ricardian behaviour 33 Ricardian equivalence hypothesis 34, 49n, 53 ‘right-wing’ governments, fiscal stance of 137–9, 141–2 robustness test 124 Rocard, Prime Minister Michel 139, 142 Rocard–Cresson era 139 Romer, D. 72 Romer, P. M. 88 Romero de Avila, D. 89 Roubini, N. 137 Rubinfeld, D. 11 Rubio, O. 12 Sachs, J. 10, 137 Sala-i-Martin, X. 10, 88 Samuelson, P. A. 188 Sanchez-Robles, B. 89 Sargan’s test 123 savings rates 166–7 ratios 166 Sawyer, M. 13, 29, 53, 85 coordination and 185, 186, 188 Say’s Law 186 Scandinavian social protection model 169, 175, 179, 180

218 Index Schclarek, A. 33 Schmidt-Phillips test 133 Schuman, Robert 6 Schwab, R. M. 8, 113, 120 Seitz, H. 115 Sekkat, K. 27 SGP see Stability and Growth Pact Shah, A. 8 short-term changes 61 Sin, C. 114 Siné, A. 142 single currency 17, 184 macroeconomic policies and 157 see also euro area Single European Act 134 single monetary policy 61, 63, 65, 172, 185 Slovenia 28, 35, 38 small and medium enterprises (SMEs) 182 Smith, S. 19 social expenditures 92, 97, 146 social-Keynesian strategy 158 ‘social partners’ 150 social protection 92, 169 systems, Bismarckian 158 social security institutions 150 ‘soft enforcement’ 191–2 Solow, R. M. 16, 25, 188–9 Sosvilla-Rivera, S. 114 Spahn, P. B. 8–10 Spain 63, 114, 131 economic divergences and 157, 159–66, 168–9, 171–2, 174, 176 public expenditures and 90, 92, 96–7 SGP and 41, 45–6 stabilization and 54, 56, 58, 60, 62, 64–5, 69 Stability and Growth Pact (SGP) 1, 71, 125 active fiscal policy rule and 73 automatic fiscal stabilizers and 25–50, 134: consolidation effects 28–34; data and 34–7; overview 2; results 41–9; two-step estimation 37–41 coordination and 192–5, 196–8, 202–4

current position 11, 12–14 economic divergences and 157–8, 167, 174, 179, 180 fiscal federalism and 17–18, 20 fiscal rule (SGPFR) 68, 72–3, 75–6 France and 131, 148, 152 public expenditures and 84, 85, 88 reform of (2005) 195 stabilization and 53–4, 55, 75–6 stability programmes 18 stabilization 53–81 macroeconomic 10, 12, 15, 54 mechanisms 18–19, 60–6, 72 monetary union model 66–9 overview 3 policy 172 programmes 180, 192 restrictive demand shock 70–4, 71, 74 short-run 54–9 see also Stability and Growth Pact Staiger, D. 188 stationarity bias 113–14 test 117, 117, 133 Sterdyniak, H. 134–5, 167, 172 Stiglitz, J. 188 Strauch, R. 89 structural factors balances 39 ‘discretion’ 134 reforms 172–3 Sturm, J. E. 112, 112–14, 117, 120 subsidiarity, principle of 9, 19 surplus budget rule 27, 39, 43–4, 43 fiscal, France 135 primary 133: structural 133 sustainability tests 132–3 Sutherland, A. 31 Swaroop, V. 89 Sweden 33, 112 economic divergences and 161, 165, 168–9 Switzerland 176 Symansky, S. 14 Tabellini, G. 26 Tatom, J. A. 113–14, 120

Index taxation 27, 29, 88–9, 91 competition 148 cuts, France 148 increases, France 145–8 rates 198 smoothing theory 196 Taylor, J. B. fiscal rule 38–41, 163 fiscal stabilizers and 37–8 rule 60–2, 69, 76, 164 univariate regression and 28 Thatcher, Prime Minister Margaret 175 Thöne, M. 89 Tiebout, C. 9 Tilford, S. 176 TINA (There is no alternative) 175 trade-offs 11 transversality condition 133 Treasury (US) 188 Treaty on European Union see Maastricht Treaty Treaty of Rome 7 Truger, A. 18, 185, 198 two-step estimation 37–41 unemployment 160, 176 France and 129 NAIRU 29, 186, 188–9 natural rate of (NRU) 188 rates 161, 162, 169 United Kingdom (UK) 113, 185 economic divergences and 159–61, 163, 166–9 France and 131, 133, 148, 149 United States of America (USA) 38, 133, 148, 188 economic divergences and 159, 163, 165–7, 176, 177 federal budget 14, 17 public capital and 112–14, 121 US dollar 165

219

univariate regression 28, 38 ‘unproductive’ expenditures 88–90, 97, 101 ‘unsustainability’, risks of 133 Usher, D. 11 Van Aarle, B. 33 van den Noord, P. 26–7, 38, 89, 131, 137 VAT decrease in France 147–8 resource 14 Vavouras, I. S. 137 Vijverberg, V. P. M. 112 Von Hagen, J. 19, 184–5, 191, 198, 201 Vori, S. 19 Voss, G. M. 114 Vrieze, M. D. 89 wages 163–4 flexibility 176 policies 172, 174 restraint 146 Wagner law 115 Walrasian myth 176 Weingast, B. 8, 11, 16 welfare gains, identification of 8 Wescott, R. F. 86 Wilcox, D. 133 Wrobel, M. V. 10 Wylie, P. J. 112 Wyplosz, C. 13, 172 coordination and 184–5, 191, 197–200 Yi, K. M. 89 Zagler, M. 89 Zou, H. F. 89 Zweiner, R. 26–7