Social Policy in the Smaller European Union States 9780857452641

Citation preview

Social Policy in the Smaller European Union States

Studies in Contemporary European History Editors: Konrad Jarausch, Lurcy Professor of European Civilization, University of North Carolina, Chapel Hill, and a Senior Fellow of the Zentrum für Zeithistorische Studien, Potsdam, Germany Henry Rousso, Senior Fellow at the Institut d’historie du temps présent (Centre national de la recherché scientifique, Paris) and co-founder of the European network “EURHISTXX” Volume 1

Between Utopia and Disillusionment: A Narrative of the Political Transformation in Eastern Europe Henri Vogt Volume 2

The Inverted Mirror: Mythologizing the Enemy in France and Germany, 1898–1914 Michael E. Nolan Volume 3

Conflicted Memories: Europeanizing Contemporary Histories Edited by Konrad H. Jarausch and Thomas Lindenberger with the Collaboration of Annelie Ramsbrock Volume 4

Playing Politics with History: The Bundestag Inquiries into East Germany Andrew H. Beattie Volume 5

Alsace to the Alsatians? Visions and Divisions of Alsatian Regionalism, 1870–1939 Christopher J. Fischer Volume 6

A European Memory? Contested Histories and Politics of Remembrance Edited by Małgorzata Pakier and Bo Stråth Volume 7

Experience and Memory: The Second World War in Europe Edited by Jörg Echternkamp and Stefan Martens Volume 8

Children, Families, and States: Time Policies of Childcare, Preschool, and Primary Education in Europe Edited by Karen Hagemann, Konrad H. Jarausch, and Cristina Allemann-Ghionda

SOCIAL POLICY IN THE SMALLER EUROPEAN UNION STATES

( Edited by

Gary B. Cohen, Ben W. Ansell, Robert Henry Cox, and Jane Gingrich

Berghahn Books NEW YORK • OXFORD

First published in 2012 by Berghahn Books www.berghahnbooks.com ©2012 Gary B. Cohen, Ben W. Ansell, Robert Henry Cox, and Jane Gingrich All rights reserved. Except for the quotation of short passages for the purposes of criticism and review, no part of this book may be reproduced in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system now known or to be invented, without written permission of the publisher.

Library of Congress Cataloging-in-Publication Data Social policy in the smaller European Union states / edited by Gary B. Cohen ... [et al.]. p. cm. -- (Studies in contemporary European history ; vol. 9) Includes bibliographical references and index. ISBN 978-0-85745-263-4 (alk. paper) -- ISBN 978-0-85745-264-1 (ebook) 1. European Union countries--Social policy. I. Cohen, Gary B., 1948HN373.5.S6287 2012 320.6094--dc23 2011020287

British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library.

Printed in the United States on acid-free paper

ISBN: 978-0-85745-263-4 (hardback) ISBN: 978-0-85745-264-1 (ebook)

CONTENTS

( List of Tables and Figures Preface

vii

ix

Introduction: Social Policy in the Smaller EU States

1

Gary B. Cohen, Ben W. Ansell, Robert Henry Cox, and Jane Gingrich Section I. The Social Investment Agenda: From Ideas to Policy?  1. How Globalization and the European Union are Changing European Welfare States  17 Robert Henry Cox  2. Family Policies, Education, and Female Labor Market Participation in Advanced Capitalist Democracies  33 Robin Stryker, Scott R. Eliason, Eric Tranby, and William Hamilton  3. Double Transformation: How to Adjust Institutional Social Policy?  59 Juho Saari  4. The Social Investment State: A New Trend in Social Expenditure or Merely a Popular Political Discourse?  81 Jorma Sipilä Section II. Interest Coalitions, Ideas, and Social Reform  5. Multiple Market Prescriptions: The Diverse Models of Health Care Reform in Sweden  107 Jane Gingrich

vi | Contents

 6.  Austrian Social Policy Reform in the Era of Integration and Rising Populism  131 Reinhard Heinisch  7. Of Firms and Flexibility: The Dynamics of Collective Bargaining Reform in Spain and Portugal  153 Sara Watson Section III. Diverging Institutional Legacies, Ideas, and Social Reform  8. Social Policy Change “Under the Radar Screen”: Health Care Reforms in Seven Small Countries  179 Kieke G.H. Okma, Luca Crivelli, Toni Ashton, Iva Bolgani, Tsung-Me Cheng, David Chinitz, Meng-Kin Lim, Hans Maarse, Rachel Meislin, and Tim Tenbensel  9. Humboldt Humbled? The Germanic University System in Comparative Perspective  215 Ben W. Ansell  10 Beyond the Welfare State: Consumer Protection and Risk Perceptions in the European Union and Austria  237 Paulette Kurzer Conclusion: Ideas and Social Reform Robert Henry Cox Notes on Contributors Index

271

261

TABLES AND FIGURES

( TABLES Table 2.1. Bootstrap EDF Estimates of compliers’ average causal effects of aggregate female education levels on female labor force participation  43 Table 2.2. Bootstrap EDF Estimates of compliers’ average causal effects on rates of female labor force participation, by aggregate female education levels, with strong left political tradition as the instrument in the CACE analysis  45 Table 2.3. Bootstrap EDF estimates of compliers’ average causal effects on within-country change in female labor force participation log rates, by aggregate female education levels, with strong left political tradition as the instrument in the CACE analysis  46 Table 2.4. Bootstrap EDF estimates of compliers’ average causal effects on rates of female labor force participation, by aggregate female education levels, with other political traditions as the instrument in the CACE analysis  47 Table 2.5 Bootstrap EDF estimates of compliers’ average causal effects on within-country change in female labor force participation log rates, by aggregate female education levels, with other political traditions as the instrument in the CACE analysis  48 Table 3.1. Three waves of social policy in Finland, 1945–2007  64 Table 3.2. Governance of social risks: the division of responsibilities (2006)  65 Table 3.3. Agenda setting during the primeval soup  70 Table 3.4. Emerging paradigms, agendas, and frames since the primeval soup  72 Table 4.1. Changes in social expenditure as percent of GDP, standardized by size of population age 0 to 14 and 65 plus, 1980–2003  88

Table 4.2. Social expenditures on families, education, and old age, 1991, 1998, and 2003  89 Table 4.3. Changes in social investment and family expenditure 1998–2003  91 Table 4.4. Child care, fertility, and female employment  95 Table 4.5. Correlations between public expenditure, female employment, and fertility, 1998  97 Table 4.6. Correlations between public expenditure, female employment, and fertility, 2003  97 Table 4.7. Correlations between public expenditure, female employment, and fertility, 1998–2003  97 Table 7.1. Key features of collective bargaining in Spain and Portugal, 1980s and 1990s  156 Table 9.1. Higher education attainment in the population, 2004  216 Table 9.2. The trilemma in higher education  218 Table 10.1. Health issues: official level of concern  239 Table 10.2. Smoking rates, men/women in 2005 and 1985–87 (percent of population in selected countries). Ranked from high to low as of 2005  240 Table 10.3. Ranking of selected European countries: tobacco-control measures (2007).  244 Table 10.4. Attitudes toward agricultural biotechnology, (percent) Austrian respondents  248 FIGURES Figure 3.1. Social expenditure, year-to-year percentage in real terms, percent, and social expenditure, percent of GDP  60 Figure 3.2. Paradigm, agendas, and frames during the primeval soup  67 Figure 4.1. Public expenditure on families and old age in twenty-four OECD countries, as percent of GDP, 1998–2003  88 Figure 7.1. Proposed changes to collective bargaining in Spain and Portugal  158 Figure 7.2. Percent of workers covered by firm-level contracts, 1990–2006  165 Figure 9.1. Gross enrollment as a percentage of age cohort, private spending as a percentage of total spending, and public cost as a percentage of GDP, 2002  219

PREFACE

( In recent decades, public social, educational, and health services have been among the most highly contested aspects of social policy in the European Union member states. Governments have had to respond to rising public demands for accountability, local community control, lower tax burdens, and greater freedom of individual choice. The forces of economic globalization have buffeted the existing welfare systems and the underpinning corporatist understandings. In the meantime, the growing regulatory functions of the EU have created pressures for more uniformity in social policies, and the EU expects more stringent controls on state budgets than when the welfare states were first built between the late 1940s and the 1970s. Working in the opposite direction, though, the 1992 Maastricht Treaty enshrined the principle of subsidiarity to ensure that public decisions are taken as closely as possible to the level of the individual citizen and with constant checks to assure that public action is justified in light of the available possibilities. Much has changed in European social policy during the last twenty-five years, but as the essays in this volume demonstrate, the changes have been more complex and subtle than one might initially suppose. Despite the pressures for a retreat from the traditions of the European welfare state, there has been more change and adaptation of social insurance programs and social services than the simple reduction or termination of programs. The smaller member states of the European Union offer particularly interesting laboratories for studying the dynamics of social policy, and, as will be seen, some of the smaller states have pioneered major innovations in social policy, such as “flexicurity” in Denmark and the concept of the “social investment state” in several of the Nordic countries. The essays in this volume represent revised and expanded versions of a selection of papers from the international conference, “Social Policy in the New Europe: The Experience of Austria and the Smaller EU Members,”

x | Preface

held at the University of Minnesota, Twin Cities. The University of Minnesota Center for Austrian Studies and the European Studies Consortium organized the conference, and the College of Arts and Sciences Scholarly Events Fund, the Government of Finland/Speer Fund, the Humphrey Institute of Public Affairs, the Center for the Study of Politics and Governance, the Center for German and European Studies, the School of Social Work, the Department of Sociology, and the Department of Political Science generously supported the event as cosponsors. The Austrian Federal Ministry of Science and Research graciously provided travel support for several of the Austrian participants. The editors are particularly grateful to the graduate research assistants at the Center for Austrian Studies, Lisa Peschel, Mollie Madden, Eric Roubinek, and Edward Snyder, for their patient and dedicated efforts in preparing the manuscript for publication. We are also grateful to Konrad Jarausch and Henry Rousso for accepting this volume for publication in the series they edit, “Contemporary European History,” and to Marion Berghahn and her colleagues at Berghahn Books for their continuing commitment to publish a wide range of scholarship in European studies.

INTRODUCTION Social Policy in the Smaller EU States

( Gary B. Cohen, Ben W. Ansell, Robert Henry Cox, and Jane Gingrich

In Europe and around the world in recent years, political, economic, and social changes have placed increasing pressure on social policy and welfare services. The integrative forces of globalization have strained longstanding national social policies and institutions. Ever-growing flows of goods and capital around the world have seemingly threatened the ability of governments to guarantee both the corporatist agreements protecting job security and the welfare entitlements that developed in many countries after World War II. International market integration has strengthened neoliberal ideological arguments, which have called ever more insistently since the 1970s for retrenchment in welfare guarantees and services. Neoliberal initiatives, in turn, have provoked efforts by supporters of the welfare state to defend and improve the social safety net, disseminating new models of “social investment” that claim to combine extensive social protection with support for economic growth. Just as Europe was a leader in the development of the welfare state and the supportive structures of corporatist politics from the 1920s onward, in recent decades it has been a bellwether for the particular stresses from globalization. While many popular analyses over the last twenty years have talked about the end of the European welfare state and of the corporatist concept of social partnership, academic literature traces a reality of more complex change and adaptation in social insurance and social services in most

2 | Gary B. Cohen, Ben W. Ansell, Robert Henry Cox, and Jane Gingrich

European countries. The aging of native populations, the arrival of new waves of immigrants, the changing roles of women in the workforce, and the structural economic changes connected to deindustrialization have created urgent needs for new educational and social services. Growing demands for accountability, community control, lower taxes, and individual choice have forced legislators and governmental agencies to decentralize or partially privatize the administration of many social services, but, on balance, educational and social services of all types have continued to grow in most countries. Social policy in Europe has displayed both resiliency and adaptation under the stresses of globalization and neoliberal reform. Yet those who follow these developments seldom provide systematic attention to the smaller European countries. Debates in the United Kingdom, Germany, France, and sometimes Italy over basic issues of political economy often attract wide international attention. Attention to the smaller countries—Belgium, Denmark, Austria, or Finland—emerges in single case studies or as illustrative examples of change, but rarely as part of a systematic analysis. Because the smaller countries are more sensitive to shifts in the global economy, they are seemingly forced to respond to demographic and economic changes earlier than are the bigger countries. Yet the experience of the smaller countries is highly varied. Sometimes they serve as models for reform, undertaking experiments that only later gain the attention of stymied reformers in the larger countries. However, at other times these countries fail to enact significant change, despite substantial economic and political pressure to do so. Taken together, these varied experiences can shed much light on the politics of welfare-state reform. The chapters of this volume analyze the hotly debated issues of change in social policies and the welfare state in a number of the smaller countries of Europe. They focus attention on a range of reforms: those that cut back entitlements; those that aim to maintain them; and those that claim to “modernize” welfare states. In so doing, they draw out the reasons some work and others fail, engaging in broader debates about the nature and causes of policy change. Together, these chapters make three basic points about the politics of change in advanced welfare states. First, they demonstrate that new ideas often provide a “coordination point” around which agreements for reform are reached. Second, new ideas about social policy may emerge, but they require political support coalitions to implement them and, equally, can be derailed by oppositional coalitions. Third, the success of reform also depends on the institutional and historical legacies within countries; where subnational veto players or preexisting policy frame-

Introduction: Social Policy in the Smaller EU States | 3

works exist, even a powerful political coalition may be unable to enact any policy ideas, however ostensibly appealing. The book will draw out these themes. For the remainder of this introduction, we seek to highlight the theoretical and practical significance of the smaller countries. The following chapters will then draw out these lessons.

Why the Smaller European Countries Matter What are the stakes of examining the smaller countries of Europe? Why might we expect them to behave differently than the larger states, which have traditionally captured academic interest? We argue that the smaller countries of Europe have both substantive and analytical significance that ought not be neglected. Most simply, the smaller countries are only “small” in individual terms: in aggregate, their population actually accounts for almost 60 percent of Europe’s population. In that sense, the citizens of smaller countries are a European majority. Moreover, such countries can be substantively important policy innovators. Ideas from “flexicurity” to mass public higher education to flat-rate income taxation have originated in Europe’s smaller countries. Analytically, these countries are important because they are at once more dependent on large structural forces like globalization than are the larger countries but also display a broad degree of variation in responses. Hence these countries have long been seen by at least some scholars as analytically intriguing. Both academic observers and the popular press have long noted several crucial features that draw Europe’s smaller states together, distinguishing them from the larger countries. First, Europe’s small states have more open economies (Katzenstein 1985; Garrett 1998). In order to compensate for small domestic markets, these countries have historically traded extensively with other countries, with exports plus imports typically amounting to well over half their national income. This approach has not only generated wealth, but also income volatility. Second, most of Europe’s small states have developed consensual political systems. Political scientists and historians studying the Nordic countries and the low countries of the Netherlands, Belgium, and Luxembourg, as well as Austria, Ireland, and non-EU member states such as Switzerland and Norway have long noted their tendency to employ power-sharing arrangements within government and across social actors in the economy (Lijphart 1968; Powell 1982). Third, many of these countries developed large welfare states in the postwar period. Like high trade dependence and consensual political systems, large welfare states are not exclusive to the small countries,

4 | Gary B. Cohen, Ben W. Ansell, Robert Henry Cox, and Jane Gingrich

but the small countries do tend to spend more on social policies and have developed extensive systems of social protection (Hicks and Swank 1992). In 1985 Peter Katzenstein published a classic book on international trade and political economy in small states. Small States in World Markets argued that the conjunction of open economies, consensual political systems, and large welfare states in Europe’s small countries was no accident. Europe’s small countries’ dependence on global markets exposed citizens to the vagaries of the world economy. This vulnerability and its political consequences led both market actors and governments to cooperate to mitigate the consequences. The result was that small countries developed systems of social protection and labor market institutions that protected workers, alongside systems of corporatist bargaining that provided mechanisms for resolving social tensions. The world, though, has changed since Katzenstein first wrote. Small countries have always traded extensively, but since the mid-1980s, flows of foreign investment among countries have exploded, as have the options for multinational production. Moreover, the nature of production has radically shifted, as all countries have moved away from manufacturing and toward a service-based economy with lower rates of growth. Furthermore, since the mid-1980s the European Union (EU) has expanded its regulation of economic affairs both in and between member states. While small states remain relatively wealthy, the combination of growing global economic openness, slower domestic growth rates, and the rise of global governance structures has put new pressures on their systems of social protection. The exact nature of these pressures, though, is hotly debated. One line of reasoning in recent scholarship portrays social policy in small states as in “big trouble” (Schwartz 1994). Proponents of this argument point to the twin pressures of globalization and Europeanization as posing a particular challenge to the welfare state in small countries. The combination of growing capital mobility with a longstanding dependence on international trade means that these countries’ economic health is increasingly linked to that of global capital. As firms and investors find it easier to locate in regions where tax rates and labor costs are lower, welfare states that guard workers against the vagaries of the market become untenable. These processes are at play everywhere, but they emerge most vigorously in the highly globalized economies of Europe’s small states, which simply cannot afford to ignore the demands of global capital. These market pressures combine with those emanating from the EU. Unlike the larger countries, small states have less direct influence on the rules issued by the EU, and they lack the ability to ignore its dictates. Globalization and Europeanization thus reduce the autonomy of small states and push them in the direction of market-conforming policies.

Introduction: Social Policy in the Smaller EU States | 5

While the above argument emphasizes the changing global environment as forcing convergence among small countries, a second line of thinking emphasizes the crucial role of domestic institutions in preventing change. Many observers argue that small countries’ consensual political systems and corporatist economic governance, which created a framework for social accommodation during the postwar era, make change difficult in the current era. Governance systems that involve corporate actors—such as unions, employers, and social groups—in the policy process have many “veto points” over policy and give voice to those wishing to block change. These institutions not only empower those who wish to defend the status quo, but they also bestow particular advantages on existing strategies. Small states have often developed particular economic production niches, such as high-value goods, which rely on specific labor market and welfare institutions (Streeck 1991). As different actors coordinate around these strategies, existing structures may exhibit an “increasing returns” logic, which makes change more difficult over time (Pierson 2000). While proponents of this view debate whether these processes constitute positive resilience in the face of global pressures or create sclerotic environments that leave small states unable to adjust to the demands of the global economy, they uniformly predict that change itself is unlikely. A third line of scholarly thinking presents small states as nimble innovators on the forefront of reform that neither replaces nor simply replicates existing structures of social protection. This work builds on Katzenstein’s original insights, emphasizing the link between global economic demands and the need for compensatory domestic policies. It suggests that social protection in the small counties remains alive and well and, in fact, is a necessary consequence of both their historical trajectory and their position in the global economy. However, as the needs of the global economy change, so, too, must the systems of social protection. Instead of protecting workers from unemployment caused by changes in global demand, governments must facilitate the conditions of employment by ensuring that workers possess competitive skills (Boix 1998) and do not face barriers to economic participation, such as insufficient child care (Estevez-Abe 2006). Recognizing these shifts, small countries have innovated, developing more flexible labor markets while also emphasizing active labormarket policy, investment in human capital and skills, and programs to address the needs of women, immigrants, and other groups with historically lower rates of labor-market participation. Academic observers hold up the successes of Denmark, the Netherlands, Finland, Ireland, and other small countries as a model for Europe, demonstrating that an emphasis on “social investment” in skills can in fact create a “win-win-win” situation

6 | Gary B. Cohen, Ben W. Ansell, Robert Henry Cox, and Jane Gingrich

that combines high growth, global competitiveness, and social inclusion (Iversen 2005). These different theoretical approaches share one commonality: they all presume that “smallness” has a more or less uniform effect on policies. Global forces drive all small countries toward consensual politics and extensive welfare provision; all small countries are at risk from globalization and Europeanization and hence are in “big trouble”; or all small countries are nimble innovators. Despite these uniform expectations, these arguments often draw on regionally and historically specific case studies. Nordic countries, accordingly, are often the inspiration for studies of leaders and innovators, whereas Southern European small states are considered to be “laggards” and stuck with poor-performing states particularly vulnerable to cutbacks. The small continental countries fall between these stereotypes: neither basket cases nor pathbreakers but exemplars of a consensual, albeit stagnating, government. This volume, in examining small countries from each of these groups— the Nordic countries, continental Europe, and Southern Europe—shows that the variation within groups of small countries, and among small countries more generally, is extensive. Sometimes Southern European states are innovators—as in the case of Portugal’s labor-market reform. Some continental countries adopt radical reforms, whereas in others, reform withers on the vine. It is crucial then, that we take a new look at the experience of the smaller European countries, asking both what unites them analytically but also what explains the very real differences in social policy reform among them. In understanding varying paths to social policy reform among the smaller countries of Europe, we need to ask what kinds of reforms have been on the agenda. This volume shows that two strands of policy types have emerged on the agenda of European states since the early 1990s: “market liberalization” and “social investment.” The first idea, “market liberalization,” marks an array of policies, building on the privatizations of the 1980s, which extend markets into the public sector itself. Examples include the creation of internal markets among public hospitals, school choice, vouchers for elderly care, and other implementations of market incentives in public services. The concept of “social investment,” by contrast, sees a much more extensive role for an active state. It marks a turn away from government spending that supports consumption and aggregate demand management—for example pensions, unemployment insurance, and public ownership of industry—to public spending that supports individuals in entering and succeeding in the private sector. Investment in human capital and active labor-market policies, moving beyond passive income transfers and toward benefits that aim to help individuals be more

Introduction: Social Policy in the Smaller EU States | 7

competitive in the changing labor market, are the hallmarks of this strategy. Typical policies include education spending, child-care spending, and parental leave. This marked a recalibration of strategies for center-left parties—as global and electoral pressures prevented widespread government intervention with market outcomes, these parties turned toward policies that impacted market opportunities. Though these two sets of reforms appear antagonistic, both diffused widely among European countries during the 1990s and the first decade of the twenty-first century. The articles in this volume thus examine the adoption, or lack thereof, of both social investment and market-liberalization reforms among the smaller European countries. Previous research on small countries has tended to focus on “social consumption” policies like pensions and redistributive spending, and where it has analyzed market reforms, it has done so in terms of regulation of the private sector. Instead, the analyses in this volume examine a distinct set of policy areas. In terms of social investment, various chapters examine education investment, active labormarket policy, child care, maternity policies, and public health. In terms of market liberalization, chapters examine the introduction of markets in health care, university financing, and labor-market deregulation. Thus, this volume moves the analysis of social reform in the smaller countries of Europe to a contemporary set of reforms quite distinct from those that motivated the writings of Katzenstein. In both examining a broader set of smaller countries and a broader range of policies than much of the literature on welfare reform, this volume provides a unique window onto the broader theoretical debates around the fate of the welfare state and the politics of reform.

Ideas, Institutions, and Coalitions: A Theory of Constrained Innovation The articles in this volume not only examine the development of “social investment” and “market liberalization” policies but also situate them in the broader context of the diffusion of ideas about social policy. In recent years there has been a resurgence of scholarly interest in the role of ideas in guiding economic and social reforms. Broadly, scholars agree that ideas can be usefully conceptualized as “focal points” around which political actors can coordinate their understandings of policy problems, share potential solutions, and develop support coalitions that enable policy enactment (Béland 2009; Culpepper 2008; Weyland 2008). The emergence of ideas about social investment and market liberalization provides a lens into the process of policy emergence and diffusion. Innovative policies

8 | Gary B. Cohen, Ben W. Ansell, Robert Henry Cox, and Jane Gingrich

commonly emerge in response to perceived failures of the existing policy framework. In the European case, both the “social investment” and “market liberalization” policy agendas originated in response to the diagnosis of “Euro-sclerosis” as European states experienced both stagnant growth and high unemployment through the 1980s and 1990s. These agendas were packages of ideas, from respectively left and right, about how government might best respond—offering policymakers a common understanding of the problems facing the country and a set of policies that promised particular distributional and economic outcomes that actors could coordinate around. The diffusion of these ideas through transnational communications between bureaucrats, scholars, business leaders, and think tanks and the process of learning from early adopters help to explain the eventual adoption of new social policy reforms. Together the articles show that ideas like “social investment” and, conversely, “market liberalization” have often been first created by or diffused to the smaller European countries. Smallness can be a virtue in terms of policy innovation. Whereas larger countries have commensurately larger bureaucracies or several federal subunits with high degrees of independence, many smaller countries are able to enact major policy changes driven by central policy makers much more rapidly. Sensitivity to international trade and Europeanization also provides a demand from these countries for policy innovations that enable them to maintain social policies under increasing global economic pressure. However, social policy reforms do not emerge fully formed like Athena from Zeus’s head. They are the product of prevailing ideas about viable social policies that are themselves prior to any actual reforms. The process of idea diffusion is not frictionless, however. Two important constraints affect both how these ideas diffused and whether they diffuse at all. First, policy ideas can only be implemented if they have the support of a large enough coalition of political and private actors. Sometimes ideas like social investment find interest in a dominant center-left party and a business community seeking workers with more productive or flexible skills. Sometimes, however, no such demand exists and the idea withers on the vine. Second, preexisting political and policy institutions determine whether an idea can be effectively implemented. Where policies must pass a blockade of veto players, policy makers may struggle to implement popular policy ideas. Similarly, when a policy idea meets preexisting policy institutions that are incompatible, the idea may not take root. Simply put, the diffusion of social policy ideas, even among the smaller countries, is dependent on the existence of permissive political coalitions and institutions.

Introduction: Social Policy in the Smaller EU States | 9

Lessons from the Smaller Countries: Chapter Outline The chapters of this book show that policy reforms in the smaller countries of Europe demonstrate more complex outcomes than existing arguments would suggest. Collectively, the authors show that the experience of European—and non-European—small countries is highly varied. In no case, even in the face of what appear to be extensive reforms conforming to market pressures, has social protection disappeared or even been dramatically eroded. Equally, though, the small countries have hardly been immune to change. While some chapters do suggest “path-dependent” effects, with change largely replicating existing structures, others demonstrate more substantial shifts. The changes, however, do not uniformly move toward a “social investment” model or to “market liberalization.” While these ideas have been powerful focal points, political coalitions and institutions have proved to be important mediating forces and constraints to change. The chapters in this volume set out this story in three parts. We begin with a set of analyses of one particularly important policy idea, the “social investment” model, with four chapters that set out the extent to which the model has been widely adopted, employing both statistical analysis and focused cross-country historical comparison. The second section of the volume examines the important role of political coalitions in either facilitating or blocking change toward social policy ideas. The final section of the volume examines how preexisting policy institutions and historical legacies constrained the kinds of reforms that could emerge. Robert Cox starts the first section of the book, on the “social investment” model, contrasting social policy reform in the Netherlands, Denmark, and Belgium. Cox sets out the evolution of the idea of a social-investment state, showing how in places that it took hold as an alternative model for social provision, policy makers were able to make substantial changes. In Belgium, where it failed to take hold, by contrast, policy makers continued to engage in a more path-dependent and incremental reform process. Robin Stryker, Scott Eliason, Eric Tranby, and William Hamilton examine the implications of the emerging social investment model. They argue that while increasing women’s education has, to an extent, promoted employment, the effects have not been as far-reaching as other measures. Although, rhetorically many small countries have discussed introducing more flexible labor markets, lower transfer payments, and more investment in education, the reform efforts and outcomes are more complex. Juho Saari’s chapter sheds light on the mechanisms behind the adoption of an idea. He shows that policy makers enacted dramatic changes to the Finnish welfare state that reduced entitlements to pensions and un-

10 | Gary B. Cohen, Ben W. Ansell, Robert Henry Cox, and Jane Gingrich

employment benefits while expanding entitlements for those facing new risks, such as single mothers, children, and the long-term unemployed. Change along these lines was possible precisely because reformers across the political spectrum were attracted to the concept of a “social investment state” and rapidly coordinated around it during a period of crisis, with the promise to redirect social benefits in ways that addressed new risks while also making the labor force more competitive and gaining support across the political spectrum. Jorma Sipilä’s work further shows the social-investment model has been more limited than many analyses suppose. Sipilä argues that there has been no uniform movement across either large or small countries toward investment in education, child care, or active labor-market policies over traditional spending on pensions, unemployment insurance, or other passive benefits. Social investment has been a crucial idea for reform in some, but certainly not all, contexts. The second section of the book goes “under the hood” of the process of reform, examining both the ideas of “social investment” and “market liberalization,” and showing that the emergence of new ideas does not end the importance of distributive coalitions. Indeed, this section shows that distributive battles emerge both to prevent and shape the adoption of ideas. Although ideas frequently provided a common frame for action, both social investment reforms and market liberalization depend crucially on what Sara Watson calls the conditions for “political exchange.” Jane Gingrich begins this analysis through the case of market reforms of health and elderly care in Sweden. The introduction of market forces offered a common frame to policy makers who wished to address shortcomings in the public sector status quo, but political actors on the Left and the Right maintained and pursued different forms of market reforms. The Left used markets strategically in order to improve the legitimacy of the welfare state, while the Right used them in precisely the opposite way: to support private actors. A single idea provided fodder for varying distributive aims. By contrast, Reinhard Heinisch argues that in Austria the lack of a common frame of action inhibited change. Despite growing economic stress during the last two decades, and Austria’s long famed system of political accommodation and coordination, actors could not agree on either socialinvestment or market-liberalization strategies. Here partisan political actors, perceiving the distributive implications of change, moved to block it despite strong pressures to engage in reform. Instead of rallying around a common frame providing an alternative logic to justify social protection, domestic actors resisted reform.

Introduction: Social Policy in the Smaller EU States | 11

Sara Watson’s chapter bridges both findings, comparing the case of Portugal, where market liberalization took hold, to that of Spain, where it did not. Watson shows that reformers in Spain and Portugal pursued similar legislative strategies aimed at liberalizing the labor market by decentralizing collective bargaining to the firm level, but that the impact of these reforms varied dramatically across countries based on the way existing actors used them. In Spain, employers encountered a rigid labor market with numerous pathologies, and unions built on these problems to exact concessions from employers, bartering away changes in some areas for protections in others. By contrast, in Portugal, the more decentralized starting point limited the scope for unions to exact further concessions from employers, allowing employers more space to follow their preferences and coordinate around the idea of markets. The third section follows up on these differences, looking at how institutional and historical legacies also condition the adoption of both social investment and market liberalization. The chapter by Kieke Okma and her colleagues reviews the experience of seven small countries that introduced both expanded benefits and modified forms of markets into their health-care systems. While in each case expanding coverage and expanding marketlike reforms within the publicly financed system offered a common reform frame, these countries introduced markets in dramatically different ways that preserved or extended existing diversity. Ben Ansell’s chapter develops these claims in the area of social investment, showing how existing institutions can modify or block the adoption of particular ideas. While the expansion of higher education aimed at producing skilled generalists has occurred in some European states, there has been little change in higher-education systems in many continental European countries, which still cater largely to elite students. In Germany, Austria, and Switzerland a variety of subnational veto players have been able to restrict expansion of higher education, often against the wishes of those in national government wishing to promote a social-investment model. Consequently, higher education expansion has typically only occurred under “grand coalitions.” In political systems where actors can easily block radical reforms, enacting change of the welfare state requires widespread agreement among actors on an alternative model of action to the status quo. A similar finding emerges from Paulette Kurzer’s study of the regulation of public health. Kurzer shows dramatically different responses in Austria to smoking bans and genetically modified food, with the Austrians taking early and decisive action on limiting the introduction of GMOs (Genetically Modified Organisms) into the food chain while being one of the last countries in Europe to address smoking. In holding the larger Aus-

12 | Gary B. Cohen, Ben W. Ansell, Robert Henry Cox, and Jane Gingrich

trian political context constant, Kurzer shows that shared beliefs about the dangers to the food supply provided stimulus to rapid action with respect to GMOs, but, in the case of smoking, the historical legacy of Nazi control of public health limited reform. Commonly held ideas mattered for reform, but existing institutions and historical legacies shaped how they emerged.

Conclusion The experience of the small countries, then, provides a number of clues about the future of the welfare state in Europe and abroad. Pressures from outside, the weight of past institutions, and the need for compensation do not obviate politics. Indeed, the authors show that both the extent and character of change—and nonchange—was politically driven in crucial ways. Change emerged more extensively where policy makers accepted an alternative set of ideas about the logic and purpose of the welfare state. However, ongoing bargaining among partisan and economic actors within this common framework typically shaped the specific character and implications of reform. These findings about the political dynamics of efforts to change social policy in the smaller EU member states suggest important lessons for policy makers in larger states, which may also face pressure to engage in far-reaching social policy reform.

Bibliography Béland, Daniel. 2009. “Ideas, Institutions, and Policy Change.” Journal of European Public Policy 16, no. 5: 701–18. Boix, Carles. 1998. Political Parties, Growth, and Equality: Conservative and Social Democratic Economic Strategies in the World Economy. Cambridge: Cambridge University Press. Culpepper, Pepper. 2008. “The Politics of Common Knowledge: Ideas and Institutional Change in Wage Bargaining.” International Organization 62, no. 1: 1–33. Estevez-Abe, Margarita. 2006. “Gendering the Varieties of Capitalism—A Study of Occupational Segregation by Sex in Advanced Industrial Societies.” World Politics 59, no. 1: 142–75. Garret, Geoffrey. 1998. Partisan Politics in the Global Economy. Cambridge: Cambridge University Press. Hicks, Alex and Duane Swank. 1992. “Politics, Institutions, and Welfare Spending in Industrialized Democracies, 1960–82. American Political Science Review 86, no. 3: 658–74.

Introduction: Social Policy in the Smaller EU States | 13

Iversen, Torben. Capitalism, Democracy and Welfare. 2005. Cambridge: Cambridge University Press. Katzenstein, Peter J. 1985. Small States in World Markets: Industrial Policy in Europe. Ithaca: Cornell University Press. Lijphart, Arend. 1968. Verzuiling, Pacificatie en Kentering in de Nederlandse Politiek. Amsterdam: J. H. De Bussy. Pierson, Paul. 2000. “Increasing Returns, Path Dependence, and the Study of Politics.” American Political Science Review 94, no. 2: 251–68. Powell, G. B., Jr. 1982. Contemporary Democracies: Participation, Stability and Violence. Cambridge: Harvard University Press. Schwartz, Herman. 1994. “Small States in Big Trouble: State Reorganization in Australia, Denmark, New Zealand and Sweden in the 1980s.” World Politics 46, no. 4 (July): 527–55. Streeck, Wolfgang. 1991. “On the Institutional Conditions of Diversified Quality Production.” In Beyond Keynesiansim: The Socio-Economics of Production and Employment, edited by Matzner, Egon, and Wolfgang, 21–61. London: Edward Elgar. Weyland, Kurt. 2008. Bounded Rationality and Policy Diffusion: Social Sector Reform in Latin America. Princeton: Princeton University Press.

Section I

THE SOCIAL INVESTMENT AGENDA From Ideas to Policy?

(

Chapter 1

HOW GLOBALIZATION AND THE EUROPEAN UNION ARE CHANGING EUROPEAN WELFARE STATES

( Robert Henry Cox

Globalization and Europeanization are often described as forces that compel states to adopt neoliberal reforms to their welfare states. At the same time, these forces intensify the conflict over social policy by encouraging contending political parties to dig into their positions. Change in social policy, therefore, would seem to demand a fundamental reordering of the balance of political forces, with its attendant disruption and conflict. Although there has been a great deal of conflict around the politics of welfare reform in Europe, the countries with the most acerbic welfare debates have reached stalemates in the reform process, whereas the countries that have gone the farthest to change their welfare states have done so with a comparatively greater degree of social peace. In the laggard countries, the politics of social policy continue to be characterized as a win-or-lose struggle between Left and Right, or labor and capital. For the Left, and especially for labor, globalization and Europeanization threaten to deprive labor of the hard-fought gains that welfare states represent. The Right, by contrast, sees adaptation to globalization as an imperative for dramatic change with grave consequences for failure. But in those countries where reform has gone the farthest and been most effective, this struggle did not lead to the type of entrenched struggles that one might expect. How is it that reform and consensus could have accompanied one another, indeed reinforced one another, to produce a “virtuous cycle” (Levy 1999) of reform?

18 | Robert Henry Cox

The answer, suggested in this chapter, is that reform happened where political actors engaged in a root-and-branch reconsideration of the purpose of the welfare state and forged a consensus around a new set of principles that made reform possible. Moreover, this new consensus embraced the challenges Europeanization and globalization posed and actually led to policy changes that were designed to work with these forces rather than oppose them. To put it in stark terms, where the postwar development of welfare states could be described as a struggle of “Politics Against Markets” (Esping-Andersen 1990), the recent wave of reform sees social policy as a crucial part of the package of economic adjustment rather than as a bulwark of security against the market. This chapter examines three small European countries, two of which stand out as leaders in welfare reform (Denmark and the Netherlands), while the third has become the archetype of welfare stalemate (Belgium). As these three cases demonstrate, the welfare reformers have engaged in a root-and-branch reconsideration of the purpose of their welfare state. The welfare laggard, by contrast, is not adapting because political positions are still mired in the old politics of the welfare state, largely because Belgium has been caught in a linguistic conflict that has overshadowed all political issues.

The Newest Politics of the Welfare State Globalization exposes national economies to increased competitive pressures, and this compels them to trim levels of social spending, reduce nonwage labor costs, and create a more flexible labor force. Europeanization is the integration and harmonization of practices among the countries that are members of the European Union (EU). Europeanization often produces effects similar to globalization due to what Fritz Scharpf (2002) calls the “constitutional asymmetry” between the single market and social policy. Because the European Union was designed to promote economic integration—and its treaty agreements underscore this—its instruments of compulsion are directly related to efforts to create and expand the European single market. When social policy interferes with the operation of the single market, the EU has strong instruments to compel states to adjust social policy in favor of it. To date, EU treaties have not articulated a commitment to social policy on par with the status of the single market (Daly 2006). Consequently, efforts to build and enhance the social dimension of Europe can only operate via soft mechanisms of coordination. The bias for the single market, therefore, has similar effects to that of globalization, stating a preference for the market over social policy.

How Globalization and the European Union Are Changing European Welfare States | 19

Furthermore, under the EU system of multilevel governance, social policy is the competence of nation-states. Yet at the national level, pathdependent forces, specifically the impact of “lock-in effects” and “increasing returns” (Pierson 2000) built into both the politics and the programs of the welfare state, make it difficult to introduce change. At the political level, welfare states are part of the postwar settlement between labor and capital, which created lock-in effects—strong incentives that deter either side from vigorously renegotiating the terms of the settlement. At the program level, specific aspects of program design, such as PAYGO methods of funding public pensions, create increasing returns, making it difficult to consider a different formulation of the program that might be more fiscally durable. These path-dependent elements of welfare states tend to reinforce the political status quo and leave the political space essentially unchanged since the beginning of welfare expansion. On the Left are advocates of welfare expansion who act on behalf of those adversely affected by unfettered market forces. On the Right are liberals who favor market solutions and strive to limit the encroachment of social policy on markets, as well as conservatives who believe that social policy has eroded social obligations. In this sclerotic environment, most efforts to reform social programs have been minor and incremental. As characterized by Paul Pierson (1996), the popularity of welfare states has led political leaders to adopt strategies of “blame avoidance” by not reforming major programs, instead focusing their reform energies on programs that have few beneficiaries or by devising reforms that are technical and not easily understood by the general public. In addition, there is a credibility gap between Left and Right in the reform process. The public is suspicious of calls for reform when they come from the political right, whereas, when similar suggestions come from the Left, they tend to be taken more seriously (Ross 2000). Essentially, in the “new politics” of the welfare state, the traditional lines of political cleavage have not changed. The political equilibrium of the postwar settlement is intact, pitting Left and Right in the roles they have embraced for half a century. Also, the basic understanding of the scope and purpose of the welfare state remains unchallenged and unchanged. Welfare reform is incremental, piecemeal, and modest. In some European countries, however, the “newest” politics of the welfare state take a step beyond incremental adjustment. In these countries the original purpose and scope of welfare programs is under fundamental reconsideration, and this has reconfigured the political coalitions. Whether the welfare state continues is no longer the issue that divides. Instead, the questions center on where the welfare state fits in relation to other economic and social concerns. In these innovative countries, parties on the

20 | Robert Henry Cox

Left recognize the need to maintain a globally competitive workforce, not simply protect jobs in sheltered industries. These parties know that creating new jobs for a new economy is the way to boost employment and that protecting jobs in declining industries is only postponing the pain. Parties on the Right recognize that social policy plays a crucial part in keeping a workforce technologically advanced and adaptable to changing circumstances (Schmidt 2002). Both of these concerns come together in a few important new principles that drive welfare reform. First, the newest vision of social policy places more emphasis on programs that boost employment rather than those that protect income. The result of this has been to develop and expand active labor market programs and to expand the responsibilities of individuals to save for retirement and insure themselves against employment disruptions. The point here is not that active labor market programs have replaced workers’ rights with a more stingy system but that it is crucial to keep people in the workforce, contributing to their own and the collective well-being. Second, in a rapidly changing global economy, a competitive workforce needs to have the skills to embrace new technology. Continental European welfare states were based on an old industrial model whereby people trained for a job they would hold the rest of their lives. Today in Europe, people entering the workforce are likely to hold many jobs over their lifetimes, and policies designed to encourage “lifelong learning” are intended to help them adjust to employment transitions (Griffin 1999). In addition, flexibility, not only in the labor market but also in regulations that govern how one accrues credits for pensions and other social insurance benefits, is needed to take into account the more varied work histories of a flexible labor force. Third, notions of entitlement have changed. The greatest achievement of the postwar expansion of welfare states was the institutionalization of the idea of social citizenship (Marshall 1949; Mishra 1990). Social welfare came to be seen as a right, and this produced another lock-in effect whereby rights could only expand, not contract. The newest politics of the welfare state change the foundation on which the welfare state is legitimated by emphasizing the reciprocity between rights and duties. This is an important shift that addresses three normative criticisms about European welfare states: that welfare programs do not provide enough incentives for citizens to contribute to their societies; that within the single market, welfare programs are a magnet to draw migrants from other EU countries; and that welfare programs are an obstacle to the integration of immigrants from outside Europe.

How Globalization and the European Union Are Changing European Welfare States | 21

In this unseated political equilibrium, the institutions of the welfare state cease to be insurmountable obstacles to reform. The institutions of corporatism—tripartite councils, labor unions, and employer federations—once locked their members into the contours of the postwar settlement. However, employers and workers are finding their peak associations ill-equipped to recognize the new concerns of their membership. Decentralization in collective bargaining has fractured the institutions of corporatism, and the erosion of their memberships has prompted peak associations to redefine their purposes. At the program level, processes of institutional layering (Thelen 2004) have allowed institutions that are difficult to reform to be augmented with institutions that allow social policy to reflect the new purpose. For example, public pension schemes, notoriously difficult to reform because of their large sunken costs and a double-payment problem, have been supplemented with a multitude of new programs to encourage private savings for retirement. For the countries that are making these adjustments, the newest politics of the welfare state have lessened the tensions that arose from globalization and Europeanization. A strong emphasis on education and lifelong learning has facilitated the transition to a knowledge-based, service economy. The constitutional asymmetry between the European single market and national welfare states becomes less of a burden for those countries that reoriented social policy to support market development.

Beyond Ideology In countries where one finds the newest politics of the welfare state, parties are responding to globalization and Europeanization by fundamentally reconsidering their ideological foundations. This confrontation has been most severe for Social Democratic parties, which saw the postwar settlement as a Social Democratic victory, representing the triumph of the working class for social justice. In the words of Gøsta Esping-Andersen (1990), social welfare and labor market regulations led to the “decommodification” of labor, freeing the working class from the choice between work and lifestyle. However, the Social Democratic parties that have reexamined their ideological foundations have rediscovered a positive vision of work at the heart of their ideology. To put it in simple terms, Social Democrats take pride in having put into practice Karl Marx’s dictum “to each according to his need,” but they often overlooked the first part of the dictum, “from each according to his ability.” Today, the most innovative Social Democratic parties are augmenting their well-developed theories of distribu-

22 | Robert Henry Cox

tion with equally well-considered theories of production. The new idea in Social Democratic circles is to enact “social investment” policies that place a premium on training, upgrading skills, and reintegrating labor market “outsiders” whom existing regulations of the labor market have discriminated against (Bernard and Boucher 2007; Giddens 1998). The ironic paradox in this shift is that Social Democratic reformers employ instruments that are indistinguishable from those their opponents on the right side of the political spectrum advocate. Upgrading skills requires not only that people enroll in retraining programs, but also that these programs lead to jobs within a specified period of time. Such notions have been the hallmark of neoliberal labor market reforms. When US President Bill Clinton advocated a five-year limit for benefits, European Social Democrats derided this as a punitive reform only imaginable in America. Now, such time limits are increasingly common in Europe and receive the endorsement of Social Democratic parties. Also, Social Democrats are advocating tighter conditions for the award of benefits (Taylor-Gooby 2004). New ideas have allowed Social Democrats to reconcile labor market reforms with their ideologies. Social Democrats make a special effort to distinguish their form of labor market activation from what their neoliberal opponents proposed. The basic distinction, as articulated by many Social Democratic parties, is that they are trying to use the power of the state to allow people to return to and thrive within the labor market. Mobilizing the capacity of the state, rather than allowing market forces to make adjustments, is a hallmark of Social Democracy, and social investment policies are the ideas that link the ideology to the realities of globalization. The irony is that the mechanisms Social Democrats advocate are often those that also have the support of parties to the right of the political spectrum. Though the ideological challenge has been strongest for Social Democratic parties, the new discourse has also affected parties on the right. The challenge for the Right, at least among the neoliberal parties, is to accept that unregulated markets, especially labor markets, do not operate ideally. Rather, markets need some sort of regulation to prevent them from realizing suboptimal outcomes, an argument that has been easier to make following the global financial crisis of 2008. For example, the deregulation of labor markets is often seen to remove a major obstacle to female participation in the labor force; however, the actual participation of women in the workforce is dependent upon the availability of child-care options for those who choose to work. Scandinavian countries enjoy the highest participation of women in the workforce, largely because the state underwrites a myriad of child-care options. Indeed, publicly subsidized child care itself is a major source of female employment. Countries with more liberal labor market regulation, on the other hand, find that women are

How Globalization and the European Union Are Changing European Welfare States | 23

best able to enter the workforce if they can do so at a high income or if they have a spouse who also works. Lone mothers, especially those who have marginal labor skills, often find that the child-care options available in the marketplace make it difficult, if not impossible, for them to balance work and family duties (Lewis 2006). Consequently, liberal parties are learning to accept some ideas from the Left that require more state intervention in the labor market, such as subsidized child care or retraining programs. Parties on the right, which have a more dirigiste (statist) tradition, also are facing challenges to adapt. Christian Democratic parties have long supported heavily regulated labor markets. Indeed, they were the architects of the present forms of labor market regulation, taking credit for such ideas as the “breadwinner” concept (Kersbergen 1995) or the “Sozialmarktwirtschaft.” The breadwinner concept was the idea that a wage earner was responsible not only for himself, but also for his entire family. Its impact in policy led to high wages to ensure that the breadwinner could feed the entire family, job tenure to ensure that the income security of the family was not dependent on the whims of market forces, and generous social insurance to allow the family to maintain its standard of living in the event a calamity affected the breadwinner’s ability to work. The Sozialmarktwirtschaft, or social market economy, was the German Christian Democrats’ alternative to the free-market economy, emphasizing a regulated market that provided social security. For Christian Democrats the current challenge is to move away from the idea of corporatist economic management. The cozy relationship between the state and the social partners (business and labor) has proven to be the single largest obstacle to labor market reform in countries where Christian Democrats or dirigiste parties were responsible for creating the present labor market regimes. Corporatism allows labor unions to protect existing jobs, with the adverse result that employers are reluctant to take on new workers. Widely known as the “insider-outsider problem” (Lindbeck and Snower 2001), this cozy relationship is responsible for high levels of youth unemployment and a degradation of skills in the workforce. The entry of women into the workforce, too, challenges these parties to consider that women could have multiple roles in society and not just be mothers and homemakers.

The Newest Politics of Labor Markets The countries that have booked the most dramatic changes in labor market policies have done so because new ideas about what the labor market should achieve have been widely discussed; these new ideas have been

24 | Robert Henry Cox

embraced, and they have created a legitimate basis for the new reforms. For many years, Denmark and the Netherlands were considered to have implemented the best examples of successful labor market reform, and here is where the most innovative policy activity has taken place. In the Netherlands, the reform of welfare programs began in the 1980s and reflected a slow adjustment to a problem created by the most generous breadwinner model in Europe. The problem stemmed from a very generous system of benefits that dated from the middle of the 1970s. By the late 1970s, an economic downturn set in, and this generous system proved to be a form of hidden unemployment. To avoid layoffs, many workers were encouraged to visit their doctors and apply for disability assistance. The result was that by the middle of the 1980s, the Netherlands had the highest percentage of disabled workers in all of Europe. This high level of inactivity was especially pronounced among older and low-skilled workers (Veen and Trommel 1999). At the same time, more young women began to pursue careers and were frustrated to encounter formal regulations and informal discrimination that discouraged them from taking jobs. When the European Commission began a campaign to reduce gender discrimination in the workplace, it singled out the Netherlands for its extremely low level of female participation in the workforce (Maruani 1992). Organizations representing women pushed for reforms that would allow all workers, including women with part-time jobs, to enroll in the social insurance system. Their criticisms focused on the regulatory obstacles that discouraged women from taking jobs if they also had young children to care for. This was the origin of the idea of the “one-and-ahalf-job model,” a hallmark of the new approach to the labor market in the Netherlands, which suggested that a household could better balance work and home responsibilities if it had two part-time workers (Visser and Hemerijck 1997). Many men were changing their own ideas of how to balance work with family and found the prospects of part-time work attractive, and a wider sector of the population embraced the idea. The most dramatic change, however, came in the form of active labormarket policies. The basic idea behind this, known as activation (aktivering) both in Denmark and in the Netherlands, was that the entire system of cash transfers, which included social insurance as well as public assistance, served as a disincentive to work. The critique was controversial, but the enacted reforms sought to reduce the take-up rates in benefit programs by reducing the payments, reducing the duration of benefits, and otherwise providing incentives for people to move back into paid employment. These changes, it was argued, represented a shift “from the safety net to the trampoline,” a phrase coined by the Rotterdam Social Service Office (Gemeentelijke Sociale Dienst Rotterdam 1985). Like in Denmark,

How Globalization and the European Union Are Changing European Welfare States | 25

the major mechanism for this was a budget-based decentralization of authority to municipal governments and works councils to better supervise the implementation of their programs. To justify strict enforcement of active labor-market policies, officials in both Denmark and the Netherlands articulated a notion of reciprocity between rights and responsibilities. The basic idea was that a worker who had a right to unemployment and other benefits also had a responsibility to actively seek work—and to take any available job. In the Netherlands, activation policies included a legislative change that required people receiving unemployment benefits to seek jobs of comparable worth. This was a large departure from the old system, which allowed a person to stay on benefits if he could not return to the job for which he had been trained. In Denmark, controversy over the strictness of the activation measures led the government to temper the activation responsibilities with more worker rights. These came in the form of a job-rotation program that allowed workers to take leave for family care, retraining, or sabbatical. Except for the family-care provisions, employers disliked the program, and it was sharply curtailed shortly after its adoption (Martin 2005). In short, the pattern in both Denmark and the Netherlands was one of creative problem solving, thinking up new ideas to reconceptualize the balance between work and welfare and the balance between job and home. The creative ideas helped to reorient the positions of political actors. Social Democrats, as well as the political right, supported many forms of activation, as well as the cutbacks in cash transfers. The language of rights also resonated across the political spectrum. Ideas such as “flexicurity” and the “one-and-a-half-jobs model” encouraged people to think of labor market provisions in a new way, one that reflects the contemporary reality of working women and a workforce that requires periodic upgrading of skills. As such, these ideas mark a departure from the postwar settlement on labor market policy and build a foundation for a more globally competitive labor market. Moreover, old institutions of corporatism are not insurmountable obstacles to realizing change. The change in ideas has eroded the legitimacy of traditional institutions of corporatism, leading peak associations to scurry to find new ways of embracing their membership.

Retro Belgium In contrast to the Netherlands and Denmark, Belgium represents one of the weakest cases of welfare adjustment in Europe. Numerous reports from international agencies have criticized Belgium for its dire economic prospects and the failure of its political leaders to tackle seriously the threats

26 | Robert Henry Cox

its antiquated system of social welfare posed (OECD 2006, 2007; Commission of the European Communities 2006; IMF 2005). The basic thread running through all the critiques is that the lack of reform in Belgium is compounding problems that will be more difficult to fix in the future. For this reason, international experts recommend that action be taken to alleviate the forecasted disaster. Two policy areas have been singled out for the strongest critique: pensions and unemployment assistance. One problem affects both of these policy areas. Belgium suffers a very high level of inactivity in the labor market, meaning that a large percentage of people of working age are not working. For example, according to statistics from the year 2000, Belgium had the highest unemployment-benefit dependency ratio in Europe, at 23.5 percent (Kuipers 2004: 58). This means that one-quarter of the Belgian workforce was drawing unemployment benefits, a percentage that overwhelmed the capacity of the working population to pay enough unemployment taxes to support those receiving benefits. This low level of labor market participation is a legacy of the programs for early retirement in Belgium. During the 1980s, many European countries expanded early-retirement options to reduce unemployment levels. Belgium went further than most countries in utilizing this measure. Since 2000, Belgian governments have sought to increase the effective age of retirement, but labor unions have successfully halted all initiatives. In 2005, an effort to raise the age of early retirement from fifty-eight to sixty generated the first general strike in twelve years. The low level of inactivity also threatens the future solvency of the pension system, because a projected increase in the ratio of pensioners to people in the workforce is compounded by the low level of labor market activity. The prescription from all experts is that Belgium needs to institute reforms to get more people working and keep them in the workforce longer. Yet during the past two decades, Belgium has managed to institute only one serious reform of its social safety net, and this reform did not match the scope of reforms instituted in other countries, for example, Denmark and the Netherlands. The single major reform was the “Global Plan” adopted in 1993 by the government led by Jean-Luc Dehaene. A simple comparison with the Netherlands gives a sense of the difference in scope. During the 1990s, the Dutch government enacted three major reforms of the welfare system, each of which generated savings greater than the entire Global Plan. In addition, the Dutch reforms froze benefits in 1994 and 1995 and cut the levels of benefits in future years, whereas the Belgian Global Plan introduced no changes to the levels of benefits (ibid.: 67). Finally, the Dutch reforms severely limited the autonomy of the social partners (labor and management) to administer social insurance, while in Belgium social

How Globalization and the European Union Are Changing European Welfare States | 27

partners continued to enjoy their formal role, and the political power of the labor unions actually grew, allowing them to forestall a number of measures to make labor markets more flexible (ibid.: 103). Thus, the Belgian record was of incremental, rather than fundamental, reform, and it did not satisfy international observers. Since 2000, efforts to reform Belgian labor markets have been modest, incremental at best, and have occasionally exacerbated the problems of a passive welfare state. Belgian governments have increased spending on active labor-market policies, but the levels of social security benefits have weakened the incentive to respond to activation measures. Modest changes have been introduced into the social security system that have moved it from an insurance system to more of a basic income system. In the words of Erik Jones (2008: 169), the approach to reforming social policy in Belgium has been to “try everything else first.” There has been no grand narrative of reform in Belgium. Globalization has not been presented as a crisis, demanding response. Unlike in Denmark and the Netherlands, and more like in Germany, the pressures of globalization and Europeanization have engendered polarization rather than consensus between left and right (Kuipers 2004). Indeed, the polarization has eroded the institutional framework that once created consensus in Belgium. For much of the postwar period, Belgium was one of the best cases of a consociational democracy, a system where political and social differences were surmounted through institutions that promoted elite consultation. Today, political polarization has eroded the role of social partners in the consultation process, and the political system of Belgium is easier to describe as pluralist than consociational (Jones 2008). In this pluralist environment, political and social differences are the focus of interest organization, and the mechanisms of elite accommodation fail to produce agreements (Marier 2005). Nor has Europeanization had a significant effect on Belgian social policy. The EU focus on reforming social security and labor markets has actually worked in favor of the status quo (Beyers and Bursens 2006). Unlike in the realms of environmental and competition policy, where the EU has forced dramatic change on Belgian governments, the open method of coordination in social policy has allowed Belgium to voice commitment to broad principles of social reform while undertaking little substantive reform. Most critics blame the federal structure for creating too many veto points that allow entrenched interests to easily block reform (e.g., Swenden, Brans, and de Winter 2006). Fragmentation affects political leadership, the ability of the social partners to speak on behalf of a unified membership, and the positions taken by political parties (Béland and LeCours 2005). At

28 | Robert Henry Cox

the political level, Belgian federalism demands coalition, in which minority governments often prevail. Furthermore, federalism allows the Waloon Socialist Party to be the dominant political force, even though it receives less than 20 percent of the popular vote. In countries where there is clear political leadership, either the office of the prime minister or the minister of finance takes on the role of policy leader in welfare reform. In Belgium, the reform-phobic profile of governing coalitions has weakened the ability of these offices to lead reform. At the program level, the division of the Belgium welfare state between the federal and national level complicates efforts to implement reforms. Social insurance is supervised at the national level, but disparities between the economic performance of the Flemish and Walloon regions lead to complaints that the region that has made the necessary economic adjustments (Flanders) has to subsidize the one that has not. In this polarized debate, Belgium even lags behind other European countries in making the modest adjustments that were the hallmarks of social insurance reforms in the 1990s, such as reducing early retirement and adjusting pensions to anticipate a larger proportion of retired citizens. Active labor-market policies, by contrast, are local responsibilities that are jealously guarded by local officials. Whereas in the Netherlands and Denmark local officials work together to place people in jobs in each others’ regions, Belgian officials resist implementing work requirements when doing so would force a person to take a job in another region (de Vos 2008). Neither have the institutions of corporatist consultation helped Belgium. In Denmark and the Netherlands, corporatist councils allow the social partners to provide serious and scientifically responsible advice, relatively insulated from public pressure. In Belgium, in contrast, the major advisory body is comprised of the social partners who take stridently opposed positions on issues of pension and labor market reform (Marier 2005). For example, an effort to raise the retirement age faced a protracted struggle and led to slight changes in rules, not fundamental reform (ibid.: 4). More fundamental efforts to recalculate the pension credits for women were defeated by the social partners. Not to be underestimated in Belgian politics is the degree of political conflict that results from the linguistic division. Political conflict has caused Belgian politicians to sideline planning, including welfare reform, and respond to crises as they develop (Walgrave and Varone 2008). Following elections in June of 2007, for example, a dispute over constitutional reform led to a six-month interregnum during which none of the political parties could strike a coalition agreement. In 2008, the governing crisis persisted and was compounded by the global financial crisis that occupied attention at the end of 2008 and into 2010. During 2010, a dispute

How Globalization and the European Union Are Changing European Welfare States | 29

over voting rights for Flemish and French-speaking Belgians dominated the political agenda, exacerbating the tensions among the parties and narrowing the space for political cooperation. The result was that most attempts at reform in Belgium resembled the “new” rather than the “newest” politics of the welfare state. The Waloon Socialist Party is an unreconstructed Social Democratic party. Its ideological and policy commitments have not altered significantly in the past forty years. Neoliberal forces prevail in the Flemish half of the country, and the politics of subnational identity (Béland and LeCours 2005) divide the country into two halves that have stereotypical images of each other. In the more neoliberal and conservative Flanders, a self-image as entrepreneurial and embracing of globalization leads to an image of the Waloons as lacking initiative and dependent on social benefits. On the Waloon side, a commitment to the old politics of the welfare state leads to suspicions of efforts to enact any sort of change.

Conclusion European countries are redefining their welfare settlements. The drivers for these renegotiations are Europeanization and globalization. For the smaller democracies of Europe, these forces have extraordinary effect. As noted twenty-five years ago by Peter Katzenstein, the smaller democracies of Europe have generous welfare states because they have open economies; however, those welfare states were a cushion that protected their citizens from the vagaries of the international economy. The welfare states allowed those countries to remain internationally competitive because they removed unproductive workers from the labor force. Today, globalization makes it more costly to maintain generous welfare provisions, and Europeanization only intensifies the pressure. In countries where welfare reform is more creative and effective, political actors have rethought their orientations to social policy. The new welfare consensus strives to define welfare programs that support and reinforce the market. As this research indicates, Denmark and the Netherlands are two countries that exhibit this progressive development. In those countries, the challenges caused by globalization led to careful considerations of the consequences of nonaction, and public officials led discussions that helped people feel comfortable with some rather dramatic changes to the cozy status quo of their welfare systems. The ideas at the center of these discussions depicted social welfare not as a reaction to market failures, but as a public investment to support the normal operation of the market. For example, “flexicurity,” the idea that social security can

30 | Robert Henry Cox

be combined with a flexible labor market allowed for some long-needed reforms that recognized part-time work and still awarded pension entitlements. Moreover, even the political left embraced these new ideas, thereby fostering a new consensus. By contrast, Belgium is an example of a country where new ideas failed to permeate the political discourse. Instead, old patterns, developed and solidified during the postwar period, served to resist change. Political parties remained loyal to their ideological foundations and avoided consensus or pragmatic solutions. There were some idiosyncratic elements to the policy sclerosis in Belgium. Tensions between the country’s two linguistic groups created a situation of instability in government and overshadowed serious discussion of policy reform. Neither path is inevitable. Belgium could easily embark on reform, and Denmark and the Netherlands could have easily avoided serious reform. The important lesson to be drawn about public policy is that evidence of a need to change is insufficient to bring about change. The most important change that needs to take place is in the way people believe their welfare states should work. New ideas about how welfare states fit into modern societies need to pave the way for consensus around a new approach. At the core of this adjustment is a fundamental rethinking of the purpose of the welfare state, one that sees social policy as part of the public investment in a productive workforce. The stories of Denmark, the Netherlands, and Belgium are not unlike the stories of the other small countries of Europe. All have keenly felt the pressures of global economic adjustment. All have tried to respond by cutting back on social programs, and all have attempted innovative reforms that preserve the essential elements of their welfare states while responding to the fiscal pressures. Some have been more successful than others. In many ways, as we noted in the introduction of this volume, the experiences of the smaller democracies are often instructive. Because they are small, they are more sensitive to the global pressures than are the larger European countries. For this reason, they are often leaders in experimenting with reforms and frequently are the proving grounds for many experiments that later are embraced by larger countries. This makes them akin to the proverbial canaries in the coal mine, showing other countries where it is or is not safe to tread.

Notes I would like to thank Gary Cohen, Jane Gingrich, Lisa Peschel, and Mollie Madden for their helpful comments on an earlier version of this chapter.

How Globalization and the European Union Are Changing European Welfare States | 31

Bibliography Béland, Daniel, and André LeCours. 2005. “The Politics of Territorial Solidarity: Nationalism and Social Policy Reform in Canada, the United Kingdom, and Belgium.” Comparative Political Studies 38, no. 6: 676–703. Bernard, Paul, and Guillaume Boucher. 2007. “Institutional Competitiveness, Social Investment, and Welfare Regimes.” Regulation & Governance 1, no. 3 (September): 213–29. Beyers, Jan, and Peter Bursens. 2006. “The European Rescue of the Federal State: How Europeanization Shapes the Belgian State.” West European Politics 29, no. 5: 1057–78. Commission of the European Communities. 2006. A Year of Delivery: The European Commission’s 2006 Annual Progress Report on Growth and Jobs. Brussels: European Commission. Daly, Mary. 2006. “EU Social Policy after Lisbon.” Journal of Common Market Studies 44, no. 3 (September): 461–81. de Vos, Mark. 2008. “Werkloosheidsverzekering is ook Arbeidsmarktbeleid.” Itinera Institute Nota 2008/10. Brussels: Itinera Institute. Esping-Andersen, Gøsta. 1990. The Three Worlds of Welfare Capitalism. Cambridge: Polity Press. Gemeentelijke Sociale Dienst Rotterdam. 1985. GSD, Vangnet of Trampoline? Rotterdam: Gemeentelijke Sociale Dienst. Giddens, Anthony. 1998. The Third Way: The Renewal of Social Democracy. Cambridge: Polity Press. Griffin, Colin. 1999. “Lifelong Learning and Social Democracy.” International Journal of Lifelong Education 18, no. 5 (September): 329–42. International Monetary Fund. 2005. Belgium: Selected Issues. IMF Country Reports no. 05/76. Washington: IMF. Jones, Erik. 2008. Economic Adjustment and Political Transformation in Small States. Oxford: Oxford University Press. Kersbergen, Kees van. 1995. Social Capitalism: A Study of Christian Democracy and the Welfare State. New York: Routledge. Kuipers, Sanneke L. 2004. Cast in Concrete? The Institutional Dynamics of Belgian and Dutch Social Policy Reform. Delft: Eburon Academic Publishers. Levy, Jonah D. 1999. “Vice into Virtue? Progressive Politics and Welfare Reform in Continental Europe.” Politics and Society 27, no. 2 (June): 239–74. Lewis, Jane. 2006. “Work/Family Reconciliation, Equal Opportunities and Social Policies: The Interpretation of Policy Trajectories at the EU Level and the Meaning of Gender Equality.” Journal of European Public Policy 13, no. 3 (April): 420–37. Lindbeck, Assar, and Dennis J. Snower. 2001. “Insiders versus Outsiders.” Journal of Economic Perspectives 15, no. 1 (Winter): 165–88. Marier, Patrik. 2005. “Where Did the Bureaucrats Go? Role and Influence of the Public Bureaucracy in the Swedish and French Pension Reform Debate.” Governance: An International Journal of Policy, Administration and Institutions 18, no. 4: 521–44.

32 | Robert Henry Cox

Marshall, Thomas Humphrey. 1949, 1964. “Citizenship and Social Class.” In Class, Citizenship and Social Development: Essays, edited by T. H. Marshall, 70–134. Garden City, NY: Doubleday. Martin, Cathie Jo. 2005. “Corporatism from the Firm Perspective: Employers and Social Policy in Denmark and Britain.” British Journal of Political Science 35, no. 1 (January): 127–48. Maruani, Margaret. 1992. De Positie van vrouwen op de arbeidsmarkt: tendensen en ontwikkelingen in de twaalf landen van de Europese Gemeenschap 1983 en 1990. Vrouwen Van Europa 36. Brussels: Commissie van de Europese Gemeenschappen. Mishra, Ramesh. 1990. The Welfare State in Capitalist Society: Policies of Retrenchment and Maintenance in Europe, North America and Australia. New York: Harvester Wheatsheaf. OECD. 2006. “Belgium: Momentum Slows.” OECD Observer. Paris: OECD. ―――. 2007. Jobs for Youth: Belgium. Paris: OECD. Pierson, Paul. 1996. “The New Politics of the Welfare State.” World Politics 48, no. 2 (January): 143–79. ―――. 2000. “Increasing Returns, Path Dependence, and the Study of Politics.” American Political Science Review 94, no. 2: 251–68. Ross, Fiona. 2000. “‘Beyond Left and Right:’ The New Partisan Politics of Welfare.” Governance: An International Journal of Policy and Administration 13, no. 2 (April): 155–83. Scharpf, Fritz W. 2002. “The European Social Model.” Journal of Common Market Studies 40, no. 4 (November): 645–70. Schmidt, Vivien A. 2002. “Does Discourse Matter in the Politics of Welfare State Adjustment?” Comparative Political Studies 35, no. 2 (March): 168–93. Swenden, Wilfried, Marleen Brans, and Lieven De Winter. 2006. “The Politics of Belgium: Institutions and Policy under Bipolar and Centrifugal Federalism.” West European Politics 29, no. 5: 863–73. Taylor-Gooby, Peter. 2004. New Risks, New Welfare: The Transformation of the European Welfare State. Oxford: Oxford University Press. Thelen, Kathleen Ann. 2004. How Institutions Evolve: The Political Economy of Skills in Germany, Britain, the United States, and Japan. Cambridge: Cambridge University Press. Veen, Romke van der, and Willem Trommel. 1999. “Managed Liberalization of the Dutch Welfare State: A Review and Analysis of the Reform of the Dutch Social Security System, 1985–1998.” Governance: An International Journal of Policy and Administration 12, no. 3 (July): 289–310. Visser, Jelle, and Anton Hemerijck. 1997. A Dutch Miracle: Job Growth, Welfare Reform and Corporatism in the Netherlands. Amsterdam: Amsterdam University Press. Walgrave, Stefaan, and Frédéric Varone. 2008. “Punctuated Equilibrium and Agenda-Setting: Bringing Parties Back In. Policy Change after the Dutroux Crisis in Belgium.” Governance: An International Journal of Policy and Administration 21, no. 3: 365–95.

Chapter 2

FAMILY POLICIES, EDUCATION, AND FEMALE LABOR MARKET PARTICIPATION IN ADVANCED CAPITALIST DEMOCRACIES

( Robin Stryker, Scott R. Eliason, Eric Tranby, and William Hamilton

Assessing the relationship between education and labor market attainments has long been a staple of research by labor economists and sociologists (e.g. Jencks 1972; Becker 1975; Sewell and Hauser 1975; Stryker 1981; Rosenfeld and Kalleberg 1990; Eliason 1995; Hicks and Kenworthy 2008). Achievements in education are a central indicator of investment in labor market related human capital. So when researchers found that women often received less return in occupational prestige and earnings to their investments in education than did their male counterparts, it became a concern to be addressed through regulatory policies promoting anti-discrimination and gender pay equity (Stryker 1996). Of course, only when men and women enter the paid labor force can they receive any occupational or earnings returns to their investment in education. Women’s entry into paid employment itself has been a major preoccupation of comparative research on the welfare state, social policy, and inequality (e.g., Gornick, Meyers, and Ross 1997; Huber and Stephens 2000; Misra, Budig, and Moeller 2005; Eliason, Stryker, and Tranby 2008). In part this has stemmed from gender-egalitarianism that, from the 1960s, shaped especially Scandinavian and US policies in different ways. Where Scandinavian countries emphasized supports such as paid maternity (later parental) leave and publicly provided child care to support women’s entry into employment, the United States instead took the lead in promoting anti-discrimination and affirmative-action policies to help employed women achieve parity with their male counterparts (Gauthier

34 | Robin Stryker, Scott R. Eliason, Eric Tranby, and William Hamilton

1996; Rosenfeld and Kalleberg 1990; Stryker 1996; Eliason, Stryker, and Tranby 2008). Meanwhile, such countries as Germany, Italy, and Austria became known as exemplars of a male-breadwinner tradition in which policies supported stay-at-home caregiving by mothers and in which female labor force participation correspondingly lagged—albeit in Austria less so than in Germany and Italy, and with Italy an especially extreme laggard (Gornick, Meyers, and Ross 1997; Misra, Budig, and Moeller 2005; Scharpf and Schmidt 2000; Stier, Lewis-Epstein, and Braun 2001). Feminist scholars were the first to put gender and family policies on the agenda for research by scholars of the comparative welfare state (Lewis 1992; Orloff 1993), but the rest of the scholarly pack soon followed (Esping-Andersen et al. 2002; Myles and Quadagno 2002). As Eliason, Stryker, and Tranby noted, “[I]n much of Europe, life-span is increasing as are the costs of supporting retirement, at the same time as fertility rates have plummeted. . . . Under these conditions, even the core ‘male-breadwinner’ countries of continental Europe . . . will find it seductive to contemplate increased female labor force participation” (2008: 135–36). In short, by the beginning of the twenty-first century—albeit for diverse reasons—a broad spectrum of international academics and policy makers agreed that increasing women’s employment in the paid labor market was a desirable goal. This preoccupation with female labor force participation has become a core element of redesigning the welfare state to fit today’s imperatives rather than those of an earlier industrial and post-industrial era.1 In this chapter we extend Eliason, Stryker, and Tranby’s (2008) analysis, estimating the counterfactual causal effect of aggregate levels of female education on female labor force participation. We also estimate the counterfactual causal effects of demand and supply-side policies on female labor force participation, and conditioning on levels of female education, in fourteen advanced capitalist democracies from 1960 to1999. This allows us to see whether and how women’s aggregate educational attainments impact female labor force participation over time and across countries, and also whether and how the effects of public sector size and family policies change when female educational attainment is taken into account. The remainder of this chapter is organized as follows. We first provide brief discussions of the role of demand and supply-side policies and aggregate female education in facilitating or inhibiting female labor force participation. Given the focus of our analysis, we highlight Eliason, Stryker, and Tranby’s (2008) prior analysis throughout this discussion. We then describe the data and variables used in the current analysis, immediately followed by a description of that analysis and the subsequent results. Finally, we close our chapter with discussion and conclusions of the broader

Family Policies, Education, and Female–Labor Market Participation | 35

implications suggested by our analysis regarding the role of aggregate levels of female education in influencing female labor force participation.

Family Policies, Public Sector Expansion, and Female Labor Force Participation To understand better what facilitates, or conversely, inhibits women’s entry into the paid labor force comparative research has focused especially on public sector expansion and on family policies, including maternity leave and public provision of child care.2 Scholars have shown that a public service orientation to welfare state provision, and consequently a large and expanding public sector, is especially associated with the long-term incumbency of social democratic parties in national government (Huber and Stephens 2000; Eliason, Stryker, and Tranby 2008). As Eliason, Stryker, and Tranby (2008) explain, public sector expansion is a demand-side mechanism that increases paid employment for women through a gender-typing and labor market matching process. Because many public sector jobs in the Scandinavian, social democratic style welfare states involve some type of care-giving or otherwise people-oriented or communally oriented tasks, they are perceived to be female-typed jobs and thus most appropriately filled by women. With public sector expansion fostered by long-term social democratic governance, then, comes increased demand for women’s employment. Though the public sector is not immune to job loss during times of fiscal crisis, even governments in otherwise highly market-oriented countries such as the United States may also ramp up public sector job creation to help stimulate the economy during recession. Obviously, public provision of daycare increases the number of femaletyped jobs in the public sector. But like maternity leave policies, public provision of daycare also operates on the supply side to reduce barriers to women’s entry into the paid labor market. As Eliason, Stryker, and Tranby (2008: 143) explain: “Available, affordable daycare is predicted to loosen women’s family-related budget and time constraints, thus decreasing their reservation wages and altering their preference formation. At the same time, widespread use of publicly provided child care facilities may reshape cognitive expectations and normative evaluations about the acceptability or desirability of child care provided outside the home and by someone other than the mother.” The argument pertaining to maternity leave is more nuanced, because the impact of the leave may depend on its length and generosity level (see Eliason, Stryker, and Tranby 2008; Pettit and Hook 2005; Rønsen and

36 | Robin Stryker, Scott R. Eliason, Eric Tranby, and William Hamilton

Sunstrőm 2002). But generous leave policies should encourage pre-childbirth employment for women. As long as the leave does not extend beyond about two years (thereby preventing market-related skills from becoming perceived as rusty) while also continuing to provide very generous replacement income, it should not discourage subsequent labor market participation. Using a combination of fuzzy-set and statistical methods, Eliason, Stryker, and Tranby (2008) examined the relationship among political governance, public sector expansion, a diverse set of family policies, and female labor force participation in fourteen advanced capitalist democracies from 1960 to 1999.3 The countries included were Austria, Belgium, Canada, Denmark, Finland, France, Germany, Ireland, Italy, the Netherlands, Norway, Sweden, the United Kingdom, and the United States. In the fuzzy set analyses, Eliason, Stryker, and Tranby (2008) found that high levels of public sector size, public daycare for children ages zero to two (younger children), and maternity leave were, separately, causally sufficient for high levels of female labor force participation. From a policy point of view, fuzzy set methods are useful especially because they can distinguish between two kinds of causal relations: that of necessity and that of sufficiency. Because no family policy examined was necessary to produce high levels of female labor force participation, and because high levels of public sector size, public daycare for younger children, and maternity leave all, individually, produced high female labor force participation, Eliason, Stryker, and Tranby (2008) concluded that there were alternative available paths to produce high female labor force participation To obtain the magnitude of these effects, Eliason, Stryker, and Tranby (2008) complemented their fuzzy set analyses by estimating compliers average causal effects (CACE) (Imbens and Rubin 1997; Hirano et al. 2000). A CACE analysis operationalizes the counterfactual notion of causation. A compliers average causal effect is analogous to a treatment effect produced under conditions of experimental design, but the former is produced using nonexperimental data where some plausible mechanism can be established encouraging cases into different treatment statuses. Using this design, researchers can identify a counterfactual causal effect for the segment of the population that complies with that mechanism. Estimating bootstrap distributions of CACEs for various welfare state policies, and using high cumulative left governance as the mechanism encouraging governments to enact those policies, Eliason, Stryker, and Tranby (2008) found that the median causal effect for high levels of public sector size was an increase of 38 percent in female labor force participation across the entire data range (for the 14 countries from 1960 to 1999). Though not as large as the demand side impact of public sector size, ma-

Family Policies, Education, and Female–Labor Market Participation | 37

ternity leave had a sizeable supply side effect, with a median CACE of 18 percent. The median CACEs for public daycare for children ages zero to two and children ages three to school age were found to be 11 percent and 9 percent respectively (Eliason, Stryker, and Tranby 2008). When the encouragement mechanism was specified to be low cumulative left governance, high public sector size also had a substantial effect on female labor force participation. Indeed, when countries with non-left political legacies do expand their public sectors, the result is stronger than that produced using strong left governance tradition as the encouragement mechanism, with an estimated median CACE of 46 percent. This, in turn, suggests that countries like the United States and Canada would likely achieve even higher levels of female labor force participation than they already observe were they to expand the public sector further (Eliason, Stryker, and Tranby 2008). Again, when the encouragement mechanism was specified to be low cumulative left governance, public daycare for children ages three to school age (older children) and maternity leave had substantially greater effects than when these same policies were obtained under strong legacies of left governance as the encouragement mechanism. These median effects were a 48 percent increase in female labor force participation associated with public daycare for older children and a 39 percent increase associated with high levels of maternity leave. Public provision of daycare for younger children had a smaller, but still sizeable impact—an estimated median CACE of 14 percent—on female labor force participation. However, Eliason, Stryker, and Tranby (2008) did not consider the impact that increasing educational achievements by women might have on their labor force participation. They make clear why an omitted factor, such as education, does not contaminate assessment of the fuzzy set subset relationships. And while the CACE analysis is not susceptible to omitted variable bias in the same way that standard regression analyses are, it certainly is possible that aggregate levels of female education estimates could condition these estimated compliers average causal effects.

Education and Female Labor Force Participation Like variation in the availability of family policies, variation in women’s educational levels should shape female labor supply. Because there is a positive association between education and earnings, all other things being equal, women who are more highly educated can expect that their economic payoffs to labor market entry will be greater than those attained by less educated women. Similarly, because university and post-gradu-

38 | Robin Stryker, Scott R. Eliason, Eric Tranby, and William Hamilton

ate degrees qualify women for more professional jobs, highly educated women can expect more rewards to labor market entry in terms of occupational prestige and other job attributes, like autonomy, that typically are associated with the intrinsic, as well as extrinsic, work values and rewards that in turn are positively associated with job satisfaction (Kalleberg 1977). As well, with increasing education, and especially with obtaining postgraduate and professional degrees, may come a normative orientation and self-concept promoting women’s participation in the paid labor market. For all these reasons, raising aggregate levels of education attained by women should increase their aggregate labor force participation, at least assuming that jobs commensurate in pay and prestige with women’s investments in education are available. On the other hand, if jobs commensurate with expectations about job quality along these and other valued dimensions of work are not available, aggregate increases in women’s education might not translate into aggregate increases in women’s labor market participation rates. Prior research has shown that women’s educational attainment has increased over time; indeed, women have achieved educational parity with men in many advanced capitalist democracies (see Barro and Lee 2000). From 1960 to 1999, the average years of education for women aged twenty-five and older increased substantially in all fourteen countries that we study and that likewise were studied by Eliason, Stryker, and Tranby (2008). Among these countries, mean educational attainment for women is highest in Norway, Sweden, the United States, and Canada, and lowest in Austria, Germany, and Italy—it is especially low in Italy (Tranby 2008; Table 5). Whereas leaders in women’s education are drawn equally from what Esping-Andersen (1990) referred to as the redistributive, social democratic welfare regime and what he referred to as the liberal, marketoriented regime, all the laggards in women’s education have been classified as conservative-corporatist welfare regimes (by Esping-Andersen and others), and as having male-breadwinner/female-carer gender regimes (by feminist scholars) (see e.g., Esping-Andersen 1990; Hantrais 2004). In their analyses of these same fourteen countries using data at fiveyear, rather than annual intervals, and standard interpretations of fixed effect regressions rather than statistical methods for identifying and estimating causal effects for non-experimental data, Hicks and Kenworthy (2008) found that increased aggregate educational attainment for women was significantly and positively associated with increased aggregate female employment (see also Blau, Ferber, and Winkler 2002). Hicks and Kenworthy (2008) also found that countries with more generous family policies and large public sectors also had greater female labor force participation, even when controlling for women’s average educational

Family Policies, Education, and Female–Labor Market Participation | 39

attainment. However, when they examined within country change over time, Hicks and Kenworthy (2008) found that, once they controlled for education, family policies did not have a statistically significant impact on women’s employment. The CACE framework focuses on identification and estimation of causal effects, whereas typical regression analyses, including fixed effects models, are capable of identifying and estimating causal effects only under a strict set of assumptions, which did not hold for the Hicks and Kenworthy (2008) study. Thus, it becomes all the more important to reexamine the impact of women’s education—along with public sector size and family policies—on women’s labor force participation through the lens of a CACE analysis.4

Data and Variables With the exception of data and variables measuring educational attainment, we use the same data and measures as did Eliason, Stryker, and Tranby (2008). We refer readers there for complete descriptions of data sources and measures. Here, we underscore that, to construct policy measures for the current analysis, we used information on: 1) maternity leaves, including rates of wage replacement, number of weeks of paid leave, the proportion of employed women who were covered, and the public expenditures for paid leaves; 2) public daycare, including, separately, proportions of children ages zero to two years old and three to school age, in public daycare, as well as public daycare expenditures. As described in Eliason, Stryker, and Tranby (2008), we used these raw data compiled on maternity leave and public daycare factors and composite indices to calculate fuzzy set membership scores for provision of maternity leave, public daycare for children ages zero to two, and public daycare for children ages three to school age. For the extended leaves for child care that some countries make available beyond maternity or parental leaves, we calculated fuzzy set membership scores, as did Eliason, Stryker, and Tranby (2008), based on number of weeks of extended leave support. These fuzzy set membership scores then are used in our CACE analyses. Similarly, we use the Eliason, Stryker, and Tranby (2008) fuzzy set membership scores constructed for cumulative left cabinet incumbency and the percentage of total employment accounted for by civilian government employment (public sector size). As the primary outcome, we use raw scores on female labor force participation rates for women ages fifteen to sixty-four across all fourteen countries, for the years 1960 to 1999. For this chapter, we also collected and analyzed data on women’s educational attainment. For each of the countries in our sample, education levels

40 | Robin Stryker, Scott R. Eliason, Eric Tranby, and William Hamilton

for females ages twenty-five and older are obtained for five-year intervals starting in 1960, using the average-years-of-schooling scale developed by Barro and Lee (2000) for cross-national comparisons. With these values, we impute education levels for the intervening years using a simple linear approximation. This gives us education levels for each country from 1960 to 1999, inclusive. This simple imputation may not be reasonable for a statistical analysis using these data on the original scale. However, we do not do so and instead transform these values into fuzzy scores, using the procedure described in detail in Appendix 5.2 of Eliason, Stryker, and Tranby (2008). Thus, our CACE analysis is adversely affected by this imputation only to the extent that the true values for the imputed data vary wildly outside the range on the nonimputed values used as bounds for the imputations. For example, the Barro-Lee estimate of the level of education for females twenty-five years of age and older in the United States is 11.59 in 1990 and 12.09 in 1995. The imputed values for 1991 through 1994 are 11.69, 11.79, 11.89, and 11.99, respectively. Our CACE analysis using fuzzy scores on these imputed values would only be affected to the extent that the true values vary greatly outside the bounds given by 11.59 and 12.09. While it is possible that such sharp deviations might exist within the four years bounded inside the interval, general trends in educational attainments suggest that that would be a very unlikely scenario.

Analyses and Results Our analysis consists of two parts, each within the framework of a CACE analysis. First, we examine the effect of aggregate female education on female labor force participation rates. Second, we examine how aggregate female education levels may modify the effect of family policies on female labor force participation rates. In both cases we use government legacies— left versus non-left leaning—as mechanisms encouraging governments to enact policy and to expand education. In doing so, the education CACE results are directly comparable to the CACE results due to the different policies we study. On the one hand, education, including post-secondary education, may expand for reasons outside government policy making, especially where—as in the United States—there are large numbers of private colleges and universities. As well—and again as in the United States—government education policies may be as much or more a function of state and local government than of the federal government. On the other hand, education in Continental Europe often traditionally has been more centralized under national control—France perhaps represents the extreme

Family Policies, Education, and Female–Labor Market Participation | 41

of centralized national government control over education at all levels (Chartier and Croche 2006). And even where national educational systems are decentralized and include options for various kinds of private schooling, the national state typically provides regulatory oversight, as well as some direct and indirect funding or subsidies to educational institutions and students. Thus, using left- versus non-left-leaning national government legacies as mechanisms encouraging governments to enact policies expanding education is not unreasonable, given that it does not require us to assume that all educational expansion is due to national government policy making. Insofar as this is the case, our CACE analysis reveals the overall average causal effect of aggregate female education levels on female labor force participation rates. Thinking about educational expansion as government-promoted social policy is consistent with Wilensky’s (1975) seminal discussion of the nature and emergence of welfare states, in which he contrasted the egalitarianism of Scandinavian style welfare states with the equal opportunity ethic that made educational expansion a hallmark of countries such as the United States. According to Wilensky (1975), promotion of social mobility through expanding educational opportunity represented a policy alternative to redistributive social welfare policies. Emphasizing the need of a welfare state in the twenty-first century to make supply-side investments in education and training to increase human and social capital of children and adults alike, contemporary scholarship has recovered Wilensky’s (1975) emphasis on state education spending and regulations as a policy choice guided by political ideology (Esping-Andersen et al. 2002).

Causal Effects of Female Education Levels on Female Labor Force Participation Our CACE analysis of the aggregate female education effect on female labor force participation rates follows the approach detailed in Eliason, Stryker, and Tranby (2008). In general, the CACE estimate gives the average counterfactual causal effect were a complier to change status on the hypothesized causal variable (i.e., to change treatment status). In our case, treatment is here defined on fuzzy scores measuring membership in the set “high levels of female education” (for those ages twenty-five and older), and the control is defined on membership in the complement set. The first CACE estimate is based on left government legacies acting as the encouragement for governments to expand educational attainments for females; our second CACE estimate is based on other government lega-

42 | Robin Stryker, Scott R. Eliason, Eric Tranby, and William Hamilton

cies. Following Eliason, Stryker, and Tranby (2008), the instrument is also defined on fuzzy scores, measuring membership in the set “high cumulative left cabinet incumbency” for the first CACE estimate and its complement for the second. From this, the only assumptions necessary to obtain the first CACE estimate include the following: 1. Mechanisms exist, though need not be observed, such that high levels of cumulative left cabinet incumbency acts as an encouragement for governments to expand women’s education. 2. There are tendencies for governments to comply with these mechanisms and tendencies for governments not to comply with these mechanisms. 3. The tendency not to comply with these mechanisms is considered a mixture of a. the tendency always to expand female education regardless of the level of cumulative left cabinet incumbency and b. the tendency never to expand female education regardless of the level of cumulative left cabinet incumbency. 4. For governments complying with these mechanisms, the only non-negligible way that cumulative left cabinet incumbency at time t-1 can affect female labor force participation at time t is through compliance prior to time t. 5. Women’s educational attainment levels at time t and female labor force participation at time t+1 cannot affect cumulated left cabinet incumbency up to time t.

For the second CACE estimate, the complement set “not high cumulative left cabinet incumbency” is used as the encouragement instrument. The likelihood function used to estimate the CACE with fuzzy set scores under these types of assumptions, as well as additional assumptions necessary to identify the CACE with aggregate longitudinal data, are detailed in Eliason, Stryker, and Tranby (2008, see especially Appendix 5.4). Also as they did, we use the bootstrap empirical distribution on the estimated CACE and report medians, 5th, and 95th percentiles on the bootstrapped empirical distribution function. The bootstrap estimator is more desirable in this setting when compared to standard sampling distribution theory estimators because we need not assume asymptotic normality of the CACE for inferential purposes. Table 2.1 presents 5th, 50th (median), and 95th percentiles from the bootstrapped empirical distribution function (edf) for the estimated CACE of female educational attainment at time t-1 on female labor force participation at time t. The first panel gives the estimated CACE on levels of female labor force participation rates, while the second panel gives the estimated CACE on within-country change in the log-rate.5 Clearly, in the strong left legacies context, female educational attainment has a non-zero effect on the level of female labor force participation, with a CACE estimated to be in the range 0.03 to 0.10 and a 0.08 median CACE. That is, contrasting high to low levels of women’s educational attainment in the context of strong

Family Policies, Education, and Female–Labor Market Participation | 43

left political legacies gives an estimated causal effect on female labor force participation in the range of 3 percent to 8 percent.6 For non-left legacies, on the other hand, the estimated range of the CACE includes zero, indicating no causal impact. Similarly, for both left and non-left political legacies, we find no causal impact of aggregate female educational attainment on annual within-country change in female labor force participation. Combined, the information in Table 2.1 suggests that the non-zero CACE in the left legacy context is due to cross-national differences almost entirely. We return to this issue later below, as similar results are obtained for different policy effects. Table 2.1. Bootstrap EDF Estimates of compliers’ average causal effects of aggregate female education levels on female labor force participation.* Bootstrap EDF Estimates 5th Percentile

Median

95th Percentile

Effects on Levels Left Legacies Non-Left Legacies

0.03

0.08

0.10

-0.02

0.01

0.03

0.00

0.00

0.01

-0.04

-0.02

0.00

Effects on Within-Country Change Left Legacies Non-Left Legacies

*Bootstrapped estimates are based on 1000 replications. See text for details.

Female Education Levels and Policy Effects on Female Labor Force Participation The second part of our analysis examines the effect of different policies on female labor force participation, treating female education levels as an additional condition. Recall that these conditions need not be measured directly in our analysis but instead are given by the fuzzy set membership scores of being in the set “high levels of female education,” which in turn are based on the measure for average years of schooling described above. The policies we examine are precisely the same as those studied by Eliason, Stryker, and Tranby (2008), and the overall estimated CACE of these policies can be found there. Estimating the CACE with the additional fuzzy-set condition of aggregate female educational attainment is a simple extension of the fuzzy-set based CACE estimator Eliason, Stryker, and Tranby (2008) developed. This extension is detailed in the appendix to the current chapter.

44 | Robin Stryker, Scott R. Eliason, Eric Tranby, and William Hamilton

Left Political Legacies Table 2.2 gives bootstrap edf estimates for the CACE with left political legacies acting as the instrument encouraging governments to enact demand and supply side policies, separately for high and for low levels of female education. If female education conditions the CACE for a specific policy, say public daycare for younger children, then we would expect to see differences in the estimated CACE between high and low female education levels for that specific policy, in this illustration, for public daycare for younger children. More precisely, if the inner-90th percentile range of the bootstrap distribution does not overlap for public daycare for younger children, then we have evidence suggesting that female education does significantly condition the CACE for that policy. Table 2.2 shows evidence that female education levels significantly condition the CACE for some of the studied policies.7 The impact of female education is most noticeable for the effect of extended leaves on female labor force participation. Here we find that the inner-90th (and the inner-99th, not shown in the table) percentile range on the bootstrap CACE distribution for extended leaves is significantly lower for low, compared to high, female education levels. In fact, the estimated causal effect of extended leaves on female labor force participation, for governments operating in the context of strong left legacies, is negative, ranging from -13 percent to -10 percent, when joined with low female education levels. That effect is slightly positive, ranging from 1 percent to 4 percent, when coupled with high female education levels.8 Also evident is the significant impact of female education levels on the expanded public sector CACE. The median CACE in settings involving low aggregate female education is 0.40, or 40 percentage points; the median CACE in high female education settings is 0.32, or 32 percentage points. Evidence of a significant difference between these two settings is obtained by comparing the 5th percentile value for low female education with the 95th percentile value for high female education. Here we have six percentage points separating those two values, suggesting a significantly stronger positive effect of high public sector size when coupled with low aggregate levels of female educational attainment. Table 2.2 also shows that female education only modestly modifies the CACEs for daycare and maternity leave. Here we have one percentage point differences for both daycares, and a 2- percentage-point difference for maternity leave, separating the two inner-90th percentile ranges of the CACE distributions. In fact, if we expand the range on these distributions to capture the inner-99th percentiles for the high and low female educa tion contexts we find that the distributions for the CACEs overlap for

Family Policies, Education, and Female–Labor Market Participation | 45

Table 2.2. Bootstrap EDF Estimates of compliers’ average causal effects on female labor force participation rates, by aggregate female education levels, with strong left political tradition as the instrument in the CACE analysis.* Bootstrap EDF Estimates High Female Education 5th Percentile

Low Female Education

Median

95th Percentile

5th Percentile

Median

95th Percentile

0.31

0.32

0.33

0.39

0.40

0.42

Public Daycare— Ages 0-2

0.07

0.08

0.11

0.12

0.13

0.15

Public Daycare— Ages 3 to School Age

0.07

0.08

0.10

0.11

0.12

0.14

Maternity Leave

0.07

0.09

0.11

0.13

0.16

0.17

Extended Leave

0.01

0.03

0.04

-0.13

-0.11

-0.10

-0.23

-0.21

-0.19

-0.26

-0.22

-0.19

Demand Side Factor Expanded Public Sector Employment Supply Side Policies

Family/Child Cash & Tax Benefits

*Bootstrapped estimates are based on 1000 replications. See text for details.

both daycares and maternity leave. Therefore, at this level, there is no evidence in these data suggesting that these CACEs are different by female education levels. While Table 2.2 shows evidence of some significant differences in policy effects on levels of female labor force participation across aggregate levels of female education, Table 2.3 shows no significant differences across aggregate female education levels in the effect of policies on the withincountry change in female labor force participation, on the log-rate scale. Moreover, only three of these inner-90th percentile ranges on the CACE distributions do not include the null value of zero. These three instances of policy impacts are public daycare for both younger and older children in the high female education context and public daycare for older children in the low female education context.

46 | Robin Stryker, Scott R. Eliason, Eric Tranby, and William Hamilton

Table 2.3. Bootstrap EDF Estimates of compliers’ average causal effects on within-country change in female labor force participation log rates, by aggregate female education levels, with strong left political tradition as the instrument in the CACE analysis.* Bootstrap EDF Estimates High Female Education 5th Percentile

Low Female Education

Median

95th Percentile

5th Percentile

Median

95th Percentile

-0.02

-0.01

0.01

-0.01

0.00

0.01

Public Daycare— Ages 0-2

-0.02

-0.01

-0.01

-0.01

0.00

0.00

Public Daycare— Ages 3 to School Age

-0.01

-0.01

-0.01

-0.03

-0.02

-0.01

Maternity Leave

-0.01

0.00

0.02

0.00

0.00

0.01

Extended Leave

-0.02

0.00

0.03

-0.02

-0.01

0.00

Family/Child Cash & Tax Benefits

-0.01

0.01

0.04

0.00

0.01

0.02

Demand Side Factor Expanded Public Sector Employment Supply Side Policies

*Bootstrapped estimates are based on 1000 replications. See text for details.

While these three bootstrap CACE ranges for daycare do not include zero, the median effects themselves appear small, ranging from -0.01 to -0.02. However, translating the effects to the odds scale reveals a different story. Consider the median effect on the odds scale attributable to public daycare for older children conditional on low aggregate levels of female education, at exp(-0.02) = 0.98. This median effect indicates that publicly funded daycare programs for older children, against a backdrop of strong left political legacies and in the context of low levels of female education, drag down the odds of female labor force participation multiplicatively by 2 percent annually. So, for example, with a female labor force participation rate of 0.60, this median causal effect suggests a first year decline to 0.59. Continuing over, say, ten years, this causal effect translates into a decline of eleven percentage points, from 60 percent to 49 percent for the female labor force participation rate. Similarly, a median effect of -0.01 on the logodds scale—that attributable to both types of daycare in the context of high female education levels—translates into a 6 percentage point drop from 60 percent to 54 percent in the female labor force participation rate over the course of ten years. While small on the log-odds scale, such effects are nontrivial when translated into the odds scale over time.

Family Policies, Education, and Female–Labor Market Participation | 47

Other Political Legacies Tables 2.4 and 2.5 give estimates similar to those found in Tables 2.2 and 2.3 respectively, this time using other political legacies as the instrument in estimating the compliers average causal effects. Table 2.4 shows no significant differences in the CACEs across female education levels. This is evidenced by the overlap in the inner-90th percentiles across female education levels for each of the welfare state policies studied. Thus, there is no evidence, in the context of other political legacies, that aggregate female education modifies the causal effect of public sector size, public daycare, maternity leave, extended leaves, and family/child cash and tax benefits on female labor force participation. As we found in Eliason, Stryker, and Tranby (2008: Table 5.13), high public sector size, public daycare, and maternity leave all have significant and substantial positive causal effects on female labor force participation in the context of other political legacies. As well, extended leave has a statistically significant and substantial negative effect, and family/child cash and tax benefits have a modest negative effect, which here is highlighted with low levels of female education. Table 2.4. Bootstrap EDF Estimates of compliers’ average causal effects on female labor force participation rates, by aggregate female education levels, with other political traditions as the instrument in the CACE analysis.* Bootstrap EDF Estimates High Female Education

Low Female Education

5th Percentile

Median

95th Percentile

5th Percentile

Median

95th Percentile

0.44

0.45

0.45

0.46

0.46

0.47

Public Daycare— Ages 0-2

0.11

0.13

0.14

0.14

0.14

0.15

Public Daycare— Ages 3 to School Age

0.45

0.49

0.52

0.47

0.48

0.49

Maternity Leave

0.36

0.36

0.40

0.36

0.39

0.42

Extended Leave

-0.17

-0.15

-0.13

-0.21

-0.18

-0.15

Family/Child Cash & Tax Benefits

-0.04

0.00

0.02

-0.05

-0.04

-0.03

Demand Side Factor Expanded Public Sector Employment Supply Side Policies

*Bootstrapped estimates are based on 1000 replications. See text for details.

-0.03 -0.07 -0.04 -0.04

Public Daycare— Ages 3 to School Age

Maternity Leave

Extended Leave

Family/Child Cash & Tax Benefits -0.01

-0.02

-0.04

-0.02

-0.07

-0.05

Median

0.01

0.00

-0.01

-0.01

-0.04

-0.02

95th Percentile

-0.02

-0.03

-0.03

-0.02

-0.07

-0.03

5th Percentile

-0.01

-0.02

-0.02

-0.01

-0.04

-0.01

Median

0.00

-0.01

-0.01

-0.01

-0.01

0.00

95th Percentile

Low Female Education

Bootstrap EDF Estimates

-0.02

-0.03

-0.05

-0.03

-0.08

-0.06

5th Percentile

*Bootstrapped estimates are based on 1000 replications. See text for details.

-0.09

-0.08

Public Daycare— Ages 0-2

Supply Side Policies

Expanded Public Sector Employment

Demand Side Factor

5th Percentile

High Female Education

-0.01

-0.02

-0.03

-0.02

-0.05

-0.04

Median

Overall

0.00

-0.01

-0.01

-0.01

-0.02

-0.01

95th Percentile

Table 2.5. Bootstrap EDF Estimates of compliers’ average causal effects on within-country change in female labor force participation log rates, by aggregate female education levels, with other political traditions as the instrument in the CACE analysis.*

48 | Robin Stryker, Scott R. Eliason, Eric Tranby, and William Hamilton

Family Policies, Education, and Female–Labor Market Participation | 49

Similarly, Table 2.5 shows no significant differences across levels of female education, in the CACEs on within country change in the female labor force participation log-rate. To interpret these effects, we therefore include and focus on the overall CACEs in Table 2.5. Here, the only nonsignificant effect is obtained with family/child cash and tax benefits, as indicated by the inclusion of zero in the inner-90th percentile range for its CACE distribution. All other policies and programs show a negative effect on the log-rate of female labor force participation. Multiplicatively on the odds scale, an expanded public sector has an overall median causal effect of 0.96, corresponding to a ten-year decline from an initial female labor force participation rate of 0.60 to 0.40. Those values for public daycare for younger children, public daycare for older children, maternity leave, and extended leave are 0.95, 0.98, 0.97, 0.98, and 0.99, respectively. These correspond to ten-year declines from an initial female labor force participation rate of 0.60 to 0.36, 0.49, 0.44, 0.49, and 0.54 respectively.

Discussion and Conclusion Consistent with economic and sociological theory suggesting that increased education for women should shape female labor supply, we did find positive effects of high aggregate female educational levels on aggregate female labor force participation rates, when female educational expansion is encouraged through left government legacies. But these effects are not as large as the positive causal effects of high public sector size, high maternity leave, and high public daycare for younger and older children on women’s labor force participation rates across the fourteen advanced capitalist democracies from 1960 to 1999. When encouraged through non-left legacies, the causal effect of high aggregate female education was effectively zero. For both non-left and left legacies, we found no causal effect of aggregate female education on annual within-country change in the log-rates of female labor force participation. In sum, when analysis is refined to identify and estimate causal effects, aggregate expansion of female education appears to be less important in shaping aggregate rates of female labor force participation than Hicks and Kenworthy (2008) presumed it would be. As for the potential conditioning effect of aggregate female education on the causal effects on female labor force participation of maternity leave, daycare for younger and older children, and extended leaves, our results suggest that education’s role in moderating these causal effects is minimal. With left political legacies as the encouragement mechanism, aggregate female education had the strongest impact on influencing the effect of ex-

50 | Robin Stryker, Scott R. Eliason, Eric Tranby, and William Hamilton

tended leaves and the effect of high public sector size, and it had very weak to no impact on the effect of remaining policies and programs. With non-left political legacies as the instrument, there were no significant differences in the CACEs for the public sector and family policy measures across female education levels. Overall, whether or not it is conditioned significantly by high versus low aggregate female education levels, the causal impact of key family policies and of public sector size across advanced industrial democracies, including Austria and the smaller EU countries, as well as larger EU and non-EU countries from 1960 to 1999, remains substantial. When encouraged through left political legacies, we found especially high levels of positive causal impact for public sector employment. But we also found substantial positive causal effects for maternity leave and for public daycare for younger and older children. While these causal effects of supply and demand factors on female labor force participation rates were somewhat conditioned by aggregate female education levels, the general positive effect pattern was the same across high and low aggregate female education. Consistent with some concern expressed in the literature about the impact of especially lengthy leaves, extended leave has a negative causal effect in the context of low female education, when the mechanism encouraging the leave policy is left political legacies. Only when aggregate women’s education is high is the causal impact of extended leave levels very modestly positive. When non-left political legacies are the encouragement mechanisms for our CACE analysis, the causal impact of public daycare for children age three to school age on aggregate female labor force participation is especially large—indeed even larger than the causal impact of high public sector size. This impact of public daycare for older children is not conditioned by aggregate female education, nor is the very substantial positive causal effect of maternity leave or the lesser but still substantial impact of public daycare for children ages zero to two. Consistent with concerns expressed about extended leaves, when the instrument encouraging an extended leave policy is non-left political legacies, extended leaves have a substantial negative effect on aggregate female labor force participation. This is true both in the context of high aggregate female education and in the context of low aggregate female education. To this point, our results underscore the policy implications of Eliason, Stryker, and Tranby’s (2008) earlier research. Expanding the public sector, providing daycare for younger and older children, and providing maternity leave all appear to be effective in raising women’s aggregate labor force participation. Providing lengthy extended leaves, however, is coun-

Family Policies, Education, and Female–Labor Market Participation | 51

terproductive for the specific goal of enhancing aggregate women’s labor force participation, unless the instrument is cumulative left governance and the context is high aggregate female education. But what should we make of findings for the causal effects of public sector size, daycare, and maternity leave on annual within-country change in the log-rates of female labor force participation? Depending whether the encouragement mechanism is left or non-left political legacies, and whether the policy impacts are conditioned on high or low aggregate female education, these policy effects are either zero or negative. At first blush this seems surprising given the common presumption in scholarly and policy making circles that maternity leave and especially public provision of daycare promote female labor force participation. Results are explicable, however, when we combine the null and negative policy effects on within-country change in aggregate female labor force participation, found in Tables 2.3 and 2.5, with the positive causal effects of high levels of maternity leave and public provision of daycare for young and older children on aggregate levels of female labor force participation across countries over the entire 1960–1999 time period, found in Tables 2.2 and 2.4. In this context, these negative causal effects on within-country change over time may reflect a ceiling effect that necessarily would dampen trajectory slopes that start out higher than others. Overall, then, our results suggest that policy makers who seek to encourage high levels of aggregate female labor force participation do get enhanced female labor force participation bang for the buck devoted to public daycare for both younger and older children and to maternity leave. For extended leave, this is true only under highly restricted conditions. This does not mean that extended leaves are a bad idea, only that they must be justified by reasons other than the desire to expand female labor force participation. At the same time, policy makers’ enthusiasm for public daycare and for maternity leave based on their impact on aggregate female labor force participation should be tempered by awareness of ceiling effects such as those suggested by our analyses. That high aggregate levels of female education had a positive impact on aggregate female labor force participation when female educational expansion was encouraged by left governments is consistent with the “human capital development” reorientation to the twenty-first-century welfare state urged by Esping-Andersen and others. However, that there is no such positive impact when educational expansion is encouraged by non-left governments, and that there is no such positive impact on annual within-country change of log-rates of female labor force participation for either left or non-left legacies suggests that the human capital development approach may be substantially more limited in its payoffs than EU policy

52 | Robin Stryker, Scott R. Eliason, Eric Tranby, and William Hamilton

makers suppose. At the same time, for women in non-left legacy contexts, enhanced aggregate education has not translated into the increased aggregate employment necessary to fulfill the equal-opportunity function that Wilensky assumed for education. It is possible that more nuanced analysis would show that enhanced aggregate education for women in non-left legacy contexts is associated with shifts in the female job distribution from part to full time employment or from lower to higher paying jobs, even if those education increases do not translate into increased aggregate female labor force participation. However, for education to promote substantial upward mobility for women, there must be a sufficient number of high quality jobs to absorb a growing supply of more highly educated women. This is, to say the least, not a foregone conclusion.

Appendix

Estimating Compliers’ Average Causal Effects Using Fuzzy-Set Membership Scores on Intention and Treatment Variables with Additional Fuzzy-Set Conditions In this appendix we extend the fuzzy-set based compliers’ average causal effect estimator developed by Eliason, Stryker, and Tranby (2008) to handle additional fuzzy-set conditions. Here we detail the likelihood function for a single additional fuzzy-set scored condition. Extensions to additional conditions are straightforward. Let cj , aj , and nj be the compliance and noncompliance probabilities described in Eliason, Stryker, and Tranby (2008), but here defined on the additional fuzzy-set condition and its complement, for j = 1 and 0 respectively. Similarly, define Y1itj(Y2itj) as the female labor force participation rate for country i at time t under high (low) levels of a policy’s implementation in country-time (i,t), defined on the fuzzy-set condition and its complement for j = 1 and 0 respectively. Further, let sxit and sitz be fuzzy scores as defined in Eliason, Stryker, and Tranby (2008). Finally, let sitjc be the fuzzy score measuring the degree to which country-time (i,t) belongs to the condition set and its complement for j = 1 and 0 respectively, with j sitjc = 1. For our analysis, the condition set is given by “high levels of female education.”

Family Policies, Education, and Female–Labor Market Participation | 53

The extension of the Eliason-Stryker-Tranby fuzzy-set-based compliers’ average causal effect estimator to accommodate an additional single-set condition, is given by 1

I

sxit sitz sitjc

T

CACEC = cj fc1j Y1itj aj fa1j Y1itj j=0 i=1 i=1

sxit (1-sitz ) sitjc

aj fa2j Y1itj

(1-sxit)(1-sitz ) sitjc

 cj fc2j Y2itj nj fn2j Y2itj

(1-sxit)sitz sitjc

 nj fn1j Y2itj

(For details see Eliason, Stryker, and Tranby [2008].) The CACE is defined as usual, except here there is a CACE given separately for the set defined on the condition, CACE1 = yfc11 1it1dy  yfc21 2it1dy, and the complement of the set defined on the condition, CACE0 = yfc10 1it0dy  yfc20 2it0dy.

Notes 1. A related key issue, especially for gender-egalitarian scholars and policy makers is improving the quality of women’s employment, including achieving gender pay equity and upgrading part-time employment (Misra, Budig, and Moeller 2005; Charles and Grusky 2004). These issues are beyond the scope of the current chapter. 2. See Gornick, Meyers, and Ross 1997; Daly 2000; Korpi 2000; Stier, Lewis-Epstein, and Braun 2001; Rønsen and Sunstrőm 2002; Morgan and Zippel 2003; Misra, Budig, and Moeller 2005; Petit and Hook 2005; Eliason, Stryker, and Tranby 2008; Hicks and Kenworthy 2008. 3. The fuzzy set methods used by Eliason, Stryker, and Tranby (2008) build upon Charles Ragin’s Fuzzy Set Social Science (2000; see also Ragin and Pennings 2005), but their methods differ somewhat from Ragin’s (2000; Ragin and Pennings 2005) by, among other things, allowing for measurement error and providing goodness-of-fit tests for hypotheses of causal necessity, sufficiency and necessity, and sufficiency combined. These methods are described in detail in Eliason and Stryker (2009) and Eliason, Stryker, and Tranby (2008). The latter likewise provides detailed information on the conversion of raw data on long term social democratic incumbency of government, public sector size, public provision of daycare, and maternity leave (including information on public



54 | Robin Stryker, Scott R. Eliason, Eric Tranby, and William Hamilton

4.

5.

6. 7.

8.

expenditures, scope of coverage, leave duration, and benefit replacement levels), to fuzzy set scores. See Morgan and Winship (2007) for an informative treatment of assessing causal effects with non-experimental data. See also Heckman, LaLonde, and Smith (1999) for additional complications that arise, and solutions, when using time series data to estimate causal effects. The log-rate is more appropriate here than the rate per se when studying rates of change. See Clogg and Eliason (1987) and Clogg, Eliason, and Grego (1990) for details on log-rate and related models for over-time data. We also estimated the CACE using a five-year instead of one-year lag. This produced similar results to those reported in Table 2.1. As can be seen in Table 2.2, aggregate female educational attainment does not condition the CACE for family/child cash and tax benefits. These include family allowances, family support benefits, and child care tax relief. Thus, these policies constitute a heterogeneous policy grouping, and Eliason, Stryker, and Tranby (2008) found that overall, this group of policies had a negative causal impact on female labor force participation when encouraged through the instrument of left governance legacies. We included family/child cash and tax benefits in Table 2.2 here only for purposes of comparison with prior results in Eliason, Stryker, and Tranby (2008) that were not conditioned on female educational attainment. For comparison, Eliason, Stryker, and Tranby (2008) found that the unconditional extended leave causal effect was null. This makes sense, given that such an unconditional effect combines the negative effect for women with low education levels and the positive effect for women with high education levels.

Bibliography Barro, Robert J., and Jong-Wha Lee. 2000. “International Data on Educational Attainment: Updates and Implications.” Center for International Development Working Paper No. 42. Becker, Gary S. 1975. Human Capital: A Theoretical and Empirical Analysis with Special Reference to Education. 2nd Ed. New York: Columbia University Press for NBER. Blau, Francine D., Marianne A. Ferber, and Anne E. Winkler. 2002. The Economics of Men, Women and Work, 4th ed. New York: Prentice Hall. Charles, Maria, and David B. Grusky. 2004. Occupational Ghettos: The Worldwide Segregation of Women and Men. Stanford: Stanford University Press. Chartier, Jean-Emile, and Sarah Croche. 2006. “How European Integration is Eroding National Control over Education Planning and Policy.” European Education 37, no. 4: 7–21. Clogg, Clifford C., and Scott R. Eliason. 1987. “Some Common Problems in LogLinear Analysis.” Sociological Methods and Research 16, no. 1: 8–44.

Family Policies, Education, and Female–Labor Market Participation | 55

Clogg, Clifford C., Scott R. Eliason, and John Grego. 1990. “Models for the Analysis of Change in Discrete Variables.” In Statistical Methods in Longitudinal Research, Volume II: Time Series and Categorical Longitudinal Data, edited by Alexander von Eye, 409–42. New York: Academic Press. Daly, Mary. 2000. “A Fine Balance: Women’s Labor Market Participation in International Comparison.” In Welfare and Work in the Open Economy: Diverse Responses to Common Challenges, vol. 2, edited by Fritz W. Scharpf and Vivien A. Schmidt, 467–510. Oxford: Oxford University Press. Eliason, Scott. 1995. “An Extension of the Sorensen-Kalleberg Theory of the Labor Market Matching and Attainment Process. American Sociological Review 60, no. 2 (April): 247–71. Eliason, Scott, and Robin Stryker. 2009. “Goodness-of-Fit Tests and Descriptive Measures in Fuzzy Set Analysis.” Sociological Methods and Research 38, no. 1: 102–46. Eliason, Scott, Robin Stryker, and Eric Tranby. 2008. “The Welfare State, Family Policies and Women’s Labor Market Participation: Complementary Fuzzy-Set and Compliers’ Average Causal Effects Analyses.” In Method and Substance in Macrocomparative Analysis, edited by Lane Kenworthy and Alex Hicks, 135–95. New York: Palgrave-MacMillan. Esping-Andersen, Gøsta. 1990. The Three Worlds of Welfare Capitalism. Princeton: Princeton University Press. Esping-Andersen, Gøsta, John Myles, Anton Hemerijck, and Duncan Gallie. 2002. Why We Need a New Welfare State. Oxford: Oxford University Press. Freeman, Richard. 1976. The Overeducated American. New York: Academic Press. Gauthier, Anne. 1996. The State and the Family: A Comparative Analysis of Family Policies in Industrialized Countries. Oxford: Clarendon. Gornick, Janet, Marcia Meyers, and Katherin Ross. 1997. “Supporting the Employment of Mothers: Policy Variation Across Fourteen Welfare States.” Journal of European Social Policy 7, no. 1: 45–70. Hantrais, Linda. 2004. Family Policy Matters: Responding to Family Policy Change in Europe. Bristol: Policy Press. Heckman, James J., Robert J. LaLonde, and Jeffrey A. Smith. 1999. “The Economics and Econometrics of Active Labor Market Programs.” In Handbook of Labor Economics, vol. 111, edited by Orley Ashenfelter and David Card, 1865–2097. New York: Elsevier. Hicks, Alexander, and Lane Kenworthy. 2008. “Family Policies and Women’s Employment: A Regression Analysis.” In Method and Substance in Macro-Comparative Analysis, edited by Lane Kenworthy and Alex Hicks, 196–220. New York: Palgrave-MacMillan. Hirano, Keisuke, Guido W. Imbens, Donald B. Rubin, and Xiao-Hua Zhou. 2000. “Assessing the Effect Of An Influenza Vaccine In An Encouragement Design.” Biostatistics 1, no. 1: 69–88. Huber, Evelyne, and John Stephens. 2000. “Partisan Governance, Women’s Employment and the Social Democratic Service State.” American Sociological Review 65, no. 3: 323–42.

56 | Robin Stryker, Scott R. Eliason, Eric Tranby, and William Hamilton

Imbens, Guido W., and Donald B. Rubin. 1997. “Bayesian Inference for Causal Effects in Randomized Experiments with Noncompliance.” The Annals of Statistics 25, no.1: 305–27. Jencks, Christopher. 1972. Inequality: A Reassessment of the Effect of Family and Schooling in America. New York: Basic Books. Kalleberg, Arne. 1977. “Work Vales and Job Rewards: A Theory of Job Satisfaction.” American Sociological Review 42, no. 1: 124–43. Korpi, Walter. 2000. “Faces of Inequality: Gender, Class and Patterns of Inequalities in Different Types of Welfare States.” Social Politics 7, no. 2 (Summer): 127–91. Lewis, Jane. 1992. “Gender and the Development of Welfare State Regimes.” Journal of European Social Policy 2, no. 3: 159–73. Misra, Joya, Michelle Budig, and Stephanie Moeller. 2005. “Employment, Wages and Poverty: Family Policies and Gender Equity.” Paper presented at Annual Meetings of Eastern Sociological Association, Washington D.C., April. Morgan, Kimberly J., and Kathrin Zippel. 2003. “Paid to Care: The Origins and Effects of Care Leave Policies in Western Europe.” Social Politics 10, no. 1: 49–85. Morgan, S. L., and Christopher Winship. 2007. Counterfactuals and Causal Inference: Methods and Principles for Social Research. Cambridge: Cambridge University Press. Myles, John, and Jill Quadagno. 2002. “Political Theories of the Welfare State.” Social Service Review 76, no. 1 (March): 34–66. Orloff, Ann Shola. 1993. “Gender and the Social Rights of Citizenship: The Comparative Analysis of Gender Relations and Welfare States.” American Sociological Review 58, no. 3: 303–28. Petit, Becky, and Jennifer Hook. 2005. “The Structure of Women’s Employment in Comparative Perspective.” Social Forces 84, no. 2: 779–801. Ragin, Charles. 2000. Fuzzy-Set Social Science. Chicago and London: University of Chicago. Ragin, Charles and Paul Pennings. 2005. “Fuzzy Sets and Social Research.” Sociological Methods and Research 33 (May): 423–30. Rønsen, Marit and Marianne Sundstrőm. 2002. “Family Policy and After-Birth Employment Among New Mothers: A Comparison of Finland, Norway and Sweden.” European Journal of Population 18: 121-152. Rosenfeld, Rachel A. and Arne L. Kalleberg. 1990. “A Cross-National Comparison of the Gender Gap in Income.” American Journal of Sociology 96, no. 1: 69–106. Scharpf, Fritz, and Vivienne Schmidt, eds. 2000. Welfare and Work in the Open Economy, vol. 1. Oxford: Oxford University Press. Sewell, William H., and Robert Hauser. 1975. Education, Occupation and Earnings: Achievement in the Early Career. New York: Academic Press. Stier, Haya, Noah Lewis-Epstein, and Michael Braun. 2001. “Welfare Regimes, Family-Supportive Policies and Women’s Employment Along the LifeCourse.” American Journal of Sociology 106, no. 6: 1731–60. Stryker, Robin. “Religio-Ethnic Effects on Attainments in the Early Career.” American Sociological Review 46, no. 2 (1981): 212–31.

Family Policies, Education, and Female–Labor Market Participation | 57

―――. 1996. “Comparable Worth.” In Women and Work: A Handbook, edited by Paula J. Dubeck and Kathryn Borman, 74–77. London and New York: Garland. Tranby, Eric. 2008. “Family Policies and Women’s Employment: Welfare State Policies, Labor Markets and Female Labor Force Participation in 14 Advanced Capitalist Democracies.” Unpublished Manuscript, University of Minnesota, 10 October 2008. Wilensky, Harold L. 1975. The Welfare State and Equality: Structural and Ideological Roots of Public Expenditure. Berkeley: University of California Press.

Chapter 3

DOUBLE TRANSFORMATION How to Adjust Institutional Social Policy?

( Juho Saari

Introduction Welfare states around the postindustrialized world have faced various pressures from the mid-1980s, the end of the so-called Golden Age. Since then, in most welfare states, the annual real increases in public/social expenditures have largely resulted from commitments made some twenty or thirty years beforehand on pensions, health care, child care, unemployment, and sickness insurance. In order to maintain these built-in commitments, most welfare states introduced policies aimed at maintaining a proper balance between the resources and the costs and promoting employment and economic growth (see also Korpi 2003). Yet the institutional structures of the welfare state matter, as there is evidence indicating significant differences in the scope of these policy measures across countries, regardless of similarities in structures (e.g., postindustrialization, urbanization, aging) and challenges (e.g., globalization, unemployment, demographic shifts). Indeed, high spending levels in social expenditure do not necessarily generate cuts; in fact the opposite often (but not always) seems to be true. In this context, the otherwise rather insignificant Finland, an atypical Scandinavian welfare state of some five million inhabitants, is of considerable theoretical, institutional, and empirical interest as an extreme case study, particularly in light of the 2008 financial crisis. In many ways Finland has already experienced such a collapse. Finland had significant

60 | Juho Saari

built-in commitments to increase public expenditure when it witnessed the worst recession (1991–93) experienced in any OECD country since World War II: with an increase in unemployment from 1.8 percent of the labor force (1990) to 16.5 percent (1993), public debt from 15 percent of GDP (1990) to 56 percent (1993).1 This recession was followed by one of the longest periods of continuous prosperity. The post-recession GDP growth from 1994 to 2006 clearly exceeded the growth that the post-recession conventional catch-up effect would envisage. This implies that from the economic and fiscal points of views, Finland was able to significantly upgrade its growth and employment policies in creative ways in order to meet new demands and expectations. It did so by inventing new paradigms and implementing innovations. Around the turn of the twenty-first century, Finland regularly scored highly in different world rankings: from the world competitiveness index and business environment index to the quality of life, human development, information society, and sustainable development indexes. In fact, not a single other country in the world generated such a balanced picture (Saari 2006): the public economy showed a surplus; the employment rate significantly improved (and unemployment declined); and, although income differences did dramatically increase, the average standard of living improved according to all conventional metrics. Figure 3.1. Social expenditure, year-to-year percentage in real terms, percent (left), and social expenditure, percent of GDP (right). Created by the author.

Double Transformation: How to Adjust Institutional Social Policy? | 61

The adjustment of the welfare state through this double boom-bust transformation was far from a smooth one. To illustrate this, consider changes in expenditure on social protection as a percentage of GDP. Figure 3.1 shows that it fluctuated to a remarkable extent during the 1990s. In 1990, 25.1 percent of GDP was allocated to social expenditure. Only three years later, the figure was no less than 34.6 percent, an increase of 9.5 percentage points. This trend alone is an interesting story. However, in 2000, Finland’s expenditure on social protection was down to 24.9 percent; that is lower than in 1990. In the late 1990s, there was virtually no real annual increase in social expenditure. Since then Finland has reached its pre-recession levels, which has more recently been followed by some increase in relative terms (the latest figure is 26.2 percent for 2006). The puzzle of retrenchment policies that this chapter analyzes is how adjustments to the Finnish welfare state—to the state’s institutions and objectives—both maintained the welfare state’s basic institutional structures and yet remained conservative in terms of increases in real expenditure through this double transformation. In particular, it focuses on the factors that explain Finland’s capability to adapt to into new circumstances by reforming its policy paradigms, agendas, and policy frames. To do so, I investigate this change at three institutional levels. First, I introduce the basic institutional framework of the Finnish welfare state for readers unfamiliar with the Scandinavian welfare state tradition. Second, I show how its governance structures differentiate and “cage” various social risks. Third, I examine new policy paradigms, policy agendas, and policy frames that have been designed to meet these objectives since the recession of the early 1990s. Finally, I review the role of the European Union (EU) in Finnish social and health policies, showing the influence of multilevel governance and policy transfers, a key theme of this collection and, increasingly, a focus social-policy analysis in general.

The Evolution of the Institutional Framework Scholars have devoted systematic attention to the factors explaining differences in retrenchment across welfare regimes and in comparative historical perspectives. Much of this retrenchment literature focuses on institutions, constitutional veto points, path dependency, feedback loops between policies and politics, and kinds of policy complementarities (e.g., Pierson 2001b; Crouch 2005; Starke 2005; Deeg 2007). In the Finnish case, these mechanisms, and related path dependencies that result from self-reinforcing processes do explain much of the institutional inertia.

62 | Juho Saari

However, this chapter approaches retrenchment policies from a different angle. It looks at institutional change at different levels, asking how these levels adapt to retrenchment pressures. I start by analyzing horizontal principles that form the basic structures of social policies since the late 1960s. Second, I examine changes in the governance structures aimed at governing certain social risks. Finally, I focus on changes in ideas or conventional wisdoms and their roles in explaining the rise of new social policy objectives since the 1990s.

Horizontal Principles Since the 1960s (and in some cases, since the mid-1970s), Finnish social policies relied on a set of basic, or “horizontal,” institutional principles that were applied in all or most institutions and have strong path-dependent characteristics (Kangas and Saari 2008): Close links to social risks and related governance structures: Social policy programs are designed to respond to specific social risks, such as childhood, sickness, and old age, not to respond specifically to poverty or social exclusion, which are clearly secondary objectives. Individuality: Social insurance benefits and taxation are based solely on the individual’s own work history, payments, and income. There are no significant derived rights or duties, with the exception of some widows and orphans’ benefits. Independence: Individuality also means that there are no notable legal dependencies between adult generations in social or public law. However, as an exception that confirms the rule, Finland’s marriage law rules that spouses are responsible for each others’ welfare during their marriage. Collectivism: In principle, all citizens are covered by the same (or same kind of) schemes regardless of their status or socioeconomic position. The role of occupational or so-called second-pillar benefits is limited. Coordination: Individuals have a right to move from one municipality to another or from one workplace, company, or industry to another without losing any of their social rights or previous (defined) contributions. Furthermore, taxation and transfers are coordinated in earnings-related benefits. Legislated benefits: Almost without exception, Finnish social policy is based on laws and statutes, and the role of occupational or individual social protection (in jargon, second and third pillar) is negligible, although their roles in some policies are rapidly expanding.

Double Transformation: How to Adjust Institutional Social Policy? | 63

Residence-based benefits and services: The fundamental criterion for receiving services and transfers is residence, not employment. Thus, the coverage of social policy (social security, in jargon, first pillar) is wider than in many other EU countries. Emphasis on income transfers and social services: With very few exceptions, assets have no impact on rights to social transfers or services.

Partly by historical accident and partly by design, many of these characteristics have proved to be quite flexible and adaptive, allowing for significant institutional adjustments without major retrenchment both during the booms and the recessions. Consequently, they remained intact, experiencing only minor adjustments during the double transformation. In fact, the Constitutional Reform Act of 1995, which introduced several social rights as constitutional rights, further strengthened several of these basic institutions.

Governance Structures The next level of institutional analysis focuses on governance structures. Much of social policy research examines particular institutional arrangements or administrative structures (like social insurance or services in kind, like child care). Such approaches are important, but they neglect the fact that transfers and services are sometimes functional substitutes or have complementarities to each other. Furthermore, in some cases, instead of transfer or services, the different forms of indirect regulation may play a crucial role, like the employers’ duties to look after their aging employees. For instance, some forms of social insurance, like sickness insurance, developed rather late in Finland (compared to other similar industrializing countries, see Kangas and Palme 2005) quite simply because functional substitutes (e.g., publicly provided health services and the socially endorsed duty of employers to pay various benefits) fulfilled their roles. Clearly, the history of social policy looks quite different if we take different functional substitutes into account and examine the interplay of different substitutes in governing different social risks. When we approach social policy from this angle, we see that it has expanded to meet new social risks by designing new governance structures in three waves (see Table 3.1). First, until the mid-1960s the bulk of social policy focused on social risks that were integrally connected to (small) children, the aging (including frail elderly), and the sick. Policy addressed these risks through extensive packages of social transfers (means-tested benefits, needs-tested, additional benefits, universal and earnings-related benefits, tax deductions, and fringe benefits), and social and health ser-

64 | Juho Saari

vices of different kinds (institutional and noninstitutional). Second, from the mid-1960s until the late 1980s, the main focus of social policy was on housing, unemployment, education (and related subsidies), and reconciling work and family life. Again, an extensive network of institutionalized social transfers and social, health, and educational services worked to “cage” these risks. Finally, since the 1990s, three new groups of social risks were identified: long-term unemployment (or the excluded ones), excessive indebtedness, and immigration (Jäntti, Saari, and Vartiainen 2006). The institutional model evolved and expanded smoothly, first, from those outside of the labor force to those inside of it, and, second, to those who cannot be classified in those terms. From another angle, one can see social policy as expanding from the clearly deserving and vulnerable groups toward those who have more alternatives and responsibilities over their choices and behaviors and, finally, to those who have traditionally been treated with some suspicion. As the new risk groups have been integrated into socio-political systems, institutional incentives and moral hazards have become politically more central. The pressure created by these rapidly generated groups that could not—for a variety of reasons—be governed by already existing governance structures explains the branching points, timings, and sequences of the most reforms (Saari 2001). Table 3.1. Three waves of social policy in Finland, 1945–2007 Mid-1960s

Mid-1960s to late 1980s

Early 1990s to today

Social risks

Childhood Sickness Aging

Education Housing Unemployment Work/family

Indebtedness Long-term unemployment Immigration

Deserving

Clearly

Partly

Unclear

Labor market status

Passive

Active

Mixed

Fiscal costs (relatively) Large

Significant

Minor

Incentives

Not significant

Partly significant

Significant

Moral hazard

Small

Mixed

Large

During Finland’s double transformation these governance structures continued to evolve and expand even during and after the recession of the early 1990s. However, much of this expansion occurred outside the con-

Double Transformation: How to Adjust Institutional Social Policy? | 65

ventional boundaries of social policy, resulting in the myth of permanent stagnation, or austerity, in social policy (illustrated in Figure 3.1 above). Furthermore, from the administrative point of view, other ministries aside from the Ministry of Social Affairs and Health (MSAH) assumed major responsibility for the new governance structures. For instance, the responsibility for immigration was divided between the ministries of Interior and Labor; for long-term unemployment, between the Ministry of Labor and the MSAH; and responsibility for excessive indebtedness was allocated to the Ministry of Justice (see Table 3.2 for details). Table 3.2. Governance of social risks: the division of responsibilities (2006)

MSAH Childhood

Labor*

Education

Interior

Justice

XX

Sickness

XX

Old age

XX

Housing

X

X

XX

Unemployment

XX

Family/Work

XX

X

X

XX

Education Indebtedness

XX

X

Long-term unemployment Immigration

Environment

XX

XX

XX

X

XX

XX

X XX

XX = major responsibility, X = minor responsibility * The Ministry of Labor was merged with the Ministry of Trade and Industries in 2008.

New governance structures were also funded through means that were not included in social expenditure statistics (like active labor-market policy measures, agreements between the banks and the persons on excessive debts, and immigration costs). Therefore, significant sociopolitical expansion occurred out of social politicians’ sights and used new methods of financing (and regulation), resulting in a biased picture on the evolution of Finland’s socio-political model and its expenditure levels since the mid-1990s. This does not imply that there have been no revisions within the governance structures. Most governance models did experience minor revisions, and unemployment and activation policies were drastically revised. Furthermore, the rapid evolution of three new governance structures affected the existing structures, redesigning and redefining their bound-

66 | Juho Saari

aries. Finally, individual institutions experienced an endless number of minor revisions aimed at more efficient and transparent allocation of resources and curbing of the rising costs of social transfers and social and health services. Quite often, these changes lowered replacement levels to the levels of the late 1980s and slightly modified access to social and health services by introducing user charges that made the costs more transparent to service users. However, in several cases such reforms have been misinterpreted as cuts, when in practice functional substitutes have taken stronger roles in governing certain social risks. Much of the policy implemented during and after the recession was in this respect quite “rational” by its political nature (for the details, see Saari 2001).

Changes in Ideas Analysis of changes in the horizontal principles and governance structures does not fully cover the rather fundamental change in social policy thinking. Recently, more attention has been paid to the roles of policy paradigms, agenda-setting skills, and policy frames in explaining the internal restructuring of welfare states. Such internal restructuring is a way to reallocate public resources to increase (or decrease) the long-term adaptive efficiency in a society (North 1990). Furthermore, as the decision-making process or broader political environment become more complex, arguably, factors like paradigms, agenda setting, and policy framing start to play more important roles in explaining choices. Under such circumstances decisions are less rational and stochastic and rely more on different kind of belief systems. Clearly, in order to understand the sociopolitical reform process of the 1990s, one should also focus on the other side of the complexity boundary, where (shared) mental models shape the paradigms, agendas, and frames in ways that explain the choices among policy alternatives during various “rare moments” characterized by deep (or Knightian) uncertainty (see Denzau and North 1994; Ostrom 2005; see also Belánd 2007). The following section argues that the role of key policy makers was quite influential in designing new paradigms, agendas, and frames during the short interim period—the rare moment—that took place after the deepest recession but before the late 1990s boom. In looking at this time period one should focus on the role of ideas, in addition to structures, power resources, and institutions.

Double Transformation: How to Adjust Institutional Social Policy? | 67

STRUCTURES (Convergence)

POWER-RESOURCES (Divergence)

INSTITUTIONAL STRUCTURES (Path-making and -breaking processes)

Complexity boundary

POLICY PARADIGM

POLICY AGENDAS

POLICY FRAMES

POLICY OUTCOMES

Figure 3.2. Paradigms, agendas, and frames during the primeval soup. Created by the author.

The Primeval Soup of 1992–95: Paradigms, Agendas, and Frames During the darkest moments of the free fall in (public) economy from 1991 to 1992, policy makers implemented reforms on an hourly basis without much strategic thinking. In fact, the government’s own forecasts indicate that it did not believe that the recession would be so deep (see Saari 2001: Chapter 5). Rather, it forecasted a minor and temporary recession, actually proposing to expand public spending in its government program (March 1991). Then, the reality kicked in. From late 1991 to 1993 Finland experienced an annual negative growth rate of 6 to 8 percent, with Knightian uncertainty shadowing its future (see Blyth 2002). The government responded with policies in three different areas, a difficult task for a center-right government (1991–95) that had assumed power just months before the collapse. First, there was conflict over monetary and economic policy. The government had to decide whether to devalue the currency (saving export industries and sacrificing domestic

68 | Juho Saari

industries and indebted households) or stick with the strong currency (with the opposite effect); in the end, the government of Finland opted for devaluation, with all its consequences, including banking crises. Second, the government saw the exceptionally rapid escalation of public debt as a key challenge (it rose from 15 percent to 58 percent of the GDP between 1990 and 1994), and it advanced policies that combined expenditure cuts, reliable long term policy commitments, and extensive borrowing from global markets (mainly from Japan). Finally, the relationships between the social partners and the government were quite tense. The conservative government had several times violated an established division of work. Yet the social partners (both employers and the unions) often promoted their narrow interests at the expense of the common good—typical for encompassing organizations, even those supposedly representing broad societal interests. This conflict resulted in short-termism and political cost shifting, generating some major strikes and other irregularities. Once the most urgent crisis passed, economic growth gradually turned positive, and the escalation of public debt stabilized within fiscally acceptable limits. In late 1992 and early 1993 new strategic approaches emerged. All the key actors (as well as a scientific community and other minor players) began to formulate new approaches. The history of Finnish social policy research from late 1992 to early 1995 represents a kind of primeval soup (Kingdom 1995) of competing policy paradigms, policy agendas, and policy frames. This process started with some proposals from the Confederation of the Employers in 1991 and was followed by others in 1992 and 1993. Yet by 1995, when the Social Democratic/ conservative/ left-wing (socalled “rainbow”) government entered office and hammered out its policies in its governmental program, the time of primeval soup was over. A new national consensus evolved, which the politics—and selection processes—of the elites maintained until the 2008 recession. This rapid consolidation of ideas resulted from the “rainbow” government pushing a compulsory consensus of an almost religious nature, allowing very little debate on the key guidelines of policies, even within the government, let alone between the government and oppositions. In fact, within the government there was an inner circle of like-minded senior politicians that excluded potential dissidents from decision-making processes. Furthermore, close connections between the employers and the conservatives on the one hand, and the employees’ confederation and the Social Democrats on the other hand, generated partnerships and dense networks that streamlined their policy objectives. Finally, the consolidation of the new consensus was strengthened by the seeming success of these policies; indeed, the governments of the late 1990s were quite popular.

Double Transformation: How to Adjust Institutional Social Policy? | 69

During the 1992–95 period, a number of different policy paradigms were proposed. At one end, several analysts offered models with neoliberal flavors, with some promoting New Zealand’s liberalization of key markets and models of interest mediation as a good practice. These analysts also argued that basic income schemes should replace existing institutional solutions. At the other end of the political spectrum and economic thinking, other analysts underlined a need to increase public expenditure through semi-Keynesian fiscal decisions in order to counterbalance the decreasing demand of domestic markets (this was Sweden’s strategy). Despite these differences, the issue of economic/fiscal and, consequently, social policy paradigms was quite rapidly settled: the government, together with social partners, focused on competitiveness, the stability of public economy, employment growth, etc., and paid quite little attention to redistribution, equality, equal opportunities, and social justice, all of which virtually vanished from the Finnish political (government’s) vocabulary. The recession years also witnessed varying comprehensive agendas (or “policy packages,” as they are commonly called in Finland). Different groups made extensive proposals. The most important ones were “the Satanic verses” of the employers’ confederation, the “Sailas lists” (named after Raimo Sailas, the head of the Budget Department and later the permanent secretary at the Ministry of Finance), the “Sorsa mediation” (named after Kalevi Sorsa, the mediator and previous prime minister), and several proposals by the government that all included lists of cuts in social transfers and social services (see Table 3.3). Here we see several straightforward findings. First, during the recession, power of the social partners to set key agendas in social policy shifted from the employees’ to employers’ confederations and has remained there (see first column from the left in Table 3.3). Second, within the government there was a rapid shift in power balance in social policy–agenda setting from the MSAH to the Ministry of Finance (MF), as the MSAH was not able to rapidly set alternative agendas (see second column from the left in Table 3.3). This shift can be seen in the changes of 1994: When the MF was already implementing micro-level policies aimed at simultaneously promoting employment and curbing the expansion of the costs and macrolevel cuts in social expenditure, the MSAH was proposing to increase the absolute levels of social assistance and other minimum benefits—changes that would have significantly increased costs. During the critical years until the late 1990s, the MF designed the government’s key social policy agendas.

70 | Juho Saari

Table 3.3. Agenda setting during the primeval soup “Satanic verses” (Employers, 1991)

Sailas (Ministry of Finance, 1992)

The government proposal in the context of the Sorsa mediation (6 November 1992)

The new government proposal as part of incomes policy agreement (December 1992)

• The abandonment of annual indexation-based increases in all social transfers in 1992. • The abandonment of employers’ national pension contribution. • The (then) temporary reduction of employers’ health insurance contribution (1 percent) to be made permanent. • The reduction of employers’ socialsecurity (pension) contribution by 2.5 percent percentage points and the introduction of employee pension contribution both in private and public sector. • Rearrangement of the unemployment security– financing model so that the share of the state would rise from 48 to 56 percent, the share of employers would decrease from 47 to 22 percent, and the share of the insured would increase from 5 to 22 percent.

• The abandonment of indexationbased increases in social transfers. • The abandonment of all planned reforms, including the child-care reform. • The restructuring of state-municipality grants system. • The increase in user charges, including health care charges. • The revision of pension policy so that the maximum pension would be 50 percent of previous wage: the lowest retirement age-limits would be increased; pension indices would be redefined so that they would be less linked to the increase in wages and more linked to the cost-of-living index. • The abandonment of all major tax deductions in family policy, cuts in home-care support, and more means testing. • The revision of government-housing policy (Arava).

• A new competition law would be enacted; various regulations and licenses aimed at limiting competition would be abandoned. • The investments in social housing by providing interest subsidies and implementing other measures would be put forward. • The financing structure of an earnings-related unemployment benefit would be revised so that the share of the insured, the state, and employers would be 5.5 percent, 47.5 percent, and 47 percent, respectively. • The earningsrelated share of unemployment would be 42 percent of remainder between the former wage and basic unemployment allowance. • Employees’ pension security contribution would be deductible in taxation. • Proposals aimed at revising the early retirement scheme would be cancelled.

• The government would cancel its agenda dated 6 November 1992. • Labor market organizations and the state would make a proposal aimed at raising the layoff threshold. • Implementation of waiting periods in unemployment insurance for persons entering the labor markets for the first time without vocational or professorial training. • The health insurance contribution would be raised, and wage earners’ unemployment social-security contribution would be introduced. • Cuts in the pensions for public-sector workers to make them comparable with those of the private sector. • Private pension systems would be further revised following the simultaneous agreement between the labor market organizations. • All further reforms would be negotiated with labor market organizations.

Double Transformation: How to Adjust Institutional Social Policy? | 71

Third, the social partners and government decided—after serious confrontation—to maintain and restore an existing Finnish model of interest mediation (known as an extended incomes policy model, or incomes policies with social packages), which protected earnings-related benefits over means-tested and flat-rate benefits. In 1992 the government made it clear that it would cut earnings-related benefits (in order to lower labor costs) if the social partners could not complete an incomes policy agreement with similar total economic and fiscal effects. The social partners did complete this agreement, and in response the government retargeted the cuts to the state and municipality-funded (means-tested and flat-rate) benefits. This moved reestablished “the conventional Finnish model,” which has not been seriously challenged since then (see third and fourth columns in Table 3.3). Fourth, the government systematically applied several forms of blameavoidance policies (see Weaver 1986). For instance, instead of cutting benefits, the government allowed inflation to do the job (for instance in child allowances), or, alternatively, it revised indexations so that minor changes at the time would result in large savings in the future. Finally, it prioritized services in kind over increases in cash benefits as services generated employment, whereas transfers resulted in additional social security contributions (see also fourth column in Table 3.3). Taken together, these findings show how existing interests and institutions filtered new ideas about social policy reform in particular ways.

The Post-recession Social Policy Many subsequent reforms came on the political agendas during the recession years of 1991 to 1993; however it would be misleading to argue that nothing of importance occurred subsequently. In fact the process of collective-belief formation (or collective/social (un)learning, see Hall 1993) has continued. In what follows, I introduce five key policy objectives emerging from the primeval soup: an emphasis on competitiveness, public economies, employment, market paradigms, and demography. Discussions of these factors have dominated Finnish social policy since the mid-1990s, and attention to them explains much of the political ability of Finnish governments to “curb” the increases in social expenditure during the late 1990s (see Figure 3.1, above; for a summary, see Table 3.4). Each of these objectives reflects certain policy paradigms and consequent agendas and frames. Together they comprise the post-recession framework of social policy making and its recalibration toward combining public costs and policy objectives (see Pierson 2001a: 425). Signifi-

72 | Juho Saari

cantly, none of these objectives is typically a sociopolitical objective (like socioeconomic equality); rather, they interpret social policy as a means to achieve other objectives that are considered more important in modernizing the Finnish model. Table 3.4. Emerging paradigms, agendas, and frames since the primeval soup Policy Objectives

As Applied to Social Policy Paradigm

Agenda

Competitiveness

Focus on research and development innovations instead of devaluations; avoiding additional social expenditures

Shifting costs from employers to employees; integrating social and health policy to national innovation systems (NIS)

Competitiveness and /productivity seen as mainstreamed policy objective.

Frame

Public finance

Commitment to stability and predictability

Curbing state social expenditure

Emphasizing “national urgency” and budget constraints

Employment

More emphasis on supply and quality of labor rather than on demand (as such)

Incentive reforms, the implementation of activation measures, lifelong learning, and (recently) flexicurity

Underlining disincentives as morally and economically devastating

Markets in social and health policies

Deregulation, partnerships, and “market neutrality” in social and health services

Implementing quasi-markets in social and health services; strengthening quasi-markets in earnings-related pensions

Efficiency, productivity, and cutting waste, while protecting the workers when possible

Demography

Proactive policies to deal with the challenges of aging society

Pension and social and health care reforms; policy programs for aging workers

Rapid aging; fairness between the generations; more efficient allocation of resources, cutting fat

The political and business elites (and many leaders of labor unions) accept competitiveness as a ruling paradigm. This has resulted in an agenda of shifting from employer social-security contributions to employee contributions and, more broadly, from social security contributions to taxfunded benefits. More recently, much attention has been devoted to the interplay between the welfare state and national innovation systems, indicating that some added value could be identified here. Finally, competi-

Double Transformation: How to Adjust Institutional Social Policy? | 73

tiveness is commonly framed as something that everyone benefits from greatly. National competitiveness is in the people’s common interest. The surplus of public economies is considered a key part of the socioeconomic paradigm. Here, the government underlines the need to generate surpluses in social security funds, maintain at least a balance in municipalities (that have a right to levy taxes), and keep the state budget permanently in surplus (however, the government allowed some public borrowing to cover public investments). In terms of agenda setting, this paradigm prioritizes earnings-related benefits and social and health services that social security and municipalities fund over the benefits and services that the state funds (mostly flat-rate and means-tested benefits). The frame emphasizes the need to keep things “under control” and underlines the role of responsible policy making. Employment has been a top priority since the recession. There has been a paradigm shift from the macro level and semi-Keynesian management of labor demand to a more micro-level emphasis on the supply of highquality labor. From a sociopolitical point of view, both micro and macro incentives to increase participation rate, especially among the young and aging, have dominated this agenda. Government documents are full of references to the reconciliation of private/family and working life, activation measures, lifelong learning, policies for the young entering first-time labor markets, and aging workers, active aging, etc. All key actors seem to accept this frame of thinking. The role of the market paradigm in social and health services has rapidly evolved since reforms in 1992 allowed quasi-markets or private-public-partnership (PPP) models. The key concept here is market neutrality: the equal treatment of service providers in various tenders regardless of their legal status (profit, nonprofit, private, third sector). A paradigm dominated by (this interpretation of) competition law has rapidly replaced one that underlined the importance of close cooperation between municipalities and third-sector organizations (with only a limited role for private service providers).2 The Finnish earnings-related pension system is a peculiar mix of private pension funds and public responsibilities (for details, see the section on EU policies below). Here, attempts to implement the neutrality principle while simultaneously protecting the vested interests of various organizations have dominated the agenda. The frame of markets is complex. On one hand, there are ideologically motivated claims that emphasize the better quality of private services, etc. (an argument that is actually not supported by any systematic evidence). On the other hand, many documents refer to higher productivity, efficiency, and cutting “fat.”

74 | Juho Saari

In demography, a new paradigm has evolved to address the rapid aging of the Finnish population during the 2010s, as the baby boomers of postwar generations retire. Gradually, policies aimed at discouraging labor supply have been replaced by policies aimed at lengthening work careers. The agenda has consisted of different measures from pension reform to the new models of rehabilitation and the programs for better working life. In health and social services, systematic attention has been given to a better match between the needs of patients and the different models of social and health services. Framewise, the economic sustainability of pensions and health care systems has played crucial role.

European Integration and the Finnish Welfare State Finland joined the EU in 1995. Geopolitical changes, namely the relative weakness of the then Soviet Union, allowed Finnish politicians to integrate more closely with the EU. Additionally, Finland had obvious economic interests in integration, as a significant proportion of its exports went to the EU. (At that time, Europeanization was exclusively considered an advanced expression of globalization.) This was, nevertheless, not as crucial as the previous agreements (1989) on European Economic Area, which allowed Finland’s industries to enter the EU on equal terms. Since then, Finland’s EU-policy paradigm has developed into a rather sketchy collection of general principles and broad objectives that demonstrate a limited amount of up- and down-loading (to use EU jargon) during the 1990s and the early 2000s. Rather than building its own European paradigm, Finland has until very recently adopted mainstream thinking designed by the Commission, France, and Germany. In addition, the European Monetary Union (EMU) and related Stability and Growth Pact (SGP) generate stability for currency and thereby promote economic growth and employment. The frame has strongly relied on the assumption that small member states like Finland benefit from the close cooperation with the (strong) Commission, which is able to control the great powers (Germany, France, Italy, the United Kingdom) of the EU. The policy documents of the 1990s are quite frank: the membership of the EU is considered as “mainly” positive as it shall have only a limited impact on the welfare states’ institutional structures, whereas it shall clearly strengthen its financial basis through some additional economic growth. The Ministry of Finance allowed little critical in-house scrutiny over these assumptions; in most cases European policies also fit quite well with emerging paradigms, agendas, and frames.

Double Transformation: How to Adjust Institutional Social Policy? | 75

Despite this stance, some actors expressed doubts over the impact of the EU on the welfare state. Much of the debate from 1992 to 1994 was dominated by ungrounded fears over the competencies of the EU to design Finnish social and health policies. Finland also had its own “Polish plumber”—a Portuguese fisherman with eight children, who would exploit the social security system and destroy social cohesion. European integration also had some important ramifications for the institutional characteristics of the Finnish welfare state (Leibfried and Wolf 2005). The EU demarcates between social insurance and noncontributory benefits (including social and medical assistance of different kinds and social services). The former is governed by the Social Security Coordination Regulation (1408/71), whereas the latter is left outside. The central issue for Finland was one of symmetry, with Finland trying to avoid a situation where it would pay a benefit to a country where such a benefit does not exist. In Finland there are several universal and residence-based flat-rate benefits that do not fall clearly under either social insurance or social assistance, including child allowance, national pension, basic unemployment allowance, and labor market support. Under the pressure to maintain the symmetry requirement, Finland changed the institutional structure of these benefits to accentuate either social insurance or social assistance and meet the spirit of the symmetry principle. The main issue was if—and if so, to what extent—these benefits were included in the scope of Regulation 1408/71 and whether they were exportable. For example, national pensions fell under the coordination regulation. In response, the government revised the entitlement rules of the national pension to tighten the link between the years of residence in Finland and the benefit levels. Child allowance is regarded as a family benefit and is also under the scope of Regulation 1408/71. (This contradicts the Finnish tradition since 1982, which considers child allowances to be a universal family benefit to be funded by the state and paid to the mother, rather than additional salary.) The basic unemployment daily allowance (flat-rate unemployment benefit) was classified as insurance, whereas labor market support (a means-tested flat-rate unemployment benefit) was defined to be social assistance and, consequently, outside the scope of the coordination regulation. The Finnish child–home care allowance was regarded as compensation for families that do not use public day-care facilities, putting it outside of Regulation 1408/71. However the European Court of Justice (ECJ) recently ruled (in case 333/00) that that the home care allowance should be regarded as a family benefit and thus should be included in the scope of the coordination regulation.

76 | Juho Saari

In addition to the coordination regulation, a number of definitional and conceptual problems have emerged over the public-private mix in certain employment-related benefits. The key issue is the extent to which insurance directives dictated by the four freedoms apply to Finnish private-sector employment-related pensions (TEL) or the statutory accidentinsurance scheme. Since these schemes are thoroughly legislated and mandatory but are run by private or semiprivate insurance companies, confusion emerged over how to classify them. On the one hand, TEL pensions are partly organized through private for-profit insurance companies, which could, in some cases, be interpreted as occupational pensions, so that the companies running TEL pensions 3 should be subject to certain insurance directives. On the other hand, according to the Finnish interpretation, the main objective of the TEL system is to safeguard social security, and, therefore, these pensions are considered to be outside the scope of these directives. When Finland applied for membership in the EU, it negotiated a special clause against demands set by life-insurance directives. In addition to direct social-policy impacts, there are indirect impacts with long-run consequences for social policy. Alcohol policy (ALKO) and the monopoly of the Slot Machine Association (RAY) are two such cases. In the Nordic countries (with the exception of Denmark) alcohol policy has been regulative, emphasizing health and other social policy aspects. The state has collected a substantial amount of tax revenue through its alcohol monopoly. In contrast, in the EU, alcohol policy belongs to the realm of agricultural policy. In its entry negotiations Finland negotiated the right to limit the amount of imported alcohol per traveler coming from abroad. The exception was extended to the end of 2003, but since 2004, both Finland and Sweden abolished the import restrictions. The question of the alcohol monopoly, whose position has not (yet) been challenged (by the Commission, although there have been cases like 189/95 in the European Court of Justice), remains important. Like alcohol, gambling has been a state monopoly (the government allows only one nonprofit gambling association, the RAY). From its profits the RAY has subsidized many third-sector organizations, which are heavily dependent on its funds. So far, ECJ rulings (e.g. case 124/97) have been mostly positive for the monopoly; however, the competition law prevented the RAY from subsidizing third-sector organizations in ways that would distort the conditions for competition among service providers (see the neutrality principle introduced above). Thus, although the EU has not taken direct measures against the RAY and the case law of the European Court of Justice remains conclusive, indirect measures may be significant in the field of social-service delivery for some specific groups of the population.

Double Transformation: How to Adjust Institutional Social Policy? | 77

A final case involves the EMU, which was passionately debated in Finland in the late 1990s. Socio-politically, there were concerns that the EMU convergence criteria would curtail public expenditure. Furthermore, detailed attention was devoted to so-called asymmetrical shocks, as Finnish business cycles do not necessarily follow those of European markets. As the European Central Bank (ECB) follows key European markets, it was feared that this asymmetry could result in an over- and under-heated Finnish economy. Furthermore, as an (external) devaluation option would no longer be available, nominal salaries and wages would have to adapt. In order to compensate this risk, so-called “buffer funds” were established in unemployment funds and pension funds as a measure to counterbalance such policies (see also Saari 2001; Hagfors and Saari 2006). The above discussion shows that Finland’s transition to the EU was largely smooth, as the policies and paradigms from the EU fit with those in Finland. Yet, where they did conflict, there was a process of mutual adjustment. As in the domestic sphere, this process followed the interplay of Finnish organizations and political exchange among key actors.

Conclusion Finland’s double transformation is an interesting extreme case study that sheds light on social policy–reform processes under rapidly evolving or punctuated circumstances. The Finnish welfare state was able to adjust, first, because its basic institutions were flexible enough to meet new demands and various economic and political pressures; second, because it was able to govern new social risks after the recession by differentiating them from old governance structures; and third, because the new social policy objectives were in line with the hegemonic policies on competitiveness, employment growth, and other policy arenas. This third point is the most interesting from a comparative perspective. Clearly, the early 1990s represent one of those primeval soups during which new social policy paradigms, agendas, and frames rapidly evolved. Interestingly, after only a few years of multidimensional debate and policies the main players agreed to a new set of policy paradigms, agendas, and frames. Since then, there has been a significant political consensus of an almost religious kind on key policy priorities in the field of social and health policies. New policy objectives came in packages with mutually strengthening complementarities. The self-reinforcing path dependencies of horizontal principles and governance structures played their roles, but so did idea-related paradigms, agendas, and policy frames.

78 | Juho Saari

When it comes to the double transformation, it seems that from the conventional social policy point of view emphasizing rights, equality, and justice, the boom was actually more harmful than the recession. The recession resulted from the failure of macro-economic policy, governance, and some external transformations, and there is an emerging consensus that the government’s policies were too strict during those years (see Jäntti, Saari, and Vartiainen 2006). Consequently, the social costs of the recession were perhaps unnecessarily high. However, the welfare state did adjust, and in most parts functioned quite well, in terms of equality and rights during the recession. During the boom years, however, the government’s approach to social policy became more instrumental, allowing more inequality and divisions within the governance structures. It is a matter of debate whether this development was, strictly speaking, necessary for economic and employment growth. In evaluating the politics of postrecession social policy, sufficient attention should be devoted to new governance structures located outside the conventional scope of social policy. Although social policy often stagnated within its conventional limits, outside of them, no fewer than three new governance structures emerged, indicating a strong ability to implement new policies even under rapidly changing circumstances. Finally, so far, the EU policies have had only limited impact on social and health policies, although they have strengthened some institutional cleavages between insurance and assistance and generated some defensive policies in pensions and the national monopolies (in alcohol and gambling). In most cases, the EU policies were well in line with the emerging consensus, so there was no tension between the EU guidelines and the national policies on any key issue. Needless to say, this situation may change, as the impact of recent European strategies aimed at promoting growth and jobs, eco-efficiency, “flexicurity,” and the stability of public economies become more visible. In particular, more attention should be devoted to the integrated guidelines of the EU that combine economic and employment policy objectives, as well as their role in designing new paradigms, agendas, and frames of the 2010s.

Notes 1. A number of factors contributed to the crisis: bad policies (the failure of monetary policy), bad institutions (that were not able to adjust because they were aimed at expansion), bad corporate and governmental governance (poorly allocated responsibilities), and bad luck (the collapse of the Soviet Union). Nobody who mattered saw it coming. All figures on the recession are derived from the Statistics Finland database.

Double Transformation: How to Adjust Institutional Social Policy? | 79

2. Private companies have always existed and played some role in certain specialized social and health services, and sickness insurance has long reimbursed these services to individual customers. However, since the 1990s, private companies have entered into mainstream services that were traditionally provided by municipalities and hospital districts.

Bibliography Belánd, Daniel. 2007. “Insecurity and Politics—A Framework.” Canadian Journal of Sociology 32, no. 3: 317–40. Blyth, Mark. 2002. Great Transformations—Economic Ideas and Institutional Change in the Twentieth Century. New York: Cambridge University Press. Crouch, Colin. 2005. Capitalist Diversity and Change—Recombinant Governance and Institutional Entrepreneurs. New York: Oxford University Press. ―――. 2007. “Neoinstitutionalism—Still No Intellectual Hegemony?” Regulation and Governance 1, no. 3: 261–70. Deeg, Richard. 2007. “Complementarity and Institutional Change in Capitalist Systems.” Journal of European Public Policy 14, no. 4: 611–30. Denzau, Arthur T., and Douglass C. North. 1994. “Shared Mental Models—Ideologies and Institutions.” Kyklos 47, no. 1: 3–31. Eggertsson, Thrain. 2005. Imperfect Institutions—Possibilities and Limits of Reform. Ann Arbor: University of Michigan Press. Hagfors, Robert, and Juho Saari. 2006. Social Policy in the Economic and Monetary Union—Social Expenditures and Public Indebtedness in 15 EU Countries. Social Security and Health Reports 72. Helsinki: Social Insurance Institution. Hall, Peter A. 1993. “Policy Paradigms, Social Learning, and the State—The Case of Economic Policymaking in Britain.” Comparative Politics 25, no. 3: 275–96. Jäntti, Markus, Juho Saari, and Juhana Vartiainen. 2006. Equity and Growth in Finland, UNU-WIDER Discussion Paper No. 2006/06. Helsinki: UNU-WIDER. http://www.wider.unu.edu/stc/repec/pdfs/rp2006/dp2006-06.pdf (accessed 11 November 2009) Kangas, Olli, and Joakim Palme. 2005. “Coming Late—Catching Up—the Formation of a Nordic Model.” In Social Policy and Economic Development in the Nordic Countries, edited by Olli Kangas and Joakim Palme, 17–59. Basingstoke: Palgrave. Kangas, Olli, and Juho Saari. 2008. “Krisenbewältigung mit Langseitfolgen? Der finnische Wohlfahrtsstaat.” In Europäische Wohlfahrtssysteme—Ein Handbuch, eds. Klaus Schubert, Ursula Bazant, and Simon Hegelich, 239–62. Wiesbaden: VS Verlag für Sozialwissenschaften. Kingdon, John W. 1995. Agendas, Alternatives, and Public Policies. 2nd ed. New York: Harper Collins. Korpi, Walter. 2003. “Welfare-State Regress in Western Europe—Politics, Institutions, Globalization and Europeanization.” Annual Review of Sociology 29: 589–609.

80 | Juho Saari

Leibfried, Stephan, and Dieter Wolf. 2005. “Europeanization and the Unraveling European Nation States—Dynamics and Feedback Effects.” European Foreign Affairs Review 10, no. 4: 479–99. North, Douglass C. 1990. Institutions, Institutional Change and Economic Performance. Cambridge: Cambridge University Press. Ostrom, Elinor. 2005. Understanding Institutional Diversity. Princeton: Princeton University Press. Pierson, Paul. 2001a. “Coping with Permanent Austerity—Welfare State Restructuring in Affluent Societies.” In The New Politics of the Welfare State, edited by Paul Pierson, 410–56. Oxford: Oxford University Press. ―――. 2001b. The New Politics of the Welfare State. Oxford: Oxford University Press. ―――. 2004. Politics in Time—History, Institutions, and Social Analysis. Princeton: Princeton University Press. Saari, Juho. 2001. Reforming Social Policy—A Study on Institutional Change During the 1990s. Helsinki: Publications of Social Policy Association 56. Saari, Juho. 2006. Suomen malli vertailevissa indekseissä. In Suomen malli – Murroksesta menestykseen?, edited by Juho Saari, 38-66. Helsinki: Helsinki University Press. Saari, Juho, and Olli Kangas. 2007. “Finland—Toward Pro-active Policies.” In The Europeanization of Social Protection, edited by Jon Kvist and Juho Saari, 178–202. Bristol: Policy Press. Starke, Peter. 2006. “The Politics of Welfare State Retrenchment—A Literature Review.” Social Policy and Administration 40, no. 1: 104–20. Weaver, R. Kent. 1986. “The Politics of Blame Avoidance.” Journal of Public Policy 6, no. 4: 371–98.

Chapter 4

THE SOCIAL INVESTMENT STATE A New Trend in Social Expenditure or Merely a Popular Political Discourse?

( Jorma Sipilä

Introduction The aim of the welfare state appears to have taken a complete U-turn. In the 1980s the welfare state was still largely understood as a political mediator that made the consequences of the market easier for citizens to bear. This function was neatly condensed in Esping-Andersen’s title Politics Against Markets (1985) and in the concept of “decommodification,” which was made famous by his epoch-making book, The Three Worlds of Welfare Capitalism (1990). However the principle of decommodification is also undeniably the reason why the welfare state has always been an object of economic and moral criticism. In the 1990s welfare state researchers became more enthusiastic about the importance of increasing labor force participation, especially among women, and some of them even ventured to regard “commodification” as an aim of social policies (e.g., Streeck 1999: 6). Increasing the commodification of labor, indeed, is a central function of the “social investment state,” a concept that rose to fame through Giddens’ book The Third Way (1998). In this book Giddens presented the juxtaposition of the old welfare state that sought to protect people from the market alongside the social investment state, which would make people stronger actors in the market. The idea of the social investment state has been presented as an antidote to the neoliberal critique, as well as to the threat of “welfare state retrenchment” (Jenson and Saint-Martin 2003). The shift from the decommodify-

82 | Jorma Sipilä

ing welfare state to the commodifying social investment state has become an integral part of the strategic political response of Western countries to the competitiveness of low-cost economies (Giddens 1998: 1), which even asserts that this change in the global order can be managed so as to produce social solidarity and prosperity. On the other hand, critics have pointed out that human beings are not born solely to swell the labor force and that social investment is not a panacea for all needs of social protection (e.g., Olk 2006). A shift in welfare state rhetoric1 has undeniably taken place, but this is only a first indicator of change. Have there also been changes in the objects of public financing? Do welfare states actually focus more on social investment now than in the past? The purpose of this chapter is to show that such a shift cannot be detected in social expenditure and that the rhetoric on social investment has far overstated the reality.

The Social Investment State The discourse on the social investment state arose in the 1990s. One of the pioneers was Esping-Andersen, who, in his numerous works, stressed the importance of social services in maximizing participation in the labor force. In 1996 he compared the Continental model—“welfare states without work and children”—to the Scandinavian model, which, “consists [of] shifting welfare state resources from passive income maintenance to employment and family promotion. The era of public employment growth has clearly ended and, instead, policy is directed to active labor market measures, such as training and mobility, and wage subsidies. . . . Nordic welfare states may be said to spearhead a ‘social investment’ strategy” (Esping-Andersen 1996: 25). “Social investment aims to move beyond redistributive, consumption-based social welfare centered around benefits and rights, to one that, through investment in human capital, enhances people’s capacity to participate” (Perkins, Nelms, and Smyth 2004: 3). Increasing social investment is the primary strategic response to modernizing social policy, which, following Palme (2006: 160) should deal with work, aging, and the new “gender balance.” The increased focus on social policy as an investment and a better integration of social and economic goals was already visible in Clinton’s and Blair’s (and, in Finland, Lipponen’s) programs and later clearly articulated in the Lisbon Strategy of the European Parliament (2000). The strategy aims to modernize the European social model by investing in people and building an active welfare state. This requires the integration of economic,

A New Trend in Social Expenditure or Merely a Popular Political Discourse? | 83

social, and employment policies, with the clear goal that these policies should be mutually reinforcing (European Commission 2004). The idea of using the means of the welfare state to promote economic growth by investing in the labor force was not, in fact, entirely new but had been presented in many variations throughout the twentieth century (Jenson and Saint-Martin 2003). Perhaps the most striking example came from the capital logic stream of Marxist theory, which insisted that the main function of social policy in capitalism is to reproduce the labor force (e.g., Lehtonen 1983). The discourse on the social investment state may also be conceived of as a rhetorical device meant to legitimize social policy in a new political setting. The aim of any public expenditure needs to be presented to the powerful decision makers using their own terms. At present, convincing the economic elite of the advantages of rational social policy may be the central aim for the social investment discourse. But what should the growing elderly population think about a new approach that aspires to leave them by the wayside?

Social Investment and Social Consumption: Desirable and Undesirable Expenditure Researchers and politicians speak of fundamental changes toward the social investment state, indeed of a new architecture of welfare (e.g., Olk 2006). Jenson and Saint-Martin (2003) have demonstrated convincingly that an international shift in policy objectives has already taken place. But are the governments in the OECD countries making serious attempts to implement the social investment state, or are they merely talking about what they should do? In order to determine whether the social investment state has been implemented or not, the concept must be operationalized. The imperative of operationalization forces us to consider what is actually meant by social investment—we should be able to distinguish social investment from social consumption. Investment is an economic term referring to results located in the future, whereas consumption is something that occurs in the present. “For state spending to be effective, and therefore worthwhile, it must not simply be consumed in the present, to meet current needs, but it must be an investment that will pay off and reap rewards in the future” (Jenson and Saint-Martin 2003: 83). A social investment state, therefore, spends public money to derive direct or indirect economic benefits. How do we know when expenditure will produce direct or indirect economic benefits in the future and when it is just a short-term consump-

84 | Jorma Sipilä

tion? What kind of social expenditure is desirable in a social investment state and what kind is not? Distinguishing social investment from social consumption is not simple, but in most cases researchers tend to mention two particular categories of expenditure as examples of social investment: investments that increase human capital and investments that increase labor market participation. Olk makes a basic distinction between building human or social capital by investing in individuals and protecting “at risk” groups in a reactive manner (such as granting social security benefits): “In this drive to provide human capital investment through, for example, education and training, there is a marked orientation to the future with enhanced opportunities for children,” says Olk (2006: 1). Indeed, increasing the supply of labor force with the support of welfare state services (activation, child care, eldercare, employment services) is very prominent in the present political discourse (e.g., European Commission 2004; Palme 2006). Evidently, the social investment state should promote children, students, working mothers, immigrants, and people who return from unemployment, sick leave, or maternity leave to the labor force. It is more difficult to say what sort of social expenditure should be disfavored as plain consumption. Spending on passive welfare such as unemployment benefits is generally seen as “bad” (Perkins, Nelms, Smyth 2004: 9). At the top of the list we might find the unemployment benefits aiming at early retirement (de la Porte 1999: 3), particularly if they are earnings related. Another example of dubious public consumption consists of allowances paid for individual care provided by family members, because those allowances prevent women from taking employment (OECD 2007a, 2007b). More generally, the social investment state does not encourage people to leave employment nor does it invest in those who have definitely left working life. Thus, the social investment approach means, at the very least, implicit disadvantages for the elderly (Jenson and Saint-Martin 2003: 93). A good example of the new criticism against social consumption is Kvist’s comment in which he calls into question the habit of assuming that a person exceeding a certain age always needs public support (2006: 137). For the most part, it is difficult to find social consumption that does not have the character of investment whatsoever. A good example of the difficulty is the case of pensions. On the one hand, the pension is an apparent example of social consumption, but on the other hand, receiving an earnings-related pension requires earlier participation in the labor force. Thus, the existence of earnings-related pensions does not fully contradict the idea of social investment.

A New Trend in Social Expenditure or Merely a Popular Political Discourse? | 85

To understand the role of the social investment state it is important to notice that the state is not the only entity that invests in human capital; families also make such investments. Societies with a strong, liberal political culture like the United States generally emphasize the responsibility of families to invest in education and are less likely to listen to international recommendations concerning social investment politics. Investment by families provides human capital, but it must be called private, not social, investment. Compared to social investment, the private investment in education generally leads to a less even distribution of human capital. The concept of a social investment state refers to the government allocating resources of its own. The term “social investment” accurately describes those benefits that make long-term effects on a society. The social character of such benefits is reflected in the fact that they are typically universally provided in order to patch up both market and individual failures. As such, they are typically financed by taxes. If the shift toward the social investment state has really happened, it should already be noticeable in changes in social expenditure and detectable in developed welfare states. In this chapter I shall examine if the desired expenditure has actually increased at the expense of the undesired expenditure. Another task is to find cases in which the desired changes of expenditure have occurred: what about the expected consequences—are they already discernible?

Data and Methods To answer these questions I analyze social expenditures in twenty-four countries between 1998 and 2003, and partly since 1980 or 1991.2 The monitoring period begins in 1998, because at that time the governments of developed welfare states were well aware of the needs for change caused by the opening of the international economy and an aging population. An abundant body of literature on both aging and globalization was available in 1998.3 Although the concept of the social investment state was broadly used only after 1998,4 the necessities of creating a qualified labor force and of using women’s competencies in the labor market were definitely known much earlier. The government program in Finland, for instance, included a deliberate social investment program in 1995. To access a variety of countries and policies I use data from twentyfour OECD countries. They include fifteen older EU member states, four new EU member states, plus Norway, Switzerland, Canada, Japan, and the United States. Thus the European OECD countries5 and the three most important non-European member countries are included in this study.

86 | Jorma Sipilä

Following the rhetoric of the social investment approach, I selected expenditures on education and on family to represent categories of social investment. In contrast, expenditure on old age is an example of social consumption not investment. A problem with this crude distinction is that family policies involve measures that both support women’s participation in the labor force (child care, earnings-related maternity benefits), which can be regarded as social investment, but also support their staying out of the labor market (flat-rate cash subsidies), which is a form of social consumption (Ferrarini 2006). I elaborate on this issue later when I discuss family policy in detail. Not every increase in expenditure is a sign of a policy change. If, for example, the child population grows and a government therefore spends more on family allowances than before, it does not mean that the average child gets more support. A social investment approach means that each child enjoys higher expenditure than before. Therefore the expenditure on families and the elderly must be standardized to take into account the differences in age-related needs. Family expenditure in each country is standardized by the proportion of children aged 0 to 14, and old-age expenditure by the proportion of elderly aged 65 plus. All calculations concerning the expenditure on families and old age are based on these standardized values. The expenditure on education, however, is not standardized (e.g., according to the proportion of young people or the number of students), because social investment in education clearly means that there will be more students in society. There are some problems in the data that cannot be solved. As is well known, the SOCX data is based on gross expenditure, which means that sometimes the benefits include taxes, and sometimes not. Researchers would prefer net expenditure, but only some general net figures have become available so far (see Adema and Ladaique 2005). Many comparative studies on welfare state development compare social expenditure as a proportion of GDP. This has raised much discussion because the measure is affected by the development of GDP. However, this problem is of minor significance in the present study, because I compare three kinds of expenditure to each other separately in each country. A change in GDP affects all kinds of expenditure in the same way. Of course, the possibility remains that an economic boom might raise spending, for example on family services more than on pensions. If research is interested in the outcomes of the expenditure there is also a “time lag” problem to consider. The expenditure has a gradual rather than immediate effect (Green-Pedersen 2007: 19–20). The length of the time lag is a contextual issue that cannot be solved theoretically, but only empirically in each case.

A New Trend in Social Expenditure or Merely a Popular Political Discourse? | 87

Two competing hypotheses can be presented: first, we may believe that the social investment state is on the rise and that since 1998 the governments have increased their expenditure on education and families more than their expenditure on old age. The contrasting null hypothesis is that in spite of new rhetoric, the phenomenon of path dependency is strong, and old structures are still allowed to dominate—governments have not increased their expenditure on education and families more than they have on their expenditure on old age. To test these contrasting hypotheses I created a “social investment index” that describes the differences between the variables measuring social investment and social consumption. Public social expenditure on families and education are added together, and then the expenditure on old age is subtracted: Social investment index = (PSEfamilies + PSEeducation−PSEold age) My second hypothesis is that since 1998 the governments have redirected their expenditure on families to support female employment and fertility. In 2003 they spent more on child care and less on family allowances than in 1998. The contrasting null hypothesis in this case is that no such change has happened; the rhetoric on social investment has not transformed family policies. These hypotheses direct the attention toward outcomes, and a few bivariate analyses are conducted to reveal relationships for preliminary inspection.

Results Increasing Public Expenditure on Families and Education, Decreasing Public Expenditure on Old Age? Examining the development of social expenditure for the entire period of 1980 to 2003, we see that the age-standardized expenditure on old age has grown faster than the expenditure on families. This happened mainly because the expenditure on families was decreasing between 1991 and 1998.6 There is no comparable data on education for 1980, but after 1991 the expenditure on education did not change much. Table 4.1 shows that there has been an increase in family, education, and old-age expenditure during the period of 1998 to 2003. The expenditure on families increased faster than the others. Figure 4.1 clearly shows the relative growth of family expenditure compared to old-age expenditure.

88 | Jorma Sipilä

Table 4.1. Changes in social expenditure as percent of GDP, standardized by size of population age 0 to 14 and 65 plus, 1980–2003 Year

Expenditure on families

Expenditure on education

Expenditure on old age

1980 1991

1.7

..

5.9

2.1

5.2

6.9

1998

2.0

5.1

7.6

2003

2.2

5.3

7.9

Years

Expenditure on families, percent change

Expenditure on education, percent change

Expenditure on old age, percent change

1980–1991

22

..

18

1991–1998

−5

−2

10

1998–2003

9

4

4

1980–2003

27

..

35

Figure 4.1. Public expenditure on families and old age in twenty-four OECD countries, as percent of GDP, 1998–2003. Created by the author. 8

7

6

old age old age/families

5

families

4

3

2

1

0 1998

1999

2000

2001

2002

2003

A New Trend in Social Expenditure or Merely a Popular Political Discourse? | 89

When we compare the period from 1991 to 1998 to the period from 1998 to 2003 we find that the expenditures on families and education increased faster in the latter period, whereas the expenditure on old age increased more slowly. A most interesting question is whether these trends reflect deliberate policies or whether they are a consequence of demographic changes.7 The suggestion that these trends reflect the influence of demographic changes gains some support from the statistically significant negative correlation (−0.49*) between the proportion of children in the population and the expenditure on families while the analogous correlation between the proportion of elderly people in the population and the expenditure on old age was not significant. Thus the expenditure trends may indicate the existence of a conscious social investment approach.

Differences in Expenditure on Families and Education Now I conduct an overview of the data for all twenty-four countries. Table 4.2 shows expenditure on family support, education, and old-age pensions between 1991 and 2003, highlighting also the relative change between 1998 and 2003 in these indices. Table 4.2. Social expenditures on families, education, and old age, 1991, 1998, and 2003 Public expenditure on families (standardized)1

Public expenditure on education as percent GDP2

Public expenditure on old age as percent GDP (standardized)3

1991

1998

2003

Change 2003/1998

1991

1998

2003

Change 2003/1998

1991

1998

2003

Change 2003/1998

Scandinavia

4.5

3.6

3.2

0.90

6.9

7.4

7.4

1.00

6.9

7.7

8.0

1.05

Central Europe

2.5

2.4

2.5

1.05

5.0

5.1

5.2

1.04

8.0

8.4

8.2

0.97

Southern Europe

0.8

1.2

1.5

1.28

3.5

4.4

4.5

1.03

7.4

8.4

8.6

1.02

Region

Eastern Europe

2.4

1.6

2.3

1.44

5.6

4.4

5.1

1.14

7.2

8.7

9.6

1.10

United States

0.6

0.7

0.6

0.86

5.1

5.0

5.6

1.11

5.9

6.1

6.7

1.10

Canada

0.6

0.9

1.0

1.14

6.5

5.4

4.7

0.87

5.0

5.0

4.8

0.95

Japan

0.5

0.7

0.9

1.33

..

3.6

3.7

1.04

4.7

5.4

6.4

1.20

United Kingdom

2.4

2.5

2.7

1.11

4.8

4.6

5.1

1.10

4.6

5.3

5.6

1.07

Austria

3.2

2.8

3.2

1.14

5.3

6.2

5.5

0.89

10.4

11.8

12.7

1.08

Belgium

2.7

2.9

2.7

0.93

5.0

5.1

6.0

1.19

6.2

6.3

6.4

1.03

Canada

0.6

0.9

1.0

1.14

6.5

5.4

4.7

0.87

5.0

5.0

4.8

0.95

Country

90 | Jorma Sipilä

Public expenditure on families (standardized)1 1991

1998

2003

Change 2003/1998

Czech Republic

2.8

2.2

2.1

Denmark

4.3

3.9

3.6

Finland

4.3

3.3

France

2.7

Germany Greece

Public expenditure on education as percent GDP2

Public expenditure on old age as percent GDP (standardized)3

1991

1998

2003

Change 2003/1998

1991

1998

2003

Change 2003/1998

0.97

..

3.9

4.5

1.17

6.2

8.2

8.5

1.05

0.94

6.9

8.3

8.3

1.01

6.5

7.5

7.4

0.99

2.9

0.88

6.5

6.2

6.4

1.04

8.3

8.1

8.5

1.06

3.1

2.8

0.88

5.5

5.8

5.9

1.01

9.3

10.1

9.8

0.97

2.6

2.1

2.2

1.04

..

4.6

4.7

1.03

8.7

9.9

9.9

1.00

1.0

1.3

1.5

1.17

2.3

3.1

3.6

1.15

9.8

10.5

10.0

0.95

..

3.6

3.8

1.06

6.1

4.4

5.9

1.32

..

7.1

7.5

1.05

Country

Hungary Ireland

1.3

1.2

2.1

1.69

5

4.4

4.4

1.00

4.0

2.9

3.9

1.36

Italy

0.8

1.4

1.5

1.12

3

4.8

4.7

0.99

7.7

9.5

9.2

0.97

Japan

0.5

0.7

0.9

1.33

..

3.6

3.7

1.04

4.7

5.4

6.4

1.20

Luxembourg

2.8

3.0

3.8

1.27

3

3.7

3.5

0.96

9.3

8.6

6.2

0.72

Netherlands

1.8

1.5

1.5

1.03

5.6

4.8

5.1

1.06

6.4

6.0

6.0

1.00

Norway

3.6

3.8

3.0

0.79

7.1

7.5

7.5

1.00

6.0

6.9

7.3

1.06

Poland

2.0

0.8

1.5

1.91

5.2

5.0

5.6

1.12

8.8

9.4

13.7

1.46

Portugal

0.9

1.1

1.7

1.62

4.6

5.4

5.6

1.0

4.74

6.3

8.1

1.29

Slovak Republic

..

1.8

1.8

1.00

5.6

4.5

4.3

0.97

..

8.6

8.5

1.00

Spain

0.3

1.0

1.2

1.27

4.1

4.4

4.3

0.97

7.3

7.5

7.1

0.95

Sweden

5.8

3.5

3.4

0.98

7.1

7.6

7.3

0.97

6.8

8.3

9.0

1.09

Switzerland

1.3

1.5

1.5

1.03

5.3

5.3

6.0

1.13

5.3

6.5

6.6

1.01

United Kingdom

2.4

2.5

2.7

1.11

4.8

4.6

5.1

1.10

4.6

5.3

5.6

1.07

United States

0.6

0.7

0.6

0.86

5.1

5.0

5.6

1.11

5.9

6.1

6.7

1.10

Country Mean

2.1

2.0

2.2

1.09

5.2

5.1

5.3

1.04

6.9

7.6

7.9

1.05

1. OECD social expenditure, aggregated data. Data for Hungary 1999 instead of 1998. The Czech Republic Ministry of Labour and Social Affairs has provided the data for 1991 when the Republic still was a part of Czechoslovakia. 2. OECD online education database. Data for Belgium from 1999 instead of 1998. Data for 1991 from UNESCO. 3. OECD social expenditure, aggregate data. Data for Finland 1998 and 2003 and Luxembourg 2003 from Eurostat. Data for Hungary 1999 instead of 1998.

Beginning with family policy, Scandinavian countries have been at the top of the family expenditure list since 1991, but the old image of Scandinavia as the only group of European countries in which social expenditure trends favor the young over the old (Esping-Andersen 1996: 14) has faded.

A New Trend in Social Expenditure or Merely a Popular Political Discourse? | 91

Family expenditure per child as a percentage of GDP actually decreased in all Scandinavian countries between 1998 and 2003. In terms of levels, in 2003, Hungary, Luxembourg, Denmark, and Sweden were heavy investors in family policy. The non-European countries were all at the other end of the scale. The United States was exceptional because of the absence of family allowances, but neither Japan nor Canada spent noteworthy amounts of public money on family policy. In terms of public expenditure on education, the increase in expenditure was very slight (4 percent) during the observation period of 1998 to 2003. Nevertheless, even this was more than the decline from 1991 to 1998, meaning that there was an overall gradual rise in spending. It is worth mentioning that the slow growth of public expenditure on education was not compensated by private expenditure. The latter actually diminished in the OECD from 1995 to 2003 (OECD Factbook 2007). Table 4.3. Changes in social investment and family expenditure, 1998–20038 Social investment index

Family expenditure in cash/in kind

Change 1998

Change

2003

2003/1998

1998

2003

2003/1998

Region Scandinavia

3.4

2.6

-0.8

-0.1

-0.1

0.0

Central Europe

-1.1

-0.4

0.6

0.8

0.9

0.1

Southern Europe

-2.8

-2.6

0.2

0.1

0.0

-0.1

Eastern Europe

-1.7

-2.2

-0.5

1.0

0.6

-0.4

United States

-0.5

-0.7

-0.2

-0.5

-0.5

0.0

Canada

1.7

0.9

-0.8

0.5

0.7

0.2

Japan

-1.1

-1.8

-0.7

-0.1

-0.1

0.0

United Kingdom

2.0

2.5

0.5

0.8

1.4

0.6

Austria

-3.2

-4.0

-0.8

1.7

1.9

0.3

Belgium

1.7

2.3

0.6

1.2

0.8

-0.4

Canada

1.7

0.9

-0.8

0.5

0.7

0.2

Czech Republic

-2.0

-1.9

0.1

0.9

0.7

-0.3

Denmark

4.7

4.6

-0.1

-0.7

-0.7

0.0

Finland

1.5

0.8

-0.7

0.5

0.2

-0.3

Country

France

-1.0

-1.2

-0.2

-0.3

-0.1

0.1

Germany

-3.6

-2.9

0.7

0.3

0.4

0.1

Greece

-5.8

-5.0

0.8

0.4

0.4

0.0

Hungary

1.1

2.2

1.1

0.7

0.6

-0.1

92 | Jorma Sipilä

Social investment index

Family expenditure in cash/in kind

Change 1998

2003

2003/1998

Change 1998

2003

2003/1998

Country Ireland

3.2

2.5

-0.6

1.3

2.0

0.7

Italy

-3.5

-2.9

0.5

-0.1

-0.1

0.0

Japan

-1.1

-1.8

-0.7

-0.1

-0.1

0.0

Luxembourg

-1.5

1.4

2.9

2.0

3.0

1.0

Netherlands

0.3

0.7

0.4

0.1

-0.1

-0.2

Norway

4.5

3.2

-1.3

0.3

0.4

0.1

Poland

-3.6

-6.7

-3.0

0.4

0.5

0.1

Portugal

0.1

-0.8

-0.9

0.3

-0.1

-0.4

Slovak Republic

-2.2

-2.4

-0.2

1.9

0.8

-1.2

Spain

-2.1

-1.6

0.5

-0.3

-0.3

0.0

Sweden

2.9

1.7

-1.2

-0.4

-0.4

0.1

Switzerland

-0.1

0.9

0.9

0.8

0.7

-0.1

United Kingdom

2.0

2.5

0.5

0.8

1.4

0.6

United States

-0.5

-0.7

-0.2

-0.5

-0.5

0.0

Country mean

-0.3

-0.4

-0.1

0.5

0.5

0.0

In terms of changes in education spending, the United States, the United Kingdom, and most Eastern Europe countries substantially increased their expenditure from 1998 to 2003, whereas the public financing of education was notably reduced in Canada.9 In terms of the level of spending, Scandinavian countries were heavy investors in education in 2003. Among individual countries, Denmark, Norway, and Sweden were far ahead of the others, while Luxembourg, Greece, and Japan did not invest much public money in education. Hungary and Switzerland showed a remarkable increase in expenditure in the period from 1998 to 2003, moving from average to leading levels.

Differences in Expenditure on Old Age Age-standardized expenditure on old age increased in most states between 1998 and 2003.10 Traditionally, expenditure on old age has been highest in Central Europe, and indeed it still was in 1991, but after that, the highest percentages are to be found in Eastern and Southern Europe. From 1998 to 2003 the increase of expenditure was very great in Japan,

A New Trend in Social Expenditure or Merely a Popular Political Discourse? | 93

Eastern European countries and the United States, while there was a decrease in Canada and Central European states. Compared to the period from 1991 to 1998 the increase in expenditure in the later period was generally smaller or actually negative. Among individual European countries Austria has always had very high expenditure on old age, but the upswing of Polish expenditure from 1998 to 2003 was beyond comparison and brought Poland to the top. At the lower end, Ireland and Canada represent extreme cases. Seven out of the twenty-four countries reduced their relative expenditure on old age between 1998 and 2003. This is not a particularly high proportion, as about the same number of countries also reduced their expenditure on families (eight countries) and education (seven countries).

Which States Should Be Called Social Investment States? Giddens has presented the Nordic states as successful models of the social investment state (2006: 11–15), whereas France, Germany, and Italy are his prime examples of “blocked societies” (ibid.: 32–40). The tenability of this statement can be tested with the social investment index, which measures how much countries emphasize social investment as opposed to social consumption. Table 4.3 presents the results. It is no surprise that in both 1998 and 2003 we find Denmark and Norway at the top, followed by Ireland, the United Kingdom, Belgium, and Sweden. The greatest change, however, is seen in the Central European states, which seem to have taken a leap toward the social investment state. Southern European countries show a small increase, whereas the indexes for Japan, Canada, and particularly Poland are falling, revealing a trend away from social investment. High index numbers can be achieved both by investing in families and education and by saving on old-age benefits. Denmark and Norway represent the high-investment model, but the two other countries on the list, Ireland and the United Kingdom, exemplify the low-consumption model. High index figures apply to all of them. Countries not inspired by social investment are found at the other end. Negative index numbers mean that they have spent more on old age than on families and education together. The extreme cases are Poland, Greece, and Austria, followed by Germany and Italy. In view of the conception of Scandinavia as the model for social investment, the finding that the most negative index change took place in Scandinavia raises questions. If social investment is the recipe for success, why are Norway, Sweden, and Finland not upholding this emphasis? The change in Norway and Sweden is clearly negative, and Denmark alone keeps the Scandinavian flag high. For some reason, Denmark did not

94 | Jorma Sipilä

increase old age expenditure in the period from 1998 to 2003, whereas the others did. Another important finding is that Giddens’ judgment of blocked countries is at least partly out of date: Germany and Italy present clearly positive changes in the index. Generally, if there is a trend in social investment, it is, in fact, a decrease. From 1998 to 2003 the country mean went down by 0.1 percent, and thirteen countries out of twentyfour show a negative change in the index. These findings support the null hypothesis.

Expenditure on Families in Cash and In Kind My second hypothesis was that in order to increase female employment governments would increase their spending in child care at the expense of family allowances. This cannot be counted with the present OECD data because there are no reliable time series on child-care expenditure. However, an approximation can be made by inspecting the expenditure on families in two parts: cash and kind. Cash expenditure mainly includes transfers, which do not support female employment, whereas the in-kind benefits are mostly services that facilitate women’s entry into the labor market. On average, the twenty-four countries favor cash expenditure in their family policies. Both in 1998 and in 2003 the cash expenditure was 0.5 percent units of GDP higher than the expenditure in kind (see Table 4.3). Cash benefits were relatively high in the UK and Central Europe, while the US, Denmark, and Sweden preferred benefits in kind. There were only eight countries in which benefits in kind showed an increase when related to GDP, while the opposite was true in ten countries. In six countries there was no change. Thus, even in this case, the null hypothesis was upheld: governments did not reallocate their family expenditure to support female employment and fertility.

Have the Social Investments Led to the Desired Outcomes? In this section I try to ascertain to what extent public expenditure on families, education, and old age is related to female employment and fertility, both of which were mentioned earlier in this chapter as the major goals of social investment policies. The data is small, and I only use Pearson’s correlation coefficients to describe relationships. Such an exploration only gives indicative support for the associations found. In the following statistical calculations I have two dependent variables: • share of women aged 15 to 64 in employment; • children born to women aged 15 to 49.

A New Trend in Social Expenditure or Merely a Popular Political Discourse? | 95

In addition to the four variables describing public expenditure I bring in two other independent variables to describe the volume of subsidized child care and early education based on the definitions and the data presented in the OECD Family Database (Table 4.4): • the percentage of children aged 3 to 5 in early education programs; • the percentage of children aged 0 to 2 in child care. Table 4.4. Child care, fertility, and female employment Children in child care (percent)1 Ages 3-5

Ages 0-2

Children born to women aged 15 to 492 1998

2003

Change 2003/1998

Female employment participation 1998

2003

Change 2003/1998

Region Scandinavia

77

45

1.68

1.76

1.04

68.7

70.4

1.03

Central Europe

78

20

1.54

1.57

1.02

55.5

59.3

1.07

Southern Europe

81

14

1.28

1.32

1.04

43.1

48.7

1.13

Eastern Europe

70

7

1.31

1.21

0.93

52.9

51.4

0.97

United States

62

36

2.00

2.04

1.02

67.4

65.7

0.97

Canada

..

19

1.54

1.53

0.99

63.5

67.9

1.07

Japan

86

15

1.38

1.29

0.93

57.2

56.8

0.99

United Kingdom

81

26

1.71

1.71

1.00

64.2

66.4

1.03

Austria

7

7

1.34

1.38

1.03

58.5

61.5

1.05

Belgium

100

34

1.53

1.64

1.07

47.5

51.4

1.08

Canada

..

19

1.54

1.53

0.99

63.5

67.9

1.07

Czech Republic

85

3

1.16

1.18

1.02

58.7

56.3

0.96

Denmark

90

62

1.72

1.76

1.02

70.3

70.5

1.00

Finland

46

35

1.70

1.76

1.04

61.3

65.7

1.07

France

102

28

1.76

1.88

1.07

52.4

56.4

1.08

Germany

80

9

1.36

1.34

0.99

56.3

58.7

1.04

Greece

47

7

1.29

1.28

0.99

40.3

44.5

1.10

Hungary

87

7

1.33

1.27

0.95

47.3

50.9

1.08

Ireland

68

15

1.93

1.96

1.02

48.2

55.4

1.15

Italy

100

6

1.20

1.26

1.05

37.3

42.7

1.14

Japan

86

15

1.38

1.29

0.93

57.2

56.8

0.99

Luxembourg

72

14

1.68

1.63

0.97

45.6

52.0

1.14

Netherlands

70

30

1.63

1.75

1.07

59.4

64.2

1.08

Country

96 | Jorma Sipilä

Children in child care (percent)1

Children born to women aged 15 to 492

Female employment participation

Ages 3-5

Ages 0-2

1998

2003

Change 2003/1998

1998

2003

Change 2003/1998

Norway

85

44

1.81

1.80

0.99

73.6

72.7

0.99

Poland

36

2

1.44

1.22

0.85

52.2

46.2

0.89

Portugal

78

24

1.46

1.44

0.99

58.3

60.6

1.04

Slovak Republic

72

18

1.38

1.20

0.87

53.5

52.2

0.98

Spain

99

21

1.16

1.31

1.13

36.5

46.8

1.28

Sweden

87

40

1.50

1.71

1.14

69.4

72.8

1.05

Switzerland

45

..

1.47

1.39

0.95

68.8

70.7

1.03

Country

United Kingdom

81

26

1.71

1.71

1.00

64.2

66.4

1.03

United States

62

36

2.00

2.04

1.02

67.4

65.7

0.97

Country Mean

76

22

1.52

1.53

1.01

56.2

58.7

1.05

1. Source: OECD Family Database and OECD Education Database. Data for Canada, Germany and Poland from 2001; France, 2002; Greece, Luxembourg, Norway, and the Slovak Republic, 2003; and Denmark and the United States, 2005. 2. Source: OECD Factbook 2007: Economic, Environmental and Social Statistics

Table 4.4 also shows changes in the female employment and fertility rates. The levels of female employment were still quite different in 2003, as in Sweden, Norway, and Denmark together with Switzerland the levels exceeded 70 percent, but Italy, Greece, Spain and Poland were below 50 percent. However, the average rise of 2.5 percentage points has been amply exceeded by most Southern and Central European countries.11 Fertility differences are also large, and indeed they increased during the observation period. The extraordinarily high American rate went even higher, and the low rates in Japan and Eastern Europe became still lower. The correlations in Table 4.5 indicate statistical associations between social investments and their anticipated consequences. Expenditure on education and the percentage of children in child care have very high positive correlations with female employment. Education also correlates with fertility, whereas early education is not connected to the dependent variables. Expenditure on old age shows negative correlations to both employment and fertility, but only the latter is statistically significant. The social investment index puts three variables together and displays strong correlations with the female employment rate and higher correlations than the other independent variables with fertility.

A New Trend in Social Expenditure or Merely a Popular Political Discourse? | 97

Table 4.5. Correlations between public expenditure, female employment, and fertility, 1998 Public expenditure on education

Public expenditure on families

Public expenditure on old age

Social investment index

Female employment rate

0.67

0.34

-0.33

0.69

Fertility rate

0.51

0.29

-0.42

0.66

Table 4.6. Correlations between public expenditure, female employment, and fertility, 2003 Public expenditure on education

Public expenditure on families

Public expenditure on old age

Social investment index

Children aged 3–5 in preschool

Children aged 0–2 in child care

Female employment rate

0.63

0.20

-0.30

0.59

-0.07

0.70

Fertility rate

0.44

0.21

-0.45

0.63

0.05

0.22

Female employment rate

0.55

Table 4.7. Correlations between public expenditure, female employment, and fertility, 1998–2003 Change of public expenditure on education

Change of public expenditure on families

Change of public expenditure on old age

Change in social investment index

Change of female employment rate

-0.24

0.03

-0.43

0.53

Change of fertility rate

-0.21

-0.39

-0.31

0.18

Change of female employment rate

0.55

All the correlations are higher when there is a five-year interval between the figures of independent and dependent variables; the expenditure in 1998 correlates more strongly with employment and fertility in 2003 than the expenditure in 2003. Looking at the variables measuring changes from 1998 to 2003 in the lower part of Table 4.5, the correlations seem to be low and even surprising. However, two results make sense: First, a positive change in the social investment index was connected to increasing female employment. Second, spending less on old age seems to be connected to an increase in female employment.

98 | Jorma Sipilä

As Canada and the United States have been weak family-policy countries throughout the period examined, social expenditure on families cannot be a sufficient cause for the growth of women’s participation in the labor market. However, we also note that Canada and the United States have been heavy investors in education,12 just like Scandinavia. In addition, Scandinavia, the United States, and the United Kingdom have higher-than-average rates of child care among children aged 0 to 2. All this supports the argument that investment in child care and education particularly increase female labor participation. This result does not require social investment but can also be achieved through private investment. The fertility rate is a notoriously complicated phenomenon to explain (Caldwell and Schindlmayr 2003), but at least it can be stated that the level of expenditure on education and the social investment index vary together with the level of fertility, whereas it may not be a surprise that the expenditure on old age is related to reduced fertility.

Discussion Has there been a turn from decommodification toward commodification and from social consumption toward social investment? In the light of changes in social expenditure it would be risky to use such strong expressions. Certainly there is one change that speaks of a growing emphasis on social investment, as expenditure on families grew faster than that on old age. Yet the increase in family expenditure per child was partly achieved without increasing the percentage of GDP spent on families because the number of children was diminishing. A sign of a reverse development is that the data reveals only a small increase in education expenditure, and, because of this, the social investment index actually decreased during the observation period. In addition, the governments did not change the content of their family policies to support female employment. Old age expenditure increased apace with expenditure on education but more slowly than before 1998. Perhaps the old age expenditure increased, as the pension systems in many countries were not fully mature in 1998, i.e., there were more women receiving earnings-related pensions. We could also discuss the possibility of a cohort effect (Chauvel 2003), which means that generations that entered the labor market in times of positive economic growth are retiring. This raises the expenditure on old age temporarily but not necessarily for long. Such reasons may still exceed the retarding influences of pension reforms and of the long-term growth of unemployment. From the political point of view we could also ask whether it is reasonable to expect a real decrease in public old-age

A New Trend in Social Expenditure or Merely a Popular Political Discourse? | 99

expenditure as long as the main age of the constituency continues to rise rapidly. Reading results with caution is sound advice. First, the social expenditure data is not optimal material to describe the growing importance of the social investment approach. Pension reforms, for instance, reflect recent political changes much more rapidly than changes in expenditure. Second, the shortness of the study period (1998–2003) is a limitation that inhibits firm conclusions. Third, the sweeping changes in social expenditure in the early 1990s caused by the political upheavals in Eastern Europe and the recession may have influenced the findings, because in some countries they have increased the likelihood of backlashes in public expenditure. Fourth, in spite of all these reservations, in a period of diminishing interest rates (2000–2005), it was rather easy to increase social investment if desired, even for debt-ridden governments. Social research has often underlined the path dependency and persistence of national social policies (Pierson 1996; Gauthier 2002). The Scandinavian countries together with France, Belgium, and Austria, at the higher end, and the United States, Japan, and Canada at the lower end, are good examples of continuity in family expenditure. Social policies—particularly policies toward children and the elderly—are deep-seated in cultural values, which limits the acceptable alternatives; however, we also find remarkable digressions from the path. Family expenditure has increased considerably, for example, in Portugal and Spain. Norway is an example of a country that has recently adopted an “explicit and comprehensive family policy,”13 while Luxembourg, in spite of increasing expenditure, has continued along the transfer path and has not invested in family services. In a similar way, in the field of old age there are both high-consumption (Austria, Germany, France, and Italy) and low-consumption countries (Canada, the UK, and Japan). However, Portugal and Poland have experienced a large relative increase in the share of the old age expenditure, whereas Luxembourg is an example of remarkable long-term decrease, and even such countries as France and Italy have shown a recent reduction in expenditure. The overall growth of Southern European social expenditure since 1980 is a good reminder of the fact that clusters of countries do not always last long (Gauthier 2002: 464–66). A surprise in the results is that the social investment index decreased in all Scandinavian countries. If social investment is as important for future development, as the politicians claim, why is Scandinavia not retaining its erstwhile lead in social investment politics? Is there a threshold difficult to exceed, since politicians cannot ignore the needs for social consumption? But if there is such a threshold, it seems to be lower in Denmark.

100 | Jorma Sipilä

The attempt to identify consequences of social investment policies yields clear results. There are systematically negative correlations between expenditure on old age, on the one hand, and both female employment and fertility, on the other hand. In light of earlier research it is no surprise that the statistical dependency between family expenditure and employment is not high; nevertheless, it is interesting to observe fairly high correlations between investment in education, female employment, and fertility. This observation suggests that welfare state researchers could benefit from paying more attention to education. A still-closer association is found between child care and female employment, but this result comes close to circular reasoning. We have every reason to believe that the supply of child-care facilities is conducive to both motherhood and employment, but it is also self-evident that working mothers need child care and that it is more and more difficult to provide it informally in a modern society. It is more surprising that even though inadequate child-care arrangements constrain mothers’ employment, there are several countries that stick to the existing policy arrangements (Randall 2000). This study, too, is proof that the blessing of social investment has not convinced all governments.

Conclusion “Slowly but surely a ‘social investment model’ is replacing the ‘social security’ paradigm inherited from the sixties,” said John Myles and Debra Street in 1994 (see Jenson and Saint-Martin 2003: 82). I am not sure that the social investment state is gaining ground, even slowly. It is not possible to say that the social investment approach somehow dominated the budgetary policies of OECD governments from 1998 to 2003. Some change did indeed occur in the expected direction, but some did not, and the development was very uneven, especially outside Europe. The empirical findings, rather, support Gauthier’s embarrassing statement: “The rapid decline of fertility from the 1960s on did not result in large increases in cash support for families, nor did the large increase in population aging reduce governmental support for families” (2002: 464). On the other hand, the international rhetoric that supports the social investment state does not exist without reason. In light of my findings, it can be said that the level of social investment is connected to the desired outcomes. Countries investing in education and families, particularly in child care, seem to maintain high female employment rates, whereas countries that spend generously on old age have apparently also accepted lower

A New Trend in Social Expenditure or Merely a Popular Political Discourse? | 101

employment and fertility rates. Nevertheless, more empirical analyses are required to confirm these preliminary findings. The supporters of the social investment state cannot be satisfied with the development hitherto, but nor should the opponents feel relieved. It is highly likely that the social investment approach will affect public expenditure in the future.

Notes 1. There are varieties of expressions—many researchers do not like to speak of a social investment state but rather of the “modernization of social policies” or “new risks” etc. (cf. Clasen and Siegel 2007). 2. The year 1991 is a good choice for presenting intermediate data between 1980 and 1998. Germany had been reunified, and the new East European republics were established in 1990. At the moment the OECD data could be extended to the year 2007. 3. Google Scholar gives 13,900 hits for “globalis(z)ation” and 604 for “aging population” in 1998. 4. Only six hits before 1999. 5. Iceland was excluded, which I regret. 6. Actually, the standardized expenditure on families increased at the same yearly rate from 1980 to 1991 and 1998 to 2003. 7. If the expenditure remains at the same level and the proportion of children in the population decreases, the expenditure on families per child increases. A reverse development might happen with the expenditure on old age per elderly person as the share of elderly in population increases. 8. OECD social expenditure aggregated data. Data for Hungary 1999 instead of 1998. Data for Poland, benefits in kind, 1999 instead of 1998. 9. The considerable loss in Canada has been partly covered by expanding private expenditure (see OECD Factbook 2007). 10. Because of a sudden change in the compilation of statistics on old age I do not use OECD figures for Finland (1998 and 2003) and for Luxembourg (2003). Instead I use Eurostat. 11. An interesting detail is that Finland and Sweden are the only countries where the female employment rate was lower in 2003 than in 1980. 12. This is particularly true if we take into account private expenditure on education that has been very high in the United States. Canadian expenditure on education was also high in 1995, but no longer in 2003. 13. The term of Kamerman and Kahn 1977.

Bibliography Adema, Willem, and Maxime Ladaique. 2005. “Net Social Expenditure, 2005 Edition: More Comprehensive Measures of Social Support.” OECD Social Employment and Migration Working Papers No. 29. Paris.

102 | Jorma Sipilä

Bambra, Clare. 2007. “‘Sifting the wheat from the chaff:’ A two-dimensional discriminant analysis of welfare state regime theory.” Social Policy & Administration 41, no. 1: 1–28. Caldwell, John, and Thomas Schindlmayr. 2003. “Explanations of the fertility crisis in modern societies: a search for commonalities.” Population Studies 57, no. 3 (November): 241–63. Chauvel, Louis. 2003. Génération sociale et socialisation transitionnelle: Fluctuations cohortales et stratification social en France et aux Etats-Unis au XXe siècle. Mémoire d’Habilitation à Diriger des Recherches. Paris: Sciences-Po. Clasen, Jochen, and Nico A. Siegel, eds. 2007. Investigating welfare state change. The ‘dependent variable problem’ in comparative analysis. Cheltenham: Edward Elgar. Esping-Andersen, Gøsta. 1996. “After the Golden Age? Welfare State Dilemmas in a Global Economy.” In Welfare States in Transition: National Adaptations in Global Economies, edited by Gøsta Esping-Andersen, 1–31. London: UNRISD. ―――. 1985. Politics Against Markets: The Social Democratic Road to Power. Princeton: Princeton University Press. ―――. 1990. The Three Worlds of Welfare Capitalism. Cambridge: Polity Press. European Commission 2004. Report of the High Level Group on the Future of Social Policy in an Enlarged European Union, Report of the High Level Group. http:// www.pedz.uni-mannheim.de/daten/edz-fd/gds/hlg_social_elarg_en.pdf (accessed 9 July 2011). European Parliament. “Presidency Conclusions.” Lisbon European Council 23– 24 March 2000. http://www.europarl.europa.eu/summits/lis1_en.htm (accessed 20 November 2009). Ferrera, Maurizio. 1996. “The Southern Model of Welfare in Social Europe.” Journal of European Social Policy 6, no. 1: 17–37. Gauthier, Anne H. 2002. “Family Policies in Industrialized Countries: Is There Convergence?” Population 57, no. 3: 447–74. Giddens, Anthony. 2007. Europe in the Global Age. Cambridge: Polity Press. ―――. 1998. The Third Way: The Renewal of Social Democracy. Cambridge: Polity Press. ―――. 2000. The Third Way and Its Critics. Cambridge: Polity Press. Green-Pedersen, Christopher. 2007. “More than data questions and methodological issues: theoretical conceptualization and the dependent variable ‘problem’ in the study of welfare reform.” In Investigating Welfare State Change: The ‘Dependent Variable Problem’ in Comparative Analysis, edited by Jochen Clasen and Nico A. Siegel, 13–23. Cheltenham: Edward Elgar. Jaumotte, Florence.2003. “Labour Force Participation of Women: Empirical Evidence on the Role of Policy and Other Determinants in OECD Countries.” OECD Economic Studies 2003, no. 37: 51–108. Jenson, Jane, and Denis Saint-Martin. 2003. “New Routes to Social Cohesion? Citizenship and the Social Investment State.” Canadian Journal of Sociology 28, no. 1: 77–99. Kamerman, Sheila B., and Alfred J. Kahn. 1977. Family Policy: Government and Families in Fourteen Countries. New York: Columbia University Press.

A New Trend in Social Expenditure or Merely a Popular Political Discourse? | 103

Kvist, Jon. 2006. “Europeiska perspektiv på den nordiska välfärdsstaten.” Nordisk Ministerråds Velferdsforskningsprogram. Programkomitéens sluttrapport. In TemaNord 2006 no 521:128–39. Copenhagen: Nordisk Ministerråd. Lehtonen, Heikki. 1983. Työvoiman käytön ja uusintamisen muodot. Tampere: Vastapaino. OECD. 2001. Employment Outlook. Paris: OECD. ―――. 2007. “Finnish Policy on Reconciling Work and Family Life Should Strengthen: A Country Note.” In Babies and Bosses: Reconciling Work and Family Life—A Synthesis of Findings for OECD Countries. Paris: OECD. http://www. oecd.org/document/0/0,3343,en_2825_497118_34916736_1_1_1_1,00.html (accessed 20 November 2009). ―――. 2007. “Matching Work and Family Commitments: Issues, Outcomes, Policy Objectives and Recommendations.” In Babies and Bosses: Reconciling Work and Family Life—A Synthesis of Findings for OECD Countries. Paris: OECD. http://www.oecd.org/dataoecd/18/27/39689983.pdf (accessed 20 November 2009). OECD Factbook. 2007. Economic, Environmental and Social Statistics. Paris: OECD. OECD Family Database 2008. http://www.oecd.org/els/social/family/database (accessed 20 November 2009). Olk, Thomas. 2006. “Children in the ‘Social Investment State.” Paper presented at the Second WELLCHI Network Conference. Centre for Globalisation and Governance, University of Hamburg, 31 March–1 April 2006. http://www. ciimu.org/webs/wellchi/conference_2/streamb/olk.pdf Palme, Joakim. 2006. “Velferdsforskningsprogrammet i ljuset av den nordiska modellens utmaningar: Forskningen och socialpolitikens modernisering.” Nordisk Ministerråds Velferdsforskningsprogram. Programkomitéens sluttrapport. TemaNord 2006: no. 521, 155–94. Copenhagen: Nordisk Ministerråd, 2006. Perkins, Daniel, Lucy Nelms, and Paul Smyth. 2004. “Beyond Neo-liberalism: The Social Investment State?” Social Policy Working Paper No. 3. Fitzroy: Brotherhood of St Laurence. http://www.bsl.org.au/pdfs/beyond_neoliberalism_social_investment_state.pdf (accessed 20 November 2009). Pierson, Paul. 1996. “The New Politics of the Welfare State.” World Politics 48, no. 2 (January): 143-179. de la Porte, Caroline. 1999. “Is there an Emerging European Consensus on Social Protection?” In European Trade Union Yearbook 1999, edited by Emilio Gabaglio and Reiner Hoffmann, 309-340. Brussels: European Trade Union Institute. Randall, Vicky. 2000. “Child Care Policy in the European States: Limits to Convergence.” Journal of European Public Policy 7, no. 3 (September): 346–368. Sipilä, Jorma, and Anneli Anttonen. 2008. “Mobilising Formal and Informal Resources in Meeting Old Age–Related Needs—a European Comparison.” In Ermunterungen, Soziale Demokratie, Zivilgesellschaft und Bürgertugenden, edited by Claus Leggewie and Christoph Sachsse, 169–201. Frankfurt a. M.: Campus. Streeck, Wolfgang. 1999. “Competitive Solidarity: Rethinking the ‘European Social Model.’” Working Paper No. 99/8. Max-Planck Institute for the Study of

104 | Jorma Sipilä

Societies, September 1999. http://www.mpifg.de/pu/workpap/wp99-8/wp998.html (accessed 20 November 2009). Väyrynen, Raimo. 2000. “Globalisation and the Politics of Labour in Finland.” In Festschrift for Dag Anckar on his 60th birthday on February 12, 2000, edited by Lauri Karvonen and Krister Ståhlberg. Åbo: Abo Akademi University Press.

Section II

INTEREST COALITIONS, IDEAS, AND SOCIAL REFORM

(

Chapter 5

MULTIPLE MARKET PRESCRIPTIONS The Diverse Models of Health Care Reform in Sweden

( Jane Gingrich

The intersection of market reforms and health care has become increasingly prominent in recent decades. As policy makers struggle to address rising costs and appeal to confident and choosy patients, market reforms have emerged on the agendas of some of the European Union’s largest and smallest member states. Much of the literature on market reform presents these changes as producing common outcomes and emerging from a common political constellation; however the experience of market reform in health care is far more varied, both in terms of the effects on service production and the politics behind them. This chapter looks at market reforms through the lens of Sweden, one of Europe’s smallest countries but largest welfare states. It argues that politicians on the Left and the Right in Sweden introduced diverse types of markets and used these differences as strategic tools to achieve varied ideological and electoral aims. This claim is at odds with much of the literature on market reforms, which tends not to systematically differentiate among types of markets. Proponents of markets present them as a panacea to the problems of both cost and quality in mature health care systems. These scholars promise that markets will improve the efficiency of health care systems by limiting the state, thereby making patients more sensitive to the costs of care and giving doctors and hospitals incentives to be more efficient (Osborne and Gaebler 1992; Savas 2000; Lundsgaard 2002). In making these promises, though, proponents often equate a range of market-inspired reforms, pledging that everything from competition among insurers to higher patient fees will improve health care systems. Even those advocates who develop detailed policy proposals aimed at getting the incentives right

108 | Jane Gingrich

draw the logic of a single well-functioning market (e.g., Preker, Harding, and Travis 2000). In contrast, opponents of market reforms take a different tack. They, too, argue that market reforms undermine the state, but far from seeing this movement as a positive development, they claim that market competition worsens the quality, efficiency, and equity of health care provision (Leys 2003; Pollock 2004; Suleiman 2003). Market reforms bring “all bad things together,” with everything from user fees to contracting with private providers producing outcomes that threaten patients, worsen efficiency and harm workers. For both proponents and opponents of market reforms, because such policies trade off state for market control, they spell the end of politics. The implications for smaller countries are particularly important, as these countries’ small size creates few counterweights to the homogenizing effects of markets. Indeed, in examining markets, both proponents and opponents often equate them with the American model: the former pointing to medical innovation and short waiting times in the United States as evidence for markets, while the latter point to inequities and high spending in the American system as a warning against them (Goodman, Musgrave, and Herrick 2004; Pollock 2004). Small countries that choose market reforms are importing an American logic, limiting the choices of domestic politicians. A third line of thinking rejects these uniform assessments of market reforms, arguing that markets are powerfully shaped by preexisting institutions. Far from claiming that market reform undermines the state or service delivery, these institutionalists focus on continuity in welfare services, downplaying the importance of novel reforms (Jacobs 1998; Tuohy 1999). While accepting the potential diversity of markets, these scholars focus on the contingent and contextual features of national health-care systems that shape markets. For these scholars, small countries can maintain their distinctive character, but largely by avoiding substantial change. Yet none of these debates speaks to what is happening on the ground. Most countries in the European Union have engaged in substantial reform of their health care systems, changing how they pay doctors and hospitals, inspect and monitor quality, manage performance, and set overall budgets. These shifts, though, have hardly been uniform, and markets have contributed to highly divergent outcomes. European countries have increased both public and private spending, made hospitals compete while also enhancing state oversight of their practices, and contracted with the private sector while simultaneously expanding the public sector (see, for instance, Saltman, Busse, and Mossialos 2002; Figueras, Robinson, and Jakubowski 2005; Saltman and Figueras 1997; Kimberly and de Pouvourville 1993).

Multiple Market Prescriptions: The Diverse Models of Health Care Reform in Sweden | 109

This chapter argues that these varying outcomes are not accidental but emerge from systematic variation in markets themselves. Market structures can empower different actors in the health care system: patients, private providers, or the state itself. Political parties use these different markets strategically to achieve their varying aims. The Swedish case provides a window into this process. Over the past two decades, Swedish reformers have introduced a series of market reforms in health care and eldercare with profoundly different effects. Some market reforms increased the quality and patient orientation of the system; others increased the power of public-sector managers to control costs; and yet others devolved power to the private sector. These differences were not a matter of chance; rather, they followed from the reform choices of politicians on the Left and the Right. The Left introduced markets in an attempt to sustain the long-run viability of the welfare state (see Klitgaard 2007 on this point), while the Right used them to target the public monopoly. These differences resulted in varying reforms, with the Left promoting more choice within the public sector and the Right supporting private entrepreneurship without strong state control. While we see similar partisan calculations play out in the eldercare sector, they were more muted because the political stakes of reform were lower. The introduction of markets, far from reducing the scope for political influence in Swedish health services, highlights the importance of political contestation over generic reform ideas. This chapter makes these arguments in three sections. Section one briefly outlines the structural features of health and care markets that give policy makers the scope to construct them differently. Sections two and three examine market reforms in the Swedish health care and eldercare systems. It builds on an analysis of primary and secondary documents and interviews of over thirty policy makers in Sweden from 2004 to 2006.

Markets in Health and Care Services Defining market reforms in health care is a difficult task. In the most general terms, market reforms involve moving from a command-and-control or networked form of governance to one where resources are allocated through competitive incentives; however, there is no single set of competitive incentives. First, some market reforms target the revenue side, reshaping public and private responsibility for financing services. The public sector, through direct expenditure, social insurance contributions, or indirect mechanisms like tax breaks and mandatory private expenditure finances most Euro-

110 | Jane Gingrich

pean health care. Reforms that allow private individuals or firms to purchase services directly, allow individuals to voluntarily insure themselves, place new fees on patients for care or insurance, or simply limit what the public sector offers (e.g., through benefit cuts or strict budgeting) encourage more private spending relative to public spending. Each of these shifts directly or indirectly privatizes part of health care spending, giving new costs to patients. Second, other market reforms reshape the delivery of publicly funded services, either allowing more private provision or introducing competitive incentives within public services. However, health is a notoriously complex good beset by a range of market failures, which means that there is no single private health care market. These complexities create systematic differences in how policy makers can structure market reforms to affect the delivery of services. In most European countries the public sector funds much of the production of these services, but ultimately, as services, they are used by individuals. The users of services (patients, elderly clients) and those who pay for the services (the government, insurers) are different. Reforms that expand patient choice (e.g., through vouchers or funding that follow patient choices) give hospitals and care providers incentives to appeal to the patients who use the services but do not pay directly for them. In contrast, reforms that make hospitals and care providers compete for contracts from the state or insurers link funding to these purchasers’ performance goals, encouraging attention to their preferences. However, health and elderly care are complex goods, with many components of quality that are hard to specify in advance in a single contract or set of rules (Eggleston and Zeckhauser 2002). This means that market reforms alone, absent strong regulation, may actually give new producers (doctors, hospitals) the space to follow their own interests. These structural features of health care services mean that reforms introducing competition for patients’ choices or strongly regulated government contracts, or that create a deregulated market are likely to operate differently. These differences in market reforms are not just a matter of academic interest—they can have profound implications for patients, providers, and the state itself. Reforms that increase private financing change how patients access care. By making individuals more responsive to the “price” of care services, these reforms encourage individuals to consume more carefully, but they also risk making lower-income patients forego beneficial care because of high costs (see Barr 2004 on arguments over cost sharing). On the delivery side, reforms that increase competition for patient choice encourage hospitals to treat patients quickly and appeal to their needs, but they risk reducing hospitals’ incentives to satisfy cost-conscious purchasers. The logic is reversed for contracting reforms, which give hospitals an

Multiple Market Prescriptions: The Diverse Models of Health Care Reform in Sweden | 111

incentive to appeal to the government or insurers who are buying services but do not necessarily encourage attention to the needs of patients. However, reforms that expand either type of competition without clear regulation of hospital behavior give hospitals and care providers the space to innovate without guaranteeing that innovations will serve the interests of either patients or purchasers. Politicians often have different goals in reforming health care systems—technocratic cost control, electoral appeal, or long-term ideological goals. Varying markets can promote particular strategies. Politicians who favor more equality across patients and elderly users will be cautious about reforms that increase fees and costs for users, while those who favor more individual responsibility for care costs will be drawn to such reforms. On the delivery side, politicians who wish to control costs may be attracted to strongly regulated contracting, while those who wish to make reforms that will be popular among the broad electorate will see choicebased reforms as favorable, even if the reforms increase spending. Finally, those looking to promote the interests of private actors will prefer looser systems of regulation. In general, parties on the Left tend to support expansive health and care programs and would like to retain public support for them. These parties tend to avoid markets that allow new sources of private control, showing preference to those that appeal to patients through choice or that increase state control over the cost structure. Both reforms promise to enhance the long-run sustainability of the welfare state, either politically or fiscally. Which of these two markets appeals most to the Left depends on the circumstances, particularly whether choice-based market reforms can be introduced without increasing segregation and divisions across classes. By contrast, the Right is more positive toward markets that promote the private sector without increasing the size or regulatory power of the state. These markets, though, can be electorally unpopular, pushing the Right to turn to other market options. The next two sections provide evidence for the two main claims outlined above—that markets vary and that political actors will use this variation strategically. The sections show that Swedish politicians did, indeed, construct different types of markets and that these differences were the result of distinct partisan strategies.

Health Care in Sweden From the initial postwar years through the end of the 1970s most reforms to the Swedish health care system looked to expand the scope of the publicly financed and delivered system. Reforms through the 1960s and

112 | Jane Gingrich

1970s put private hospitals under public control. By the mid-1980s, over 90 percent of health care expenditure was public, with fewer than one in ten thousand physicians not dependent on public revenue (OECD 2004; Rosenthal 1992). Over the past two decades, reformers on the Left and the Right moved to introduce markets into this system but did so in different ways.

The Move Toward Choice During much of the postwar period, markets in health services were far from the Swedish political agenda. Throughout this period, the left-wing Swedish Social Democrats (SAP) played a dominant role in political life, leading all postwar governments until 1976. In government, the Social Democrats constructed an expansive welfare state, first focusing developing generous transfer programs (such as pensions and unemployment insurance) and later expanding services like health and care. In building up these services, the Social Democrats resisted markets, explicitly supporting a fully publicly funded and provided system (e.g., SAP 1975). The Swedish Right, by contrast, was fragmented across several smaller parties: the centrist- Liberal and Center parties and the more economically liberal Moderate Party (and later, the Christian Democrats). Through the postwar period, these parties were largely excluded from government, and while in the opposition, they took a conciliatory stance toward much of the welfare state. However, new economic and political challenges in the 1980s led both the Left and the Right to reevaluate the role of markets in the welfare state. Unsurprisingly, the Swedish Right took the first steps toward markets. While the non-Socialist governments (1976–1982) enacted few changes to the welfare state, during this period, the Moderate party began to consider more substantial reform. Sweden’s economy was under stress, with unemployment and inflation growing in the wake of the worldwide oil crises. Despite this economic pressure, the union movement began making radical demands for more worker control of private companies (Pontusson and Kuruvilla 1992). The employers responded vociferously against these demands, leading the Moderates (with whom they had long-standing links) to take a more radical pro-market position. This criticism targeted the welfare state, with employers and the Moderates questioning the dominance of the public sector as a financier and provider of public services. The combination of Sweden’s economic difficulties, employers’ dissatisfaction, the emergence of new-Right ideas globally following the election of Margaret Thatcher and Ronald Reagan, and growing concerns

Multiple Market Prescriptions: The Diverse Models of Health Care Reform in Sweden | 113

about the quality of services all encouraged the Moderates to contemplate privatization across the welfare state (Nilsson 2003). In the early 1980s, the Moderates began to talk about the need for a “system shift” in Sweden, promoting market reforms in health and care services as well as other parts of the welfare state (Nilsson 2003). Policy advocates within the party, the employers association, and influential think tanks considered a range of options to the financing and delivery of health and care services. Some advocated expanding the private health insurance system and allowing those that could afford to do so to purchase care privately. Equally, the party debated contracting with private providers, more limited proposals for expanding competition within the public sector, and more patient choice (ibid.; Blomqvist 2002). Behind these diverse proposals lay a vision of a more limited state with greater private funding and production. New electoral and economic challenges were also leading some Social Democrats to consider market reforms. Politically, the size and loyalty of the party’s working-class base had shrunk (Bergström 1991). There were too few unionized private-sector workers to ensure electoral victory, and as Sweden’s economy faltered, supporting either private- or public-sector workers’ wage demands was increasingly difficult. The Social Democrats felt under pressure to make new appeals that extended beyond the union movement. The party had long appealed to middle-class voters because of its support for generous pension benefits, and throughout the 1970s, health and care services (for children and the elderly) had become increasingly important to these voters. The party saw that ensuring that these services continued to appeal to middle-class voters was crucial for its long-term political success. However, Sweden’s economic difficulties meant that making such appeals by increased spending was difficult. Expenditure on welfare programs had grown rapidly during the 1970s, particularly for health and education services, and the Social Democrats had committed themselves to restoring the Swedish economy through fiscal austerity. In the 1982 election the party faced the challenge of addressing middle-class demands over public services in an environment that dramatically limited spending. Initially, these challenges did not create a constituency in the party for markets. The Social Democrats fought the 1982 and 1985 elections on promises of defending the welfare state, taking a critical stance toward the Moderates pro-market proposals (Olsson 1993). However, problems in the health sector continued to grow throughout the decade. The party first addressed rising costs in the health care sector by capping expenditure, which slowed spending increases, but this move exacerbated already long waiting lists. The media began reporting rising levels of citizen frustration, particularly in the larger and more affluent urban areas (Diderichsen 1995;

114 | Jane Gingrich

Saltman and von Otter 1992). Increasingly, policy makers both within the party and in the bureaucracy saw budgetary caps as a weak tool for improving long-run productivity (Federation of County Councils 1991). These concerns were in addition to a small, but politically worrying, trend—wealthier Swedes were purchasing private health insurance. Although the number of citizens with private policies remained small, between 1986 and 1990, Sweden’s biggest insurer grew from 4,500 to 15,000 enrollees and the second biggest grew from 1,000 to between 5,000 and 6,000 enrollees (Rosenthal 1992). The Social Democrats saw this “opt out” to the private sector as distressing, threatening to reduce middle-class support for the public system (Interview: former SAP health advisor, 2006). 1 The Social Democrats, then, increasingly felt the need to engage in a more fundamental reform of the health care system. In this environment the party began to consider market reforms. Unlike the Moderates, it did not aim to introduce a “system shift,” but rather to improve efficiency and limit faltering support of the middle class for the health care system. The influential minister of finance, Kjell-Olof Feldt, presented precisely this logic. In arguing for market reforms in public services, he claimed that they could improve productivity and, crucially, sustain the public’s belief in the legitimacy of the welfare state: “We can meet this challenge [posed by the Right] only if the welfare state is witnessed to be as effective in satisfying people’s needs as is the commercial sector of society. If the public sector is not in accord with people’s needs and demands . . . the Right’s message of freedom will break through” (quoted in Heclo and Madsen 1987). The Social Democrats distinct aims led it to consider different market reforms than the Moderates. The Social Democrats, particularly within the Ministry of Finance, initially considered a wide range of proposals aimed at improving efficiency, including contracting reforms (Hakansson, Paulson, and Kogeus 1988; Blomqvist 2002; Interview: former health advisor, SAP, 2006). At the same time, several left-leaning policy advocates developed different market reforms that directly targeted the Social Democrats’ dilemma (Blomqvist 2002). Richard Saltman, an American academic, and Casten von Otter, a trade union economist, published an influential article promoting expanded choice in Swedish health care. They argued that there were structural flaws in the system that limited responsiveness to patients and decreased efficiency (Saltman and von Otter 1987). Saltman and von Otter argued that in order to sustain the public sector and limit the growth of private financing, governments should avoid fees and should expand competition not for contracts—but for patient choice (Saltman and von Otter 1989).

Multiple Market Prescriptions: The Diverse Models of Health Care Reform in Sweden | 115

As both the Moderates and the Social Democrats sifted through these proposals, their varying ideological preferences and electoral concerns led them toward different types of market reforms.

A Move Toward Choice: Markets and the Social Democrats Local governments introduced the first changes. In Sweden, twenty-one county councils run individual health care systems. In 1985, the centerright governing coalition of Sweden’s biggest county, Stockholm County, expanded women’s choice of maternity-care provider and paid hospitals, in part, based on these choices. These reforms also allowed a new private contractor to provide some primary-care services. In the late 1980s, though, it was the Social Democrats who developed competition through choice within the public sector. Historically, the counties assigned citizens to a specific clinic and hospital, allowing little patient choice across county lines and even within the county. In Stockholm, a Social Democrat–led coalition in 1988 gave patients choice of any primary care provider and hospital within the county, with money following these choices. In 1991, the Social Democrat–controlled Federation of County Councils further expanded choice across county lines and urged individual counties to allow more patient choice. Nationally, both the Social Democrats and the non-Socialist parties supported legislation that guaranteed choice across county lines to patients who had been waiting for common procedures for a fixed amount of time. This new emphasis on competition for patient choices began to reshape the Swedish care system. By 1993, all twenty-six counties had extended choice for primary care, with seventeen allowing choice of hospital without a referral and nine allowing it with a referral (Anell, Rosen, and Hjortsberg 1997). While some counties, including Stockholm, also imported a market-like contracting model into the public sector, in practice these reforms were limited (Anell 1996). Funding still followed patient demand, and the government did not limit choice or set production limits. Indeed, if the hospitals produced more than the contracts stated, the hospitals sent the bills directly to the counties (Anell 1996; Anell, Rosen, and Hjortsberg 1997; Diderichsen 1995). Both the national regulatory structure, which limited hospitals’ ability to select or refuse patients, and county control over market entry, pricing, and medical practices backed up patient choices. These reforms produced a particular set of outcomes. While some counties also increased fees, placing new costs on patients, the major thrust of this emerging market was to expand competition among hospitals for cost-insensitive patients.2 Even though only a small number of patients actually chose an alternative hospital, hospitals had an incentive to appeal

116 | Jane Gingrich

to patients, and they responded accordingly. From 1991 and 1993, waiting lists fell 22 percent, with hospital production increasing (Harrison 2004). In maternity care, hospitals in counties like Stockholm and Malmö began to compete for expectant mothers in new ways, allowing them more control over birthing practices (Rothstein 1998; Saltman and von Otter 1992). Similar patterns emerged outside of maternity care, where hospitals facing heavy competitive pressures developed a new emphasis on quality measurement and management (Garpenby 1994, 1997). This evolving market, though, came at a cost. While the Swedish counties were able to limit overall health expenditure, the choice orientation put pressure on budgeting. Budget overruns emerged in the areas where competition for patients was most developed, with some evidence that hospitals increased activity beyond what was medically necessary (Diderichsen 1995; Whitehead, Gustafsson, and Diderichsen 1997). In the parts of Sweden where competition across county lines was fierce, particularly western Sweden, counties experienced difficulties with budgetary planning (Anell 1996). At first, these counties competed to retain their patients, but as Sweden entered an economic recession in the early 1990s, many began to limit the financial incentives around choice (Anell 1996). This first move toward market reforms in Swedish health care empowered patients, even at the cost of budgetary control. These outcomes followed from a particular political calculus. Although the non-Socialists rhetorically supported choice, it was the Social Democrats who pushed it forward. This policy emerged because the Social Democrats saw choicebased markets within the public sector as a way to sustain public support for the Swedish health care system. Indeed, Gertrude Sigurdsen, the Social Democratic minister of health stated precisely this logic: “People should be able to visit the doctor, the health centre, or the hospital they wish, also across county lines. Attempts with such choice have been shown to reduce the demand for private alternatives” (quoted in Saltman and von Otter 1992). Choice promised to reduce waiting lists and improve productivity—addressing the middle class’s concerns about the health care system. The turn to the market, then, was a turn to a specific type of market with particular outcomes. As we shall see in the next section, the Social Democrats were far less enthusiastic about markets that threatened to fragment support for the public system.

Expanding the Private Sector—A Strategy from the Right While the non-Socialists initially supported patient choice—and continue to do so in theory—in practice, they moved away from it as costs increased and the choice orientation began to work in the Social Democrats’

Multiple Market Prescriptions: The Diverse Models of Health Care Reform in Sweden | 117

favor. Indeed, in the early 1990s, markets seemed to fall off the Swedish reform agenda (Whitehead, Gustafsson, and Diderichsen 1997; Harrison and Calltorp 2000). However, even as choice-based markets became less prominent, the non-Socialists, particularly the Moderate Party, pushed for reforms expanding the scope and role of the private sector, often absent strong regulations. The 1991 elections produced a national non-Socialist governing coalition, led by the Moderate Party. In 1994, this government introduced the so-called “Family Doctor” reforms. Historically, Swedish patients received primary care from a clinic, rather than from a personal doctor. The 1994 reforms proposed allowing each citizen to opt for a personal private doctor and gave private doctors the right to enter the market irrespective of local need or county planning. Locally, non-Socialist politicians were also promoting private entrepreneurship. In Stockholm County, non-Socialists had begun contracting with private firms in the 1980s, extending a contract with City Clinic AB, a private firm that specifically catered to patients working in central Stockholm (Rosenthal 1986). In the 1990s, Stockholm’s non-Socialist government expanded private contracting, requiring its purchasing offices to consider tendering out services worth more than 20 percent of their budget (Whitehead, Gustafsson, and Diderichsen 1997). In 1998, Stockholm’s non-Socialist governing coalition privatized an acute- care hospital, St. Göran’s. This move built on an earlier policy that granted St. Göran’s an independent status. While much of this contracting was heavily regulated (akin to the reforms in the eldercare sector—see below), the county tendered St. Göran’s hastily, without following Swedish public procurement laws (Öhrlings PricewaterhouseCoopers AB 2005), and allowed the new private contractor the scope to treat additional private patients. This reform and the Family Doctor reform both gave new private-actor rights in the public market: Doctors now had autonomy over key actions such as whether to enter and exit the market, charge higher fees, and treat nonpublic patients. This move contrasts to contracting in care, where the public sector maintained much control over the contractors’ actions, limiting their autonomy. In practice, the impact of both reforms was limited by further legislation from the national Social Democrats. Nonetheless, the changes did produce outcomes that differed from the early choice-based reforms. Contracting with private physicians is now well established in parts of Sweden; for instance, in Stockholm close to 50 percent of doctors have some private practice (World Health Organization 2001). While much of this contracting is heavily regulated and the counties retain much control over the private sector, there is evidence that following the Family Doc-

118 | Jane Gingrich

tor reforms, private physicians used their position to increase costs and cater to wealthier neighborhoods (Whitehead, Gustafsson, and Diderichsen1997). St. Göran’s shows a more positive story, using its new autonomy to innovate, with patients and staff positive about St. Göran’s performance (Lofgren 2002). Despite these differences, in contrast to the early choice orientation, both reforms worked to create an autonomous private healthcare sector, not to cut waiting lists or compete to attract patients. These changes demonstrate a distinct political strategy by the non-Socialists. Both reforms explicitly fit with the Moderates’ early goals for a “system shift,” cementing the position of nonpublic actors in the Swedish health care system. In the area of family doctors, although the Liberal Party had long supported personal physicians, it had not demanded giving private doctors the right to enter the market. However, the Moderates, who both supported the private sector and disliked county management, pushed hard for this policy (Nilsson 2003; Interview: former Stockholm health policy advisor 2005). Equally, in Stockholm County, the Moderates put privatizing St. Göran’s as a top political priority following the 1998 elections (Interview: Stockholm County non-Socialist politician, 2004). Neither of these reforms targeted the whole health care system, but each put a particular market strategy, supporting an autonomous private sector, on the Swedish health agenda. The Social Democrats reacted strongly to both sets of reforms, rejecting this type of market. When the Social Democrats returned to office nationally in 1994, the government rescinded the Family Doctor reforms. It allowed existing private doctors to continue operating but returned control over the private sector and market entry to the counties. The national government also responded to the St Göran’s reforms, introducing a ban (often called the “Stop Law”) on additional contracting of hospitals and later softening these reforms but disallowing private hospitals receiving public funds from also treating privately funded patients. Although the Social Democrats had both supported and expanded markets though choice, and even accepted some profit-making in the eldercare sector, the party strongly opposed a reduction in public control over the market. For the Social Democrats, allowing private actors to enter the market or cater to privately insured patients absent public permission threatened to fragment public support for the health care system (e.g., Sahlin and Johansson 2006). The Social Democrats feared that the privatization of St Göran’s, in transferring physical capital to the private sector and allowing private contractors to treat privately funded patients, would open the door for some patients to jump the public queue (Interview: SAP health advisor, 2005). In the long run, these moves risked eroding overall

Multiple Market Prescriptions: The Diverse Models of Health Care Reform in Sweden | 119

public support for the system—and for the Social Democrats themselves. The party, then, aggressively rejected this type of market.

Elderly Care Sweden has one of the most extensive elderly care systems in the OECD, spending close to 3 percent of GDP on care for the elderly and providing extensive home-care and institutional-care services (OECD 2005). This system developed slowly during the postwar years, but by the mid-1980s, the Swedish state—rather than the market or the family—was the key locus of care (Johansson, Sundström, and Hassing 2003). In practice, Sweden’s 289 municipalities largely provided and managed elderly care, with a large local public workforce. Like the health sector, a series of reforms have changed the care sector, introducing significant private contracting; however, lower levels of national attention to care sector have meant that this change has largely (albeit not exclusively) followed from local partisan calculations.

Opening the Door for Markets: Moves from the Left and the Right The Swedish elderly care system developed slowly throughout the postwar years, as both the municipal and national government expanded services to the elderly. This incremental development created a number of structural problems in the system. Crucially, the system divided responsibility for the elderly between the municipalities (for care) and the counties (for health). In an era of tighter budget constraints, this division of responsibility gave cash-strapped municipalities incentives to delay moving elderly patients out of costly hospital care thus shifting the costs onto the counties (so called “bed blocking”). These actions both increased costs, by housing the elderly in institutions, and harmed the quality of care, by not working to keep the elderly in their own homes (Interview: official, National Board of Health and Welfare, 2005). In the 1970s, the non-Socialist government established a commission to study the problem of bed blocking. However, when the commission finally reported its findings in the late 1980s, it suggested few concrete reforms (Bäck 1999). In 1988, the governing Social Democrats took another tack, launching another investigation into bed blocking, this time restricting membership on the commission in order to promote rapid reforms (ibid.). The strategy worked, with the commission delivering a comprehensive plan to “demedicalize” care by decentralizing financial responsibility for bed blocking to the municipalities (Trydegård 2003). The Social

120 | Jane Gingrich

Democrats agreed to these proposals, and in 1992 the government introduced the “Ädel Reforms,” which gave the municipalities full financial and organizational responsibility for care. Alongside decentralization, the Social Democrats also gave the municipalities new tools to organize care as they wished. The 1991 Local Government Act dramatically expanded the municipalities’ control over their internal organization, including the ability to contract. Nationally, then, the Social Democrats opened the door for some market reforms in care. The Social Democrats had long supported generous policies for the elderly (largely through pensions but also through care), but the party was worried about the rising costs of care. Decentralization not only promised to give the municipalities more incentives for cost control, but it also promised to shift some of the difficult decisions regarding expenditure to the local governments. Nevertheless, the party kept a national regulatory presence in the sector, preserved dedicated care funding, and did not force municipalities to introduce private contracting. Its goals were to limit rising cost in the sector, not radically change it. In opposition, the Moderates had been vocal proponents of markets in care, promoting reforms that expanded private financing, private delivery, and competition among providers. However the Moderates largely advocated markets in care under a larger public-sector reform umbrella (e.g., Moderata Samlingspartiets 1984). When elected in 1991, the party did not put care reform at the top of the agenda, focusing instead on the key priorities of the employers’ federation, child care, and education (Olsson 1993; Lundahl 1990). The other non-Socialists parties placed more emphasis on eldercare, but, although open to private alternatives, these parties also supported a generous public offer. The Liberal minister responsible for care, Bengt Westerberg, vocally expressed his support for the Swedish welfare state, advocating extensive public funding and regulation of elderly care services (e.g., Westerberg 1994). Thus, the non-Socialists did not prioritize markets in elderly care. As part of a broader set of reforms to local governance, the coalition removed the dedicated elderly care budget (allowing municipalities to choose how much to spend on care) and allowed municipalities to further raise fees and contract with for-profit providers (Bryntse and Greve 2002). However, despite complaints from the employers, the government did not force municipalities to tender care services and did not give private actors the right to enter the market (Lemmel 1993; Westerberg 1992). The different preferences among the non-Socialist coalition partners and the Moderate Party’s lower prioritization of the sector led to fewer reforms in this area.

Multiple Market Prescriptions: The Diverse Models of Health Care Reform in Sweden | 121

Markets at the Local Level: Cost Control and Choice Both the decentralization of care and the deregulation of municipal activity created a different logic at the local level. In the early 1990s, Sweden’s economy fell into a deep recession. The Ädel reforms provided the municipalities with some extra funding, but they also decentralized financial responsibility for care just as demands for other local services grew and overall revenue fell. The national government further limited local tax increases in an attempt to restore Sweden’s economic health. Asked to do more with less money, Swedish municipalities looked to raise revenue and cut costs (Swedish Welfare Commission et al. 2003). These actions led to changes in both the financing and production of care. In the area of financing, most municipalities raised fees. In the 1990s, one in six people over age seventy-five reported foregoing care because of fees (Trydegård 2003). Moreover, municipalities began to restrict access, with nearly one hundred thousand fewer citizens over sixty-five receiving home care during the crisis of early 1990s (Sipilä et al. 1997). As the number of recipients fell, municipalities targeted care to the most frail and highest-need elderly citizens (Szebehely 1998). However, how they did so varied substantially, with Trydegård and Thorslund (2001) finding some municipalities offering services to as few as 5 percent of citizens over eighty, while others offering services to over half of this group. Alongside these higher fees and limited access, many municipalities expanded contracting, both creating market-like systems within the public sector and contracting with private actors. By the end of the 1990s, half of municipalities had introduced a purchaser-provider split, separating the responsibility for “buying” services from providing them (Trydegård and Thorslund 2001). Most municipalities have not tendered out elderly care services to the private sector, but there has been a significant shift toward private production. In the early 1990s, less than 1 percent of care providers were private, and by the end of the decade, 8 percent of all providers and 13 percent of the care workforce were private (Swedish Welfare Commission et al. 2003; Suzuki 2001). Unlike the early market reforms in the health care system, this contracting has done little to expand choice (Svensson and Edebalk 2001). Elderly users often do not even know there is a market. This contracting has also given fewer benefits to the private actors. While many municipalities initially had problems contracting (Socialstyrelsen 1995), all municipalities retained control over market entry and user fees, and national legislation limited wage competition and, more recently, regulated quality. Municipalities did not have to privatize; they generally followed procurement law and mostly (although not entirely) limited private actors from providing extra privately funded care.

122 | Jane Gingrich

The result of these different reforms, then, is a third distinct market. In contrast to the early choice-based market in health care, which emphasized shorter waiting time and patient responsiveness, the elderly care markets have encouraged care providers to pay attention to costs. Overall, new contract-based private providers are less costly at a given level of quality; however, this cost advantage does not always continue, as municipal providers also offer lower prices to compete (Suzuki 2001). These findings suggest that municipal providers respond to competition by reducing their costs and that providers do not engage in uncontrolled cost cutting at the expense of quality. Both public and private care providers have focused on improving efficiency, disseminating commercial management and accounting models through the care sector, rather than improving (or cutting) quality (Suzuki 2001; Almqvist 2001). Furthermore, in contrast to the Family Doctor or St. Göran’s reforms, which empowered private actors at a cost to public control, these markets have often increased the role of the municipalities in steering the care system. In giving municipal managers the scope to set performance targets through contracts, they have weighed in on new issues and increased their control in the system (Interview: former official, National Board of Health and Welfare 2005). The local governments—not the patient or private providers—gained new power from these markets. At first glance, it would seem as though care markets are relatively nonpartisan. Nationally, neither party pushed markets in care. Locally, both the Social Democrats and the non-Socialist parties raised fees, with Trydegård and Thorsland (2001) finding no relation between political majority and the level of service provided. Equally, while non-Socialist municipalities have engaged in more contracting, some Social Democratic municipalities were early innovators in this area (Bäck 2003; Suzuki 2001). However, when we scratch below the surface, we see a more variegated path. For local politicians, elderly care is a crucial sector. Both Moderate and Social Democratic voters in Stockholm, for instance, have listed elderly care as the second most important issue for local governance, seeing education and housing, respectively, as the most important (Tenfält 2002). After the 1991 election brought non-Socialists into government in 212 out of 285 municipalities (Montin and Elander 1995) these local actors worked hard to put markets on the care agenda, looking to expand the number of private care providers while limiting taxes. Many chose to construct reforms similar to their Social Democratic colleagues, yet others looked to privatize care rapidly. Indeed, private providers in non-Socialist municipalities have higher costs (controlling for a range of demographic factors) than elsewhere, suggesting that local politicians put sustaining the private sector above cost saving (Suzuki 2001).

Multiple Market Prescriptions: The Diverse Models of Health Care Reform in Sweden | 123

In other cases, the non-Socialists have pursued a different agenda. About a dozen municipalities have bucked the contracting trend, introducing competition through client choice. Unlike the early choice reforms in health care, these municipalities allow users to choose among private providers, funding providers based on these choices (Edebalk and Svensson 2005). Local non-Socialists have been almost exclusively responsible for these moves. The right-controlled Stockholm suburb of Nacka pioneered the approach, first offering elderly users in need of foot-care services a choice among private providers and later extending this agenda to private and public child-care, education, and elderly-care providers (Kastberg 2001). Initially, few other municipal governments adopted the Nacka model; however, in the late 1990s, as the economy improved and local actors faced fewer immediate budgetary constraints, a number did introduce choice. Most prominently, Stockholm City, led by Moderate politician Carl Cederschiöld, expanded client choice in care and allowed care recipients to purchase “top up” care services from private (but not public) providers. Carl Cederschiöld, was able to increase care spending as part of these reforms without raising taxes by using windfall profits from privatization and the higher revenue emerging from Stockholm’s growing economy. Other municipalities, such as Linköping, introduced a flat voucher available to all citizens regardless of need. Local non-Socialists, in favorable circumstances again, built specific types of markets that aimed to support private alternatives to the state while making appeals to voters. In contrast, The Social Democrats at the national and the local level have been more critical of markets, particularly those benefiting the private sector. In the 1990s, in opposition, the party began to express reservations over market reforms. When they returned to government in 1994, the Social Democrats moved to address the problems of fiscal austerity in the sector by increasing expenditure by 20 percent and staff by 13 percent, even as the number of people above eighty increased by only 10 percent, and by later introducing new limits on fees in the sector (Swedish Welfare Commission et al. 2003; Gould 2001). After increasing funding, the government expanded quality regulations (e.g., requiring workers in homes to report problems). Local Social Democrats followed a similar strategy, parting with their non-Socialist counterparts on questions of funding and regulation. During the 1990s, most local Social Democrats moved toward managing existing markets. For instance, in Stockholm, while decrying the excesses of the non-Socialists, key Social Democrats expressed support for the principle of private alternatives (e.g., Nyberg, Torpe, and Billström 1995) but also raised taxes and expanded funding and regulation in the sector.

124 | Jane Gingrich

The Social Democrats have largely accepted private provision as a way to cut costs and improve efficiency, but they also expanded funding and regulation, as well as limited fees. This stance distinguishes the party from the Moderates, demonstrating less ideological commitment to markets. It is also different from the Social Democrats’ own stance in the health sector, where the party both more actively promoted one type of market and resisted another. The lower political stakes of elderly care made the Social Democrats more open to reforms that promised to save money and less attracted to those making broad appeals through choice that risked increasing spending. Taken together, we see that some non-Socialist governments in favorable circumstances have pushed toward choice as a way to promote private alternatives and appeal to wealthier constituents. However, most, facing more fiscal difficulties, have turned to contracting when looking to promote the private sector. Social Democrats have prioritized efficiency in the elderly care sector, but within limits, preferring to manage, and at times extend, contracting with private actors as a way to achieve more efficiency.

Conclusion The previous sections argued that not only can markets in health and care services differ, but that, as can be seen from the experience of Sweden, these differences matter in practice. Early reforms in the Swedish health care sector expanded choice, linking the funding of hospitals to patient choice. These markets created incentives for hospitals to appeal to patients, increasing production and costs. This outcome forms a strong contrast to more recent reforms in the health and elderly care sector. In the former, we see reforms aimed at creating and sustaining a private health sector, giving private health providers autonomy to determine how and what they offer. Finally, in the elderly care sector, we see a third market, which builds on heavily regulated contracts that largely (but not entirely) privilege cost cutting and efficiency without providing specific benefits to private actors. In both health and elderly care, we saw that these differences were not just a matter of chance, but followed from the strategic calculations of key partisan actors. Social Democrats, in the politically important health sector, accepted and promoted choice-based market reforms as a way to appeal to middle-class citizens concerned with quality. In contrast, in the less salient and geographically fragmented elderly care sector, the Social Democrats looked to cut costs and manage problems in the sector rather

Multiple Market Prescriptions: The Diverse Models of Health Care Reform in Sweden | 125

than make a broad-based appeal through markets. The calculus differed for the non-Socialists. These parties, largely driven by the Moderate Party, looked to support private providers and reshape the system to create more nonpublic providers. In health care, they made a number of strategic gambits in this direction, directly promoting private actors. Again, the lower political salience of elderly care limited the advantages of this strategy in the elderly care sector nationally, with local actors pursuing it under favorable circumstances. These differences across market strategies not only matter to the citizens, policy makers, and firms in Sweden, but also to our understanding of health care reform in European countries, large and small. First, the politics of market creation is not just a matter of being for or against markets, but rather, it affects how incentives are structured. The turn to markets does often draw on broader neoliberal ideas, yet within this movement there remain different ways markets are constructed. Second, politicians in small and large countries alike retain choices over how to build markets. Distributive battles—not just concerning spending but also the structures of markets—remain central to understanding the politics of health and care systems.

Notes 1. Interviews were conducted face to face, with some telephone follow-ups, from 2004 to 2006, using semistructured interview techniques. 2. Where fees did grow, they did have an impact. Through the 1980s and 1990s, more citizens reported that they did not seek medical care because of fees (Bergmark, Thorslund, and Lindberg 2000).

Bibliography Almqvist, Roland. 2001. “‘Management by Contract’: A Study of Programmatic and Technological Aspects.” Public Administration 79, no. 3: 689–706. Anell, Anders. 1996. “The Monopolistic and Integrated Model and Health Care Reform: The Swedish Experience.” Health Policy 37: 19–33. Anell, Anders, Per Rosen, and Catharina Hjortsberg. 1997. “Choice and Participation in Health Services: A Survey of Preferences among Swedish Residents.” Health Policy 40: 157–68. Bäck, Henry. 1999. “Local Politics, Markets and Parties.” Paper presented at the European Consortium for Political Research Joint Session of Workshops, Mannheim, Germany, March. ―――. 2003. “Party Politics and the Common Good in Swedish Local Government.” Scandinavian Political Studies 26, no. 2: 93–123.

126 | Jane Gingrich

Barr, Nicholas A. 2004. The Economics of the Welfare State. 4th ed. Oxford: Oxford University Press. Bergmark, Åke., Mats Thorslund, and Elisabet Lindberg. 2000. “Beyond Benevolence—Solidarity and Welfare State Transition in Sweden.” International Journal of Social Welfare 9, no. 4: 238–49. Bergström, Hans. 1991. “Sweden’s Politics and Party System at the Crossroads.” West European Politics 14, no. 3: 8–30. Blomqvist, Paula. 2002. “Idea and Policy Convergence: Health Care Reforms in the Netherlands and Sweden in the 1990s.” PhD diss., Columbia University. Bryntse, Karin, and Carsten Greve. 2002. “Competitive Contracting for Public Services: A Comparison of Policies and Implementation in Denmark and Sweden.” International Public Management Review 3, no. 1: 1–21. Burgermeister, Jane. 2004. “Sweden Bans Privatisation of Hospitals.” British Medical Journal, 328, 7438: 484. Diderichsen, Finn. 1995. “Market Reforms in Health Care and Sustainability of the Welfare State: Lessons from Sweden.” Health Policy 32, nos. 1–3: 141–53. Eggleston, Karen, and Richard Zeckhauser. 2002. “Government Contracting for Health Care.” In Market-based Governance: Supply Side, Demand Side, Upside, and Downside, edited by John D. Donahue and Joseph S. Nye, Jr., 29–65. Washington D.C.: Brookings Institution Press. Federation of County Councils. 1991. “Crossroads: Future Options for Swedish Health Care.” Stockholm: Federation of County Councils. Figueras, Josep, Ray Robinson, and Elke Jakubowski. 2005. Purchasing to Improve Health Systems Performance. Maidenhead, NY: Open University Press. Garpenby, Peter. 1994. “The Role of National Quality Registers in the Swedish Health Service.” Health Policy 29: 183–95. ―――. 1997. “Implementing Quality Programmes in Three Swedish County Councils: The Views of Politicians, Managers and Doctors.” Health Policy 39: 195–206. Goodman, John C., Gerald L. Musgrave, and Devon M. Herrick. 2004. Lives at Risk: Single-payer National Health Insurance Around the World. Lanham: Rowman & Littlefield. Gould, Arthur. 2001. Developments in Swedish Social Policy: Resisting Dionysus. New York: Palgrave. Hakansson, Stefan, Eric Paulson, and Kaj Kogeus. 1988. “Prospects for Using DRGs in Swedish Hospitals.” Health Policy 9, no. 2: 177–92. Harrison, Michael I. 2004. Implementing Change in Health Systems: Market Reforms in the United Kingdom, Sweden, and the Netherlands. London: SAGE Publications. Harrison, Michael I., and Johan Calltorp. 2000. “The Reorientation of Market-oriented Reforms in Swedish Health-care.” Health Policy 50, no. 3: 219–40. Heclo, Hugh, and Henrik Madsen. 1987. Policy and Politics in Sweden: Principled Pragmatism. Philadelphia: Temple University Press. Jacobs, Alan. 1998. “Seeing Difference: Market Health Reform in Europe.” Journal of Health Politics, Policy and Law 23, no. 1: 1–33.

Multiple Market Prescriptions: The Diverse Models of Health Care Reform in Sweden | 127

Johansson, Lennarth, Gerdt Sundström, and Linda B. Hassing. 2003. “State Provision Down, Offspring’s Up: The Reverse Substitution of Old-age Care in Sweden.” Ageing and Society 23, no. 3 (May): 269–80. Kastberg, Gustaf. 2001. “A Tool for Influence: The Effects of Introducing a Voucher System into In-home Elderly Care.” Working paper. School of Public Administration, Gothenburg University. Kimberly, John Robert, and Gérard de Pouvourville, eds. 1993. The Migration of Managerial Innovation. Diagnosis-Related Groups and Health Care Administration in Western Europe. San Francisco: Jossey Bass Publishers. Klitgaard, Michael. 2007. “Why are They Doing It? Social Democracy and Market-Oriented Welfare State Reforms.” West European Politics 30, no. 1: 172-94. Lemmel, Magnus. 1993. “All offentlig produktion bör konkurrensutsättas.” Dagens Nyheter, 29 June. Leys, Colin. 2003. Market-driven Politics: Neoliberal Democracy and the Public Interest. London: Verso. Lofgren, Ragnar. 2002. “The Swedish Health Care System: Recent Reforms, Problems, and Opportunities.” In Fraser Institute Occasional Paper 59. Vancouver: Fraser Institute. Lundahl, Lisbeth. 1990. “New Variations on Old Themes: The Swedish Conservative Party and the Battle Over Comprehensive Education 1900–1985.” Journal of Education Policy, no. 2: 157–66. Lundsgaard, Jens. 2002. Competition and Efficiency in Publicly Funded Services. Paris: OECD. Moderata Samlingspartiets. 1984. “Partiprogram.” [Political Program of the Moderate Party] Stockholm. Montin, Stig, and Ingemar Elander. 1995. “Citizenship, Consumerism and Local Government in Sweden.” Scandinavian Political Studies 18, no. 1: 25–51. Nilsson, Torbjorn. 2003. Moderaterna, marknaden och makten: svensk högerpolitik under avregleringens tid, 1976–1991. Stockholm: Södertörns Högskola. Nyberg, Lena, Tjia Torpe, and Annika Billström. 1995. “Våga främja privatalternativ.” Dagens Nyheter. OECD. 2005. Long-term Care for Older People. Paris: OECD. ―――. 2004. OECD Health Data 2004. Paris: OECD. Öhrlings PricewaterhouseCoopers. 2005. “Granskning av vårdupphandling samt de nya avtalen avseende St. Görans sjukhus AB.” Stockholm. Olsson, Sven E. (Hort). 1993. Social Policy and Welfare State in Sweden. Lund Studies in Social Welfare, vol. 3. Lund: Archiv. Osborne, David and Ted Gaebler. 1992. Reinventing Government: How the Entrepreneurial Spirit Is Transforming the Public Sector. New York: Addison-Wesley Publishing Co. Pollock, Allyson. 2004. NHS Plc: The Privatisation of Our Health Care. London: Verso. Pontusson, Jonas, and Sarosh Kuruvilla. 1992. “Swedish Wage-Earner Funds: An Experiment in Economic Democracy.” Industrial and Labor Relations Review 45, no. 4 (July): 779–91.

128 | Jane Gingrich

Preker, Alexander S., April Harding, and Phyllida Travis. 2000. “‘Make or Buy’ Decision in the Production of Health Care Goods and Services: New Insights from Institutional Economics and Organizational Theory.” Bulletin of World Health Organization 78, no. 6: 779–90. Rosenthal, Marilynn M. 1992. “Growth of Private Medicine in Sweden: the New Diversity and the New Challenge.” Health Policy 21, no. 2: 155-66. Rothstein, Bo. 1998. Just Institutions Matter: The Moral and Political Logic of the Universal Welfare State. Cambridge: Cambridge University Press. Sahlin, Mona, and Ylva Johansson. 2006. “Fredrik Reinfeldt begår löftesbrott.” Sweden: SAP. Saltman, Richard B., and Josep Figueras. 1997. European Health Care Reform: Analysis of Current Strategies. Copenhagen: World Health Organization, Regional Office for Europe. Saltman, Richard B., and Casten von Otter. 1987. “Re-vitalizing Public Health Care Systems: A Proposal for Public Competition in Sweden.” Health Policy 7, no. 1: 21–40. ―――. 1989. “Public Competition Versus Mixed Markets: An Analytic Comparison.” Health Policy 11, no. 1: 43–55. ―――. 1992. Planned Markets and Public Competition: Strategic Reform in Northern European Health Systems. Buckingham: Open University Press. Saltman, Richard B.; Reinhard Busse; and Elios Mossialos. 2002. Regulating Entrepreneurial Behaviour in European Health Care Systems. Buckingham: Open University Press. Savas, Emmanuel. 2000. Privatization and Public-Private Partnerships. New York: Chatham House. Sipilä, Jorma, Margit Andersson, Sten-Erik Hammarqvist, L. Nordlander, PirkkoLiisa Rauhala, Kare Thomsen, and Hanne Warming Nielsen. 1997. “A Multitude of Universal, Public Services—How and Why Did Four Scandinavian Countries Get Their Social Care Service Model.” In Social Care Services: The Key to the Scandinavian Welfare Model, edited by Jorma Sipilä, 27–50. Avebury: Aldershot. Socialstyrelsen. 1995. “Alternativa styr- och driftsformer i äldreomsorgen: Uppföljning, utvädering och avtal.” Stockholm, Socialstyrelsen. Suleiman, Ezra. 2003. Dismantling Democratic States. Princeton: Princeton University Press. Suzuki, Kenji. 2001. “Marketization of Elderly Care in Sweden.” European Institute of Japanese Studies, EIJS Working Paper Series. Paper Number 137. Svensson, Marianne and Per Gunnar, Edebalk. 2001. “90-talets anbudskonkurrens i äldreomsorgen: några utvecklingstendenser.” Swedish Institute for Health Economics, Lund. IHE Working Paper. Sveriges Socialdemokratiska Arbetarepartis (SAP). 1975. “Partiprogram.” [Political Program of the Swedish Social Democrats] Stockholm. Swedish Welfare Commission, Joakim Palme, Åke Bergmark, Olof Bäckman, Felipe Estrada, Johan Fritzness, Olle Lundberg, Ola Sjöberg, Lena Sommestad, and Marta Szebehely. 2003. “A Welfare Balance Sheet for the 1990s.” Scandinavian Journal of Public Health 31,Supplement 60: 10–12.

Multiple Market Prescriptions: The Diverse Models of Health Care Reform in Sweden | 129

Szebehely, Marta. 1988. “Changing Divisions of Care-Work: Caring for Children and Frail Elderly People in Sweden.” In Gender, Social Care and Welfare State Restructuring in Europe, edited by Jane E. Lewis, 257–283. Aldershot: Ashgate. Tenfält, Torbjörn. 2002. “Skola och bostäder viktigast.” Dagens Nyheter, 20 May. Trydegård, Gun-Britt. 2003. “Swedish Care Reforms in the 1990s. A First Evaluation of their Consequences for the Elderly People.” Revue française des affaires sociales 4: 443-460. Trydegård, Gun-Britt and Mats Thorslund. 2001. “Inequality in the Welfare State? Local Variation in Care of the Elderly—The Case of Sweden.” International Journal of Social Welfare 10: 174–84. Tuohy, Carolyn J. 1999. Accidental Logics: The Dynamics of Change in the Health Care Arena in the United States, Britain, and Canada. Oxford: Oxford University Press. Twaddle, Andrew C. 1999. Health Care Reform in Sweden, 1980–1994. Westport, CT: Auburn House. Westerberg, Bo. 1994. “Acceptera ine ett B-lag” Dagens Nyheter, 17 February. Westerberg, Per. 1992. “Inget avsteg från privatiseringsvägen.” Dagens Nyhter, 10 May. Whitehead, Margaret, Rolf A. Gustafsson, and Finn Diderichsen. 1997. “Why Is Sweden Rethinking Its NHS-style Reforms?” British Medical Journal 315, no. 7113 (11 October): 935–39. World Health Organization. 2001. “Health Care in Transition: Sweden.” In Health Care in Transition. London: European Observatory of Health Care Systems.

Chapter 6

AUSTRIAN SOCIAL POLICY REFORM IN THE ERA OF INTEGRATION AND RISING POPULISM

( Reinhard Heinisch

Although Austria has undergone extraordinary political changes during the past twenty-five years, its overall welfare state has proven durable, weathering better than most the neoliberal renaissance of the 1980s and the country’s subsequent integration into the European Single Market. In fact, as this essay argues, social and labor market policies have been important tools utilized by policy makers to cushion the effects of Europeanization and industrial restructuring and were essential in managing economic deregulation and liberalization. In this story, the social partners, i.e., labor and business associations, played the role of “modernization brokers” (Heinisch 1999), helping to translate, absorb, and distribute the economic and social pressures. In this respect, the Austrian social partners continued to serve as mediators of economic change, a role that they have performed throughout the postwar period in Austria. The pattern of change and continuity of the Austrian welfare model since the mid-1980s can be best explained in terms of the presence of important veto players (Tsebelis 2002), specifically the labor unions and the labor wing of the Social Democratic Party in conjunction with key institutional and structural determinants. These factors have mitigated significant and steadily growing external pressures toward welfare state retrenchment and pro-market reforms to which Austrian governments have been reacting. In short, the argument presented here follows largely an institutionalist account of welfare state development.

132 | Reinhard Heinisch

When at one point the veto mechanisms temporarily eroded, the pace of welfare restructuring accelerated, effecting more profound systemic changes. However, in Austria’s institutional structure and consensus culture, unilaterally initiated reforms proved hard to sustain, so that eventually the political system recentered itself, returning to bipartisanship and incrementalism. In this respect, Austria remains a textbook example of Katzenstein’s (1984, 1985) argument of how small, corporatist states can adapt to absorb or mitigate the pressures of liberalization.

The Causes of Social Policy Change: Externally Induced Transformations In Austria the federal government plays the dominant role in welfare and social policy. Its centerpiece is the social insurance system, which is funded principally by automatic payroll deductions and employer contributions but requires, nevertheless, large transfers from the national budget. Federal contributions to the pension funds (2008: €7.337 million/10.5 percent), social insurance (2008: €9.73 million/13.9 percent, including consumer protection and other social services), and health care (2008: €6.703 million/9.6 percent) are by far the largest budget items and most important areas of government spending (2008 budget: €69.869 million) (Bundesfinanzgesetz 2007/2008 Gesamtüberblick: Bundesministerium für Finanzen: 87–88). Because Austria is a federal republic, the competencies for social policy are formally divided between the federal government (social insurance, active employment policy, and family services) and the provinces, which have far fewer resources but have managed a secondary welfare system, called “social assistance,” and associated assisted living programs. Since its inception under the imperial Austrian government in the 1880s, the welfare state had maintained its fundamental and differentiated gestalt, albeit much expanded by Social Democratic governments (cf. Unger and Heitzmann 2003). In the neocorporatist Austrian political framework, the social partners are also powerful players who have the ability to influence the legislative process both at the preparliamentary and parliamentary level. In the area of social policy, the peak associations of labor and business have played an even greater role in that they are additionally responsible for the nation’s self-governing, employment-based social insurance systems. For nearly three decades after the Second World War the system remained unchanged, confirming the ultrastability of Austria’s social and economic organization. Since 1980, Austria has been affected by three major and (largely) externally induced transformations, which have had combined and separate

Austrian Social Policy Reform in the Era of Integration and Rising Populism | 133

effects on the national political process. For the Austrian welfare state, these changes had for the most part negative implications in that retrenchment, disorganization, and constraint followed decades of welfare state expansion. However, the trend in social policy was not uniformly restrictive, as there were also countervailing tendencies that bolstered and even expanded certain parts of the welfare regime. The first important transformation began in the early 1980s, when Austria came to be increasingly affected by what has been dubbed the “neoliberal renaissance” (Vogler 1993), which brought about changes in the normative orientations of key segments of the population. Specifically, on the part of many Austrians, the 1980s saw a reassessment of the relationship between the individual and the state. In the later 1970s the dissolution of the two traditional political camps of Social Democrats and Conservatives accelerated, accompanied by the sharp rise of protest movements. Growing prosperity made a larger cross section of Austrians subject to the mounting tax burden required to sustain the expansive welfare state (Plasser and Ulram 1996; Pelinka 1996). The fiscal commitments necessary to ensure continued prosperity and Austria’s relative economic autonomy from international economic trends became the major cause for defections among middle-class voters from the Social Democrats. Many of them believed that they were footing the bill for an increasingly wasteful system of social spending, which they saw as catering to special interests and political insiders (Heinisch 2007). This was ironic in light of the fact that, as Tálos (2006a) observed, the postwar expansion of Austrian welfare, especially in the 1970s, meant that an ever-larger share of the Austrian public became beneficiaries of government services: In 1948, only 63.5 percent had public health insurance; by 1980 that ratio had increased to 98 percent. By the same token, the share of GDP devoted to social expenditures rose from 17.2 percent in 1960 to 26.7 percent in 1980. Ever-larger transfers from the general budget were necessary to ensure the solvency of the social funds, which became increasingly vulnerable to demographic and economic change. Faced with a dramatic decline in the support of middle-class voters (Ulram 1994: 93), the Social Democratic Party (SPÖ) sought to compensate for its eroding electoral majority by creating political coalitions with bourgeois parties it could dominate—first with the pre-Haider Freedom Party (FPÖ), from 1983 to 1987, and, subsequently, the conservative People’s Party (ÖVP also referred to here as “Conservatives”), from 1987 to 2000. The shift in orientation and the growing popular skepticism about the political establishment created a fertile ground for the unprecedented growth of Jörg Haider’s Freedom Party, whose influence shaped Austrian politics for the following two decades. By positioning the FPÖ initially as

134 | Reinhard Heinisch

an anti-system protest party, Haider directed its campaigns against the corporatist welfare state in all its aspects. He sought to convince middleclass voters not only that their rising taxes were sustaining so-called fat cats and apparatchiks, but also that “welfare cheats” and “freeloaders” were taking advantage of a social system expanded beyond all proportion. The Freedom Party’s campaigns of the 1980s and the growing antigovernment mood created greater acceptance for a deregulatory agenda as entire groups of employees, especially public-service sector workers, came under scrutiny for their alleged accumulation of social privileges and benefits. A second major cause of welfare reform was the change in Austria’s macroeconomic environment, specifically global economic competition and European integration. Whereas annual economic growth rates in the three decades following the war averaged around 4 percent, they declined by half thereafter. Between 1984 and 1987, Austrian economic growth slowed to 1.8 percent, compounded both by considerable structural problems in the large state-owned industrial sector and mounting economic problems as the principal Austrian export markets in Eastern Europe and the Middle East dwindled (Luif 1995). Business with North America also suffered as a result of the appreciation of the Austrian currency vis-à-vis the US dollar. Furthermore, Austria was plagued by comparatively low rates of innovation (Kramer 1998), as well as by relatively high prices for consumer goods, which affected purchasing power and depressed economic living standards. Another area of economic concern was a public-service sector characterized by low levels of productivity and high costs (Schauer 1992). Chronic budget deficits and a lack of Austrian venture capital made the national economy highly dependent on capital imports. Yet, the nation’s geopolitical position and changes in the international markets threatened to direct international investment flows away from the country. It seemed that these trends would worsen if Austria remained outside the emerging single European market. By doing so, Austrian firms would be at a significant competitive disadvantage in terms of cost, market access, administrative procedures, and the bidding on contracts (Winckler 1989: 224).1 Moreover, analyses suggested that as a European Community member, Austria would increase its real GDP significantly (1.6 percent vs. 3.5 percent by the sixth year) while achieving notably lower consumer prices and higher job growth. As a result, Austrian elites came to see “full participation in the Single Market”2 as a spur to wholesale modernization and liberalization of the economy and its governance system (Seeleib-Kaiser 1999). Declining birthrates and increased life expectancy compounded the problem of financing the pension system. Between 1980 and 2003, the con-

Austrian Social Policy Reform in the Era of Integration and Rising Populism | 135

tributions to the pension insurance funds rose by 195.9 percent, whereas expenditures went up by 239 percent. By 2004 the ratio of over-65-yearolds to 15-year-olds was 1:4.2, and, according to current projections, it should continue to rise to 1:2.3 by 2030 (Tálos 2005: 53; Tálos 2006a). The excessively large public-service sector and nationalized industries, which had absorbed excess labor, mostly disappeared in the 1990s, driving up government expenditures on unemployment. During the same time, the ratio of Austrians in atypical job situations (part-time, contract, temporary, on-call work, etc.) rose to 20 percent of the workforce, 85 percent of which were women. The decline in full-time employees by some 65,000 people affected contributions to the various social insurance funds, pushing up payroll taxes across the board. Consequently, the system of rewarding stable employment and privileging traditional careers found it difficult to accommodate increasingly common new types of occupations, such as short-term or atypical forms of employment. The third transformation shaping Austrian politics was a growing sense of anxiety and instability engendered by the collapse of Communism in Eastern Europe and the war on the country’s doorstep in the former Yugoslavia. These events raised fundamental questions. The country’s neutrality and relative distance from both the communist East and capitalist West had provided a sense that in Austria market efficiency could be effortlessly combined with social protection and an interventionist state. The associated skepticism toward liberal capitalism, dating back to the middle and late nineteenth century, had been shared by Conservatives and the Left alike, and facilitated welfare state expansion. As Austrian exceptionalism gave way to normalization, its key components, such as the corporatist welfare state, were increasingly viewed with ambiguity (Pelinka 1995: 5). Aside from the question of identity, the transitions in Central and Eastern Europe (CEE) also had concrete economic and social consequences. Although an enormous boon to Austrian business and the economy as a whole, the transition of Eastern Europe had an uneven impact on the Austrian workforce, negatively affecting laborers with low skills and depressing wage levels overall. Stagnant wage growth and economic uncertainty coincided with a public debate about social cutbacks and budgetary retrenchment. The opening up of CEE markets resulted in an influx of labor migrants and refugees as well as low-wage competition when low-valueadded production was relocated to the East (BfWuS 1992). The uneven impact of economic integration and EU enlargement also served to undermine the social solidarity and cohesive nature that had characterized the social policy discourse in Austria. It also rekindled debates about welfare services not being transparent and missing their intended target groups.3 This made it possible for libertarian elements in

136 | Reinhard Heinisch

the Freedom Party and groups within the People’s Party elite to promote a new political agenda aimed at establishing a different social welfare system rooted more strongly in what was dubbed “individual responsibility” and “performance orientation” (Tálos 2006a: 631–32). In the People’s Party this programmatic change was represented, among others, by Chairman Wolfgang Schüssel and the head of the parliamentary caucus, Andreas Khol, who came to shape the ÖVP’s political agenda. Without this important area of agreement, the subsequent ÖVP-FPÖ coalition government probably would not have taken shape. Summing up, the principal causes driving change in the welfare state after 1980 were economic, as many of the latent problems became exacerbated in the new macroeconomic context and were reinterpreted in light of the new political discourse.

The Effects: The Social Policy of the SPÖ-ÖVP Coalition, 1987–1999 The externally induced pressure toward liberalization and deregulation prompted a corresponding policy commitment by the Social Democrats and Conservatives, who in 1987 formed a coalition government with the objective of modernizing the Austrian economy and addressing the growing structural fiscal deficits. In doing so, the government felt it had to enact politically costly policies in a fluid macroeconomic and geopolitical context with unpredictable consequences. To mitigate political exposure, the governing coalition took advantage of the neocorporatist system by developing coping mechanisms and trying to make reforms more tolerable. In Austria’s systems of “autonomous corporatism,” the dominant role is played by employer and employee organizations, whose actions are legitimated by the government either simply by its lack of intervention or by its incorporation of social-partner decisions into public policy. The Social Partnership’s principal members are the Trade Union Federation (ÖGB), the Chamber of Labor (AK), the Economic Chamber (WK), and the Association of Austrian Industrialists (VÖI). Except for the ÖGB, which is an extremely cohesive and vertically highly integrated labor organization representing almost all private and public sector employees, the chambers are public-law organizations with compulsory membership and quasi-governmental jurisdiction. They are strongly institutionalized and command large budgets and staffs. Comparative analyses show that these Austrian associations have nearly unrivaled organizational power due to their high levels of centralization and institutional development when compared with corresponding interest organizations elsewhere (van Waarden 1995: 55).

Austrian Social Policy Reform in the Era of Integration and Rising Populism | 137

The relative lack of sector-nonspecific labor-market associations in Austria, the small size of the country, and the comparatively small number of affiliates added to the effectiveness of interest coordination and concentration. Politically, the Social Partnership has the right to take legal initiatives, needs to be consulted by government prior to submitting draft legislation, and routinely exercises influence on other political actors through interelite contacts and by being well represented in respective parliamentary slates and party decision-making bodies. Although labor and business have had different views on social policy reform, the Economic Chamber also tends to be rather status-quo oriented and protectionist. This is because it is dominated by small- and mediumsize businesses, which are generally skeptical of liberalization and greater international competition. Consequently, political divisions between the social partners are less acute than their differing interests would prima facie suggest. Moreover, the institutional culture and operational logic of the Social Partnership is based on reciprocity and, thus, is averse to open conflict. Scharpf (1991) called this feature of Austria’s model of joint governance one where “joint mistakes [are] jointly corrected” (Scharpf 1991: 56). Despite a certain erosion of corporatist clout in the 1980s, the social partnership as such can still be considered as a soft veto player in Austrian social policy–making (Obinger 2002). Although the partnership constituents could not legally block retrenchment measures by themselves, they were de facto able to do so as long as the SPÖ either remained in power or was willing to comply with union preferences. Yet it should be noted that the Social Democratic leadership was actually divided, in that SPÖ leader and Chancellor Franz Vranitzky and his finance minister, Ferdinand Lacina, were very willing to embrace more far-reaching reforms but were blocked internally by the leftist wing of the party and the labor fraction (Heinisch 2002). Despite the new government’s stated reform goals, the welfare state continued to expand in certain areas. In 1987 and 1988, the coalition enhanced labor rights and strengthened codetermination in enterprises to enlist social-partner support for accession to the European Union and the Single Market, which was the single biggest economic challenge facing the country in decades (Heinisch 1999). It was especially in this context that the government used social policies, particularly the old-age pension and unemployment insurance system, as a means of cushioning the blow of economic change. With the most severe restructuring occurring in the public sector and nationalized industry, working with the social partners ensured that the negotiations and policy coordination remained limited to a small number of key actors. This enabled the government and the principal labor market associations to hammer out policies designed to internalize social conflicts and cushion the blows of retrenchment.

138 | Reinhard Heinisch

Deregulation reduced the number of Austrians working in nationalized industries from some 102,000 to 50,000 between 1986 and 1994. In agreement with the social partners, most of the older unemployed went into early retirement. In turn, the government’s contribution to the various pension insurance funds jumped by nearly 10 percent in one year, finally reaching more than a third of the entire national budget of ATS 619 billion (about US $60 billion). In fact, the pension system alone absorbed 70 percent of all social spending by the early 1990s (WIFO 1992: 421). Social policy expanded due to EU membership, namely by improvements in gender equality, the introduction of paternity leave, paternity payments, and equal severance pay. Another innovation was the 1997 introduction of a guaranteed leave of absence for educational purposes (cf. Tálos 2006a). The government also introduced a gender-equal wagereplacement rate of 57.9 percent of income for all the unemployed and allowed for greater flexibility for family purposes (Hochrainer 2004). Discriminatory clauses, such as those that denied women full unemployment compensation if their husbands had a full-time job, were summarily eliminated. In addition, both male and female foreign residents now became eligible for emergency social assistance (Tálos 2006a). It bears remembering that despite the continued expansion of the Austrian welfare state after the Second World War, the social policy model had remained structured as a “conservative welfare regime” (Esping-Andersen 1990). Based on the social insurance model, the Austrian welfare state linked claims and benefits in health care, unemployment, and old-age pension to employment and employability. That left those who were either unable or unwilling to participate in the labor market with relatively little social protection. Access to benefits and services was acquired either directly through contributions or indirectly through marriage or by being a dependent child. As a result, the social safety net for a significant part of the population depended on the stability of a marriage. This system had served to reinforce traditional gender roles of women as homemakers, mothers, and part-time employees, while men were seen as the primary wage earners (Tálos and Falkner 1992). The label “conservative” also describes the class-based nature of the Austrian welfare state model because it traditionally differentiated claims by social status, applying a richly layered system of preferences and entitlements. This meant that income differentials were rigorously maintained and that low levels of financial support persisted for the poor. Moreover, reforms generally had to be calibrated so that changes for one group would remain proportional to the claims and benefits of other groups. Generally speaking, although the welfare state expanded in some areas and major retrenchment was prevented in others, the government insti-

Austrian Social Policy Reform in the Era of Integration and Rising Populism | 139

tuted a series of reforms whose cumulative effect was felt by substantial parts of the population. The policy measures especially affected unemployment compensation and social assistance, which had seen sharply rising costs as a result of the restructuring of industry. For example, the government introduced new restrictions, curtailing the period within which to apply for unemployment pay. Overall unemployment compensation levels were reduced from 57.9 percent to 56 percent, and the waiting period for reapplication petitions was lengthened from 20 to 26 weeks. In a further move, the salary base used for calculating the amount of unemployment compensation was also considerably extended, which in effect lowered the actual payout. Under the new rules governing petitions for unemployment after 1993, individuals were required to report weekly to the employment office to prove they were actively engaged in a job search. Similarly, social-assistance allowances were reduced from 95 to 92 percent of unemployment pay. Moreover, transfer payments by spouses and other family members were henceforth taken into account. In fact, the first decade of welfare reform, from the mid-1980s to the mid-1990s, corresponds closely to Pierson’s observations that governments begin retrenchment processes by employing obfuscation techniques such as decrementalism, indirect incidence, burden shifting, automaticity, and lagged cutbacks (Pierson 1994: 21–22). Nonetheless, when measured by aggregate spending data, the evidence points to an overall expansion of the welfare state between 1980 and 1995. Total social spending as a proportion of GDP increased from 26.3 to 29.2 percent in 1995 (BMSG 2001: 40). As a result, by 1994, with Austria’s longterm fiscal stability no longer assured, the government’s credibility was in question. Moreover, there was every indication that Austria would not meet the Maastricht criteria and would thus be unable to qualify for the single European currency. However, when the SPÖ-ÖVP coalition government proposed a massive austerity package in 1995 and 1996, the social partners, especially labor, initially blocked the measures. In the process of Austria’s accession to the EU, the Social Partnership had, compared to a decade earlier, reclaimed political power vis-à-vis the government, as the social partners had proved indispensible in convincing a skeptical Austrian public (Heinisch 1999). However, in other respects, the overall liberalization and deregulation associated with European integration curtailed many powers of Austro-corporatism, the consequences of which were a loss of policy instruments; the curtailing of legal powers; and the transfer of regulatory competencies to the supranational level. These changes reduced the social partners’ macroeconomic leverage and restricted the state’s regulatory power (Karlhofer and Tálos 1996; Falkner 1993). A specific problem was that integration affected the major social

140 | Reinhard Heinisch

partners differently and also strengthened decentralization and disorganization at all levels. In addition, Austria’s integration into the Single Market’s Euro Zone required major adjustments in budgetary terms, as well as labor-market and economic policy. Given the growing dissent within the Social Partnership between the unions and the employers on how to rein in public spending, the institution was politically weakened. When this conflict spilled over into the government, the Social Democratic-Conservative coalition collapsed. Following early elections in 1995, the coalition was restored and moved quickly to implement the austerity program. The government, especially the Social Democrats, opted not to comply with labor’s preferences because it saw its entire Europe policy in jeopardy, especially since full membership in the Single Market was the raison d’être for joining the European Union. The fallout from the unpopular retrenchment measures continued to haunt the government coalition while strengthening the populist Freedom Party. These so-called Sparpakete (retrenchment packages) were designed to reduce the expense of the government’s own operations and personnel by some 32 percent. They also proposed cutting the federal subsidies for old age–pension insurance (by 13.5 percent), families with children (by 6.3 percent), and unemployment insurance (by 5.3 percent). Simultaneously, the government planned to raise the taxes on upper-middle-class incomes, capital gains, energy consumption, cars, and tobacco. With the final implementation of the austerity measures in 1996, the government’s social policy became substantially more restrictive. It introduced a series of cutbacks in social and educational support programs while increasing the taxes on energy. These changes also coincided with regressive tax measures, such as across-the-board income-tax increases coupled with higher rates of investment-based tax deductions, which tended to benefit higher incomes. The Geburtenbeihilfe (childbirth subsidy) was eliminated entirely, while other cuts affected maternity pay and subsidies for nursing care, as well as tax exemptions for overtime pay and family-emergency expenditures. In sum, the population perceived the combination of tax increases and social retrenchment as a substantial burden imposed upon them. As harsh as many Austrians perceived the measures, overall social spending was only reduced slightly to 28.5 percent in 1998 and the public debt to 64.9 percent of GDP. In short, while Austria at the time managed to meet the criteria for adopting the single currency, the budgetary situation was far from sustainable in the long run. Summing up, despite taking measures to restrict the welfare state, the government never seemed willing to depart programmatically or ideologically from its commitment to the Austrian social model. The retrenchment

Austrian Social Policy Reform in the Era of Integration and Rising Populism | 141

process was summarily justified in reference to budgetary constraints and new economic priorities.

The Effects: The Social Policy of the ÖVP- FPÖ/BZÖ Coalition, 2000–2006 The formation of the center-right coalition between Wolfgang Schüssel’s People’s Party and Jörg Haider’s Freedom Party in 2000 brought about a fundamentally new situation in Austrian politics. The increasing conservatism of the ÖVP, as reflected in the attitudes toward education, defense policy, immigration, and religion, was complemented by a significant neoliberal and anti-statist agenda in economic policies (Tálos 2006b). Moving quickly against traditional SPÖ policy objectives, both in education and social policy, and the removal of Social Democratic and union influence from key institutions and decision making were not only welcomed, but also considered by many Conservatives to be a payback for the long period of Social Democratic dominance (Müller 2000; Müller and Fallend 2004). Ironically, the Freedom Party, which had no stake in the corporatist system and had routinely denounced Austrian consociationalism, was actually in a more difficult position. On one hand, its supporters clearly relished the opportunity to curtail the influence of organized labor. On the other hand, the Freedom Party now championed the rights of the so-called “little people” and could ill afford to appear to give ground on questions of social competence. Eventually, the coalition partners agreed on a deregulatory agenda with clearly defined exceptions. In a break with longstanding tradition, the Ministries of Labor and the Economy were merged. Previously in the hands of the SPÖ and ÖVP respectively, the merging of the ministries symbolized a clear break with the consensus model and its allocating influence proportionally to labor and business. In the area of public policy, the center-right government had ambitious goals, such as reducing the public bureaucracy by some 13,000 positions to save nearly €1 billion and eliminating the deficit within two years. In a strategy dubbed “speed kills,” the government moved bills quickly through the legislature, packing them with amendments in order to overwhelm the opposition (Müller and Fallend 2004: Table 3: 823). Taking advantage of unorthodox procedures, the ÖVP-FPÖ coalition outmaneuvered the Social Democrats and their supporters at every turn. The ideological departure from the previous consensus on social policy was already foreshadowed in the government’s initial programmatic declaration:

142 | Reinhard Heinisch

We are of the opinion that it is a part of the essence of a socially oriented society to help those insufficiently capable of helping themselves. Modern social policy lies at the intersection between freedom and responsibility, between performance orientation and solidarity. The reordering of state and private responsibility is the biggest challenge for social policy, which is confronted with escalating claims coupled with unsustainable finances and decreasing individual social effectiveness. Only what is earned can be (re)distributed. Prevention must take precedence over care (author’s translation from Tálos 2006a: 631–32).

Without their SPÖ allies in a position to block legislation, labor found it difficult to effectively oppose the new government’s policy. The Economic Chamber, fearful about the future of the Social Partnership, signaled its disapproval of the way the government treated the unions, but Austrocorporatism remained weakened and sidelined in this initial phase of the ÖVP-FPÖ government. Unburdened by Social Democratic influence, the new coalition was free to pursue its own policy priorities following neoconservative ideological preferences. The Kinderbetreuungsgeld (infant care money), swiftly passed and implemented by July 2001, reflected the ÖVP’s long-standing demand that all mothers, whether working or not, should be entitled to equal financial compensation.4 This new commitment was remarkable, since it created a major new public expense (€400 per month for each child, for up to three years) at a time of fiscal retrenchment. In a move designed to weaken the Left in social policy–making, the Freedom Party social affairs minister sought to alter the composition of the main executive board of the Federation of Austrian Social Insurance Carriers, which was composed of twenty-seven social insurance funds. As the board had traditionally been headed by a Social Democratic union leader, this more than anything else aroused the ire of labor and represented a clear break with previous Austrian political practices. When public protests failed to make an impression on the government, the matter was taken to Austria’s Administrative High Court, which vetoed the move by ruling against the center-right coalition. Yet as the government enjoyed a two-thirds majority in Parliament, it subsequently passed legislation to override the court. On the whole, the programmatic agenda of the Conservative–Freedom Party government offered an eclectic mix of neoliberal, cultural-conservative, and populist elements. The boldest and most crucial initiative by the coalition was an ambitious drive to achieve a balanced budget by 2002, three years ahead of schedule. The social policy changes necessitated by the new budgetary priorities would have been impossible for the previous Social Democratic–dominated government. Raising the minimum

Austrian Social Policy Reform in the Era of Integration and Rising Populism | 143

retirement age (to sixty-one and a half for men and fifty-six and a half for women), introducing fees for all university students in 2001, the taxation of injury-related disability pensions, higher social insurance taxes for farmers, a reduction in unemployment benefits for single parents, cuts in the unemployment support for seasonal workers, and widely unpopular new hospital fees were some of the most controversial policy decisions made by the center-right government. The Social Democrats challenged social policy reforms in court, such as 2000 Pension Reform, new outpatient charges, and the tax on accidentinsurance benefits; however their only victory was the 2002 declaration that the taxation of pensions from accident insurance was declared unconstitutional. Outpatient charges were rescinded by the government independently a year later since they had been particularly unpopular. A petition drive supported by all opposition parties and the labor unions to include the welfare state as a principal policy objective in the Austrian constitution garnered some 717,000 signatures, forcing the government to submit it for debate in Parliament. Nevertheless, denied its position as veto player, labor was unable to make headway. After the government collapsed in 2002 following protracted internal turmoil in the Freedom Party, another watershed national election led to the quick restoration of the center-right government. The victorious Conservatives formed a coalition, first with the much-diminished Freedom Party and then, following its fracturing, with the new “Alliance Future Austria” (BZÖ)—a creation of Jörg Haider representing those former Freedom Party adherents who wanted to continue in public office. Bolstered by biggest the Conservative Party victory in thirty years and more determined than ever, Chancellor Schüssel embarked on an ambitious agenda. The social policy changes introduced during Schüssel’s second term aimed at a complete reorganization of the old-age pension system by moving away from the principle of maintaining the retirees’ material standard of living in favor of more basic social security. In a first step, the government eliminated the existing options of both an early full retirement and a gradual transition from long-term unemployment into retirement (with sufficient years of contributions). Given a legal retirement age of sixty-five (which now also applies to women retiring in 2033 and thereafter),5 the new pension system intended to provide for a flexible period between age sixty-two and sixty-eight, during which retirement becomes possible. Whereas retiring prior to the full legal age would result now in a 4.2 percent cut for every year of retirement prior to age 65, later retirement was to yield a bonus of an equal amount. By stretching the wage calculus for pension levels much further back into people’s careers to periods with lower salaries, the government achieved a proportionally

144 | Reinhard Heinisch

much lower retirement pay. Moreover, the number of years necessary to attain full pension benefits was raised from 40 to 45 years, and the maximum pension disbursement was set at 80 percent of the average lifetime income. The changes also aimed at harmonizing the pension system in the sense of standardizing the benefits regardless of the different classes of insurers. However, the contributions to the pension insurance were not equalized and still vary between blue-collar workers and employees (22.8 percent) on one hand and self-employed and farmers (17.5 percent) on the other. In the latter case, the government agreed to make up the difference. The firestorm of criticism that ensued after the government announced its plans resulted in several modifications designed to address the expected hardships.6 For the most part, however, the government ignored the unprecedented street protests and strikes and passed the legislation. Calls for a referendum were overruled, and only the parliament’s upper house was able to block the measure for several weeks by a suspensive veto. Even a Constitutional Court challenge came to nothing. These events more than anything else show the difference from a decade earlier, when labor and the Left were still able to exercise veto options. Another area of reforms affected health care policy. Here, the goals were to achieve cost savings, limiting total health expenditures to about 8 percent of GNP, and decentralization. For one thing, the health insurance funds were forced to cut costs—and thus services—because the government insisted on lowering the cap on budgetary allocations to the insurance system. In the current system, the various health insurance funds contribute some 50 percent to the financing of the Austrian health care system, compared to 16 percent by the public sector and another 34 percent by the private sector. Second, a model of performance-based financing was introduced to effect greater competition. This went hand in hand with the regionalization of both the administration and financing of the hospitals, which became the responsibility of the provincial governments. Here, the hope was that greater flexibility and competition would result in cost savings through best-practice models (Gottweis and Braumandl 2006). Despite steep fee increases, however, the insurance funds will continue to face large deficits for the foreseeable future, while the public appears extremely wary of further retrenchment so that the issue remains politically unresolved. Summing up, the most significant restructuring of the Austrian welfare state followed the ousting of the SPÖ in 2000. This opened the door for the ÖVP and FPÖ to unilaterally pursue majoritarian center-right strategies. This, in turn, weakened the principal veto players—the unions—who

Austrian Social Policy Reform in the Era of Integration and Rising Populism | 145

had exercised influence through their power within the Social Democratic Party. The policies enacted were consistent with neoliberal policy changes elsewhere, as the new government sought to reduce state agency while rewarding greater individual initiative and self-reliance. This did not amount to the kind of slash-and-burn neoliberalism many critics on the Left alleged. Nonetheless, in areas such as old-age pension and unemployment-insurance administration, the center-right coalition introduced fundamental and systemic policy changes. In those cases, there was a clear philosophical shift away from the principle of preserving the material standard of living toward a more basic social safety net.

Returning to the Past: The Political Recentering of Austrian Social Policies That the role of veto players cannot account for the entirety of Austrian welfare development is powerfully demonstrated in what happened following the 2006 national election. Although the ÖVP expected to win, the voters punished it largely for its social policy reforms. The Conservatives lost by the slimmest of margins to the Social Democrats, who were poorly prepared to form a government. Although numerically capable of renewing a center-right coalition, which Jörg Haider offered to join, the People’s Party decided to form a grand coalition with Social Democrats. The moderates within the Conservative Party thus prevailed, bringing about a recentering of Austrian politics. This outcome was a reminder that the institutionalized cooperation between labor and business and the long prevailing disposition in favor of bipartisan consensus and national cooperation makes majoritarian competition strategies between the two current political blocs, ÖVP-FPÖ and SPÖ-Green, hardly sustainable. In Austria countless editorials and public opinions surveys, as well as the thinking of party functionaries, continually reaffirm the consensual culture. As a result, the policy process returned to the incrementalism that had characterized the situation in the 1980s and 1990s. Yet the renewed grand coalition under the Social Democratic Chancellor, Alfred Gusenbauer, was short lived as incrementalism quickly turned to gridlock and fingerpointing. Profound dissatisfaction inside the SPÖ with its leadership led to the ouster of Gusenbauer in 2008 and national elections in the fall of that year. The new SPÖ leader, Werner Faymann, lost no time signaling increased skepticism vis-à-vis European integration and embracing “bread-andbutter” issues in his election campaign. While emphasizing especially the high cost of food and gasoline, he promised a greater focus on popular

146 | Reinhard Heinisch

social policy demands. A few years earlier it would have been considered fiscally irresponsible even by Social Democrats to present a €1.3 billion package of measures aimed at lessening the social and economic burdens faced by many Austrians, but these policies clearly gave the SPÖ political momentum. The Social Democrats demanded rapid improvements in home nursing care by raising government subsidies 5 percent and extending them to cases of mental disability and minors in need of permanent care. Along with proposing to rescind university fees and cut in half the VAT on all food items, the Social Democrats also advocated extending the family allowance by a month. In a sharp departure from Austrian political practice, the SPÖ also announced that it would tear up its still-existing coalition agreement and instead seek free-floating majorities with any party willing to help pass the social policy legislation as soon as possible. Populist competitors such as the Freedom Party and Haider’s BZÖ also campaigned on relieving social and economic hardship while branding the ÖVP and its party leader, Finance Minister Wilhelm Molterer, as technocratic and cold.7 In response to declining poll numbers, the Conservatives had no choice but to compromise so that the SPÖ and ÖVP surprisingly managed to reach eleventh-hour agreements on several social policy agenda items that had eluded them previously.8 When hard-line conservative groups in the ÖVP worried about backsliding into fiscal irresponsibility, the People’s Party was unable to deliver on all issues. In response, the SPÖ was able to find varying support for most of its initiatives in a memorable marathon session of the Austrian Parliament on 24 September 2008. 9 The Social Democrats and the People’s Party formed yet another grand coalition despite plunging to record lows in the following elections (SPÖ: 29.3 percent/ÖVP: 26 percent), while the two populist rivals, the FPÖ (17.5 percent) and BZÖ (10.7 percent), together captured nearly a third of the electorate. Fearful of voter anger and further gains by the populist parties, Social Democrats and Conservatives largely shied away from major social policy reform initiatives. Moreover, with the sudden death of Jörg Haider in October 2008 and the withdrawal of former Chancellor Schüssel and Deputy Chancellor Molterer from center stage, the three major right-ofcenter protagonists of retrenchment policies after 2000 were gone. In the wake of the global financial crisis, the Social Democrats pushed through levels of public spending on labor market policies and social stabilization measures not seen in a decade. Already on the defensive, the Conservatives were unable to thwart these measures and abandoned their rhetoric about the need for fiscal discipline and market-driven solutions, so prevalent in the Schüssel-Haider era. The most sweeping innovation planned by the new government and pioneered by the Ministry of Social

Austrian Social Policy Reform in the Era of Integration and Rising Populism | 147

Affairs was the planned replacement of social assistance with bedarfsorientierte Mindestsicherung or Grundsicherung (basic supplemental security income), referring to twelve installments of some €752,94 (€1,129,41 for couples) to those unable to provide for themselves or rely on third-party support. The government introduced this new program in September 2010, which has largely replaced the hodgepodge of regional social assistance programs and their widely varying funding levels. Although the implementation process is still not yet complete and some regional differences in social assistance legislation will continue to play a role, basic supplemental security income clearly represents a new chapter in Austrian social policy.

Conclusions Generally speaking, the Austrian welfare state has demonstrated an astonishing durability at the overall systemic level. Most of the changes that occurred fall into a category that Oliver Rathkolb (2005) called the “adaptation of the postwar reconstruction model.” This remarkable persistence is largely the result of structural and institutional factors that made more sweeping changes nearly impossible or politically too costly. The presence of powerful veto players in the form of the Social Partnership and their political allies is the primary factor, along with the structural bias of the Austrian model toward grand coalitions. Their tradeoffs and side payments prohibit sweeping changes and necessitate policy incrementalism. The underlying normative orientation in favor of gradualism, avoiding winners and losers, and making agreements by consensus also undercuts any attempt at achieving a radical restructuring. Moreover, the differential nature of the social insurance systems in combination with varying entitlements and benefit levels has made across-the-board cuts and one-sizefits-all solutions either impractical or too difficult to implement. It is thus not surprising that the most dramatic change in the past twenty-five years occurred when the center-right government came to power and purposefully saw its role as going against several of the established political taboos. Yet, as has become clear, the results proved politically so costly that the government did not survive the subsequent elections, and the system returned to its centrist orientation. Nonetheless, the fact that the fundamentals have not changed as dramatically as they could have should not be seen as something entirely positive. Indeed, as Tálos (2006a) suggests, the intransigence of the Austrian social system can also be regarded as a missed opportunity for restructuring the model to achieve greater inclusiveness and social equality while

148 | Reinhard Heinisch

being mindful of demographic and economic factors. About 4 percent of Austrians live in poverty, and a further 13.2 percent are in acute danger of slipping into poverty. In other words, while from a system perspective, the adjustment measures have been less severe than expected, from the standpoint of certain stakeholders, the cumulative effects of retrenchment may have been quite dramatic. In the final analysis, the Austrian welfare state has become substantially more restrictive so that the sum total of incremental cuts and changes have certainly altered the level and scope of social policy since the 1970s. Notwithstanding a trend toward greater cross-class harmonization and individualization, Austrian social policy is likely to remain wedged between the stickiness of its underlying contribution-based, differential insurance model, an increasingly populist political discourse, and growing fiscal constraints.

Notes 1. Estimates of the cost advantage enjoyed by companies in the Single Market compared to those in European Free Trade Association (EFTA) countries ranged between 1 percent and 3 percent of the value of a given good. 2. This phrase, rather than “EC/EU membership,” was the official terminology first used to make Austrians comfortable with the idea of European integration. 3. There were charges that services were provided based on the “watering can principle” (Gießkannenprinzip), meaning that they were applied indiscriminately and that “meeting the mark” (Treffsicherheit) was not assured (cf. Stenographisches Protokoll 160. Sitzung des Nationalrates der Republik Österreich XX. Gesetzgebungsperiode, 25 February 1999: 6). 4. The SPÖ had always rejected this idea (Heinisch 2002), claiming that it promoted a woman’s return to the “home and hearth” because it was an incentive for females to quit their careers. 5. Traditionally, women achieved their full retirement age at sixty. 6. After a storm of protest, a more generous solution was found for certain types of workers, the so-called Hackler-Regelung (hard physical laborer regulation). The resulting arrangements extended the time period of child care counted as a contribution toward retirement from eighteen to twenty-four months. Moreover, the government also relented by capping the expected losses in individual pension income until 2024. 7. Cf. http://diepresse.com/home/politik/innenpolitik/nrwahl/86355/index.do. 8. Cf. http://diepresse.com/home/politik/innenpolitik/354988/index.do and http://diepresse.com/home/politik/innenpolitik/nrwahl/86355/index.do. 9. It was decided to rescind university fees, substantially increase the home nursing allowance, pay a thirteenth installment of the family allowance, and defuse some of the hardship clauses in the new pension system. An initiative by the BZÖ to halve the VAT on medication found the unanimous support of all parties; however the decision of the ÖVP and BZÖ not to support the other parties in lowering the general VAT on food items thwarted one of the Social Democrats’ most ambitious and costliest goals.

Austrian Social Policy Reform in the Era of Integration and Rising Populism | 149

Bibliography BfWuS (Beirat für Wirtschafts und Sozialfragen). 1992. Ostöffnung. Vienna: Der Beirat. BMSG (Bundesministerium für soziale Sicherung und Generationen). 2001. Bericht über die soziale Lage 1999. Vienna: BMSG. Bundesfinanzgesetz 2007/2008 Gesamtüberblick. 2008. Bundesministerium für Finanzen. http://www.bmf.gv.at/Budget/Budget20072008/Gesamtueberblick_BFG_ 2007_2008.pdf Esping-Andersen, Gøsta. 1990. The Three Worlds of Welfare Capitalism. Princeton: Princeton University Press. Falkner, Gerda. 1993. “Sozialpartnerschaftliche Politikmuster und Europäische Integration.” In Sozialpartnerschaft: Kontinuität und Wandel eines Modells, edited by Emmerich Tálos, 79–103. Vienna: Verlag für Gesellschaftskritik. Gottweis, Herbert, and Elisabeth Braumandl. 2006. “Gesundheitspolitik.” In Politik in Österreich—Das Handbuch, edited by Herbert Dachs, Peter Gerlich, Herbert Gottweis et al., 753–77. Vienna: Manz. Heinisch, Reinhard. 1999. “Modernization Brokers: Austrian Corporatism in Search of a New Legitimacy.” Current Politics and Economics of Europe 9, no. 1: 65–94. ―――. 2002. Populism, Proporz and Pariah: Austria Turns Right: Austrian Political Change, Its Causes and Repercussions. New York: Nova Science. ―――. 2007. “The Structure and Agency of Populism in Austria—Its Causes and Repercussions.” In Twenty-first Century Populism: The Spectre of Western European Democracy, edited by Daniele Albertazzi, Duncan McDonnell et al., 67–83. Houndsville-Basingstoke: Palgrave. Hochrainer, Klaus. 2004. “Umverteilung der Erwerbsarbeit durch gesamtstaatliche Arbeitszeitpolitik: Ein zukunftsträchtiges Instrument der Beschäftigungspolitik.” PhD diss., University of Vienna. Karlhofer, Ferdinand, and Emmerich Tálos. 1996. Sozialpartnerschaft und EU: Integrationsdynamik und Handlungsrahmen der österreichischen Sozialpartnerschaft. Vienna: Signum Verlag. Katzenstein, Peter J. 1984. Corporatism and Change: Austria, Switzerland, and the Politics of Industry. Ithaca: Cornell University Press. ―――. 1985. Small States in World Markets: Industrial Policy in Europe. Ithaca: Cornell University Press. Kramer, Helmut. 1998. “The Current Situation and Future Perspectives.” Special edition: Austria and the European Union, Europäische Rundschau 26, no. 2: 17–26. Luif, Paul. 1995. On The Road to Brussels—The Political Dimension of Austria’s, Finland’s, and Sweden’s Accession to the European Union. Vienna: Braunmüller. Müller, Wolfgang C., and Franz Fallend. 2004. “Changing Patterns of Party Competition in Austria: From multipolar to bipolar System.” West European Politics 27, no. 5: 801–35.

150 | Reinhard Heinisch

Müller, Wolfgang C. 2000. “Das österreichische Parteien System: Periodisierung und Perspektiven.” In Die Zukunft der österreichischen Sozialpartnerschaft: Trends, Prognosen und Szenarien, edited by Anton Pelinka, Fritz Plasser, and Wolfgang Meixner, 283–305. Vienna: Signum. Obinger, Herbert. 2002. “Veto Players, Political Parties, and Welfare-state Retrenchment in Austria.” International Journal of Political Economy 32, no. 2: 44–66. Pelinka, Anton. 1995. “Die Entaustrifizierung Österreichs: Zum Wandel des politischen Systems 1945–1995.” Österreichische Zeitschrift für Politikwissenschaft 24, no. 1: 5–16. ―――. 1996. “Kammern und Sozialpartnerschaft in Österreich.” In Kammern auf dem Prüfstand: Vergleichende Analysen institutioneller Funktionsbedingungen, edited by Anton Pelinka and Christian Smekal, 11–24. Vienna: Signum. Plasser, Fritz, and Peter A. Ulram. 1996. “Akzeptanz und Unterstützung sozialpartnerschaftlicher Interessenvertretung in Österreich.” In Kammern auf dem Prüfstand: Vergleichende Analysen institutioneller Funktionsbedingungen, edited by Anton Pelinka and Christian Smekal, 225–42. Vienna: Signum Verlag. Pierson, Paul. 1994. Dismantling the Welfare State. Cambridge: Cambridge University Press. Rathkolb, Oliver. 2005. Die paradoxe Republik Österreich 1945–2005.Vienna: Zolnay. Scharpf, Fritz W. 1991. Crisis and Choice in European Social Democracy. Ithaca, NY: Cornell University Press. Schauer, Reinbert. 1992. “Aktuelle Probleme der Öffentlichen Wirtschaft und Gemeinwirtschaft.” In Öffentliche Wirtschaft und Gemeinwirtschaft, edited byVerband der Öffentlichen Wirtschaft und Gemeinwirtschaft, 77–114. Vienna: Manz. Seeleib-Kaiser, Martin. 1999. “Wohlfahrtssysteme unter Bedingungen der Globalisierung: Divergenz, Konvergenz oder divergente Konvergenz.” Zeitschrift für Sozialreform 45, no. 1: 3–23. Tálos, Emmerich. 2005. Vom Siegeszug zum Rückzug. Sozialstaat Österreich 1945– 2005. Vienna: Studienverlag. ―――. 2006a. “Sozialpolitik: Zwischen Expansion und Restiktion.” In Politik in Österreich—Das Handbuch, edited by Herbert Dachs; Peter Gerlich; Herbert Gottweis et al., 624–37. Vienna: Manz. ―――. 2006b. Schwarz—Blau: Eine Bilanz des “Neu-Regierens.” Vienna: LIT. Tálos, Emmerich, and Gerda Falkner. 1992. “Politik und Lebensbedingungen von Frauen.” In Der geforderte Wohlfahrtsstaat, edited by Emmerich Tálos, 195–234. Vienna: Löcker. Ulram, Peter. 1994. “Political Culture and the Party System in the Kreisky Era.” In The Kreisky Era in Austria, edited by Günter Bischof and Anton Pelinka, 79–95. New Brunswick, NJ: Transaction Publishers. Tsebelis, George. 2002. Veto Players: How Political Institutions Work. New York: Princeton University Press and Russell Sage Foundation. Unger, Brigitte and Karin Heitzmann. 2003. “The Adjustment Path of the Austrian Welfare State: Back to Bismarck?” Journal of European Social Policy 13, no. 4: 371–87.

Austrian Social Policy Reform in the Era of Integration and Rising Populism | 151

Vogler, Heinz. 1993. “Sozialpartnerschaft: Entwicklung, Probleme, Perspektiven.” Gemeinwirtschaft 4 (August): 17–28. van Waarden, Frans. 1995. “The Organizational Power of Employers’ Associations: Cohesion, Comprehensiveness and Organizational Development.” In Organized Industrial Relations in Europe: What Future? edited by Colin Crouch and Franz Traxler, 45–97. Aldershot Hants, UK: Avebury Ashgate Publishing. Wirtschaftskammer Österreich (WIFO). 1992.“Auswirkungen des EU-Beitrittes auf Arbeitsmarkt und Wirtschaft.” Monatshefte 8 (August): 421. Winckler, Georg. 1989. “Die Schaffung des EG-Binnenmarktes und seine Folgen für Österreich.” Österreichische Zeitschrift für Politikwissenschaft 18, no. 3: 223–30.

Chapter 7

OF FIRMS AND FLEXIBILITY The Dynamics of Collective Bargaining Reform in Spain and Portugal

( Sara Watson

Although in recent years a great deal has been written within the field of comparative political economy on the preferences of employers, the focus of most of this work has been on why we should expect employers to pursue greater flexibility in labor relations.1 The dominant literature in this vein suggests that since Fordist modes of production have given way to diversified quality production, the centralization of bargaining and other forms of “rigid” labor market institutions have become an impediment— rather than an aid—to growth. Hence, employers, faced with increasing pressures to compete in international markets, are increasingly favoring a liberalized labor market. As some scholars have pointed out, however, the absence of any clear trend toward bargaining decentralization across advanced industrialized countries has been striking (Wallerstein and Golden 2000; Traxler, Blaschke, and Kittel 2001). After nearly a decade of major international organizations such as the OECD extolling the virtues of decentralized and liberalized labor markets, one of the most dramatic changes in European political economies in recent years has been the reemergence of centralized forms of wage determination (Hassel 2003). This clearly begs the question of when and why employers make the choices that they do concerning centralized wage negotiation. Under what conditions do employers cooperate with unions to protect centralized bargaining, and under what conditions do they choose to repudiate such systems? When presented with the opportunity to flexibilize labor relations, why might employers choose to maintain the status quo? This chapter examines the recent efforts of Spanish and Portuguese governments to promote more extensive

154 | Sara Watson

firm-level collective bargaining by empowering employers over unions, thereby shedding light on when and why employers might choose to support or repudiate centralized forms of collective bargaining. By the late 1990s, governments in both Spain and Portugal had come to similar conclusions about the problems plaguing their collective-bargaining systems: they were overly centralized and suffering from excessive state regulation. The stated position of governments in the two countries was that bargaining should be decentralized so that firms could be given greater latitude to use wages to reward productivity and so that they could actively negotiate a reorganization of work processes. The responses of employers to these common governmental efforts to encourage the liberalization of collective bargaining, however, have been quite different. Whereas Portuguese employers took the opportunity presented by government action to declare war on the existing system and are currently promoting a shift to a model of “disorganized decentralization,” Spanish employers chose not to attack the existing system. Instead, what we have seen in Spain in the past ten years is the maintenance of the status quo with respect to collective-bargaining institutions. Sectoral bargaining at the provincial level remains the norm, and unions and employers have largely refrained from implementing major changes relating to internal labor-market flexibility. In accounting for the divergent responses of Spanish and Portuguese employers to the liberalizing opportunities granted them by their respective governments, this chapter advances two related claims about employers and the politics of collective bargaining. First, in contrast to existing literature, which tends to focus on employers’ need for flexibility along a single dimension—typically, the need to use salary dispersion in order to reward productivity gains—I argue that there exist several dimensions of flexibility, including internal, functional, external, and salary flexibility.2 Although employers are likely to care about all of the dimensions of flexibility just mentioned, quite sensibly, they may also be willing to trade off greater flexibility in one dimension for areas that they care more about in another dimension. Indeed, these multiple rigidities explain the divergent collective-bargaining outcomes in Iberia. In Spain, unions were able to capitalize on the fact that employers enjoyed very little flexibility in two key areas—macroeconomic wage moderation and firing costs—and offered employers tradeoffs in these areas in exchange for leaving the existing system of collective bargaining untouched. In Portugal, where employers already enjoyed a fair amount of flexibility in these areas, employers were much less willing to compromise in the area of collective-bargaining reform. Thus, whereas Spanish employers entered into a bargain of political exchange

The Dynamics of Collective-bargaining Reform in Spain and Portugal | 155

with unions in which they traded off more flexibility in one area (external flexibility) for limitations on flexibility in another (internal flexibility via the decentralization of bargaining), Portuguese employers pushed aggressively for bargaining decentralization. The remainder of the chapter proceeds in four steps. The next section shows that explanations of collective bargaining that focus on employer interests and social pacts fall short in accounting for the divergent fate of collective bargaining in Iberia. Sections three and four describe the common effort of Iberian governments to shift to a more decentralized system of collective bargaining and highlights how the many rigidities in the Spanish political economy made possible the emergence of political exchange between employers and unions, in which the maintenance of centralized bargaining structures and limited firm-level flexibility in the reallocation of labor was traded for greater external and macroeconomic wage flexibility. In contrast, the relative absence of labor market rigidities in Portugal meant that the fodder for political exchange did not exist, and thus employers were committed to bargaining decentralization. The last section concludes with some implications that this analysis has for our understanding of employer preferences and of the “new” politics of social pacts.

Collective Bargaining in Iberia: Problems and Proposed Solutions Recent work on labor market reform in Iberia has tended to focus more on the form than the content of labor market reforms. An excellent indicator of this is the explosion of work during the past ten years on the resurgence of national “social bargaining” in the two countries (Rhodes 2003; Royo 2002; Hamann and Kelly 2007). Despite the scholarly interest in the emergence of more cooperative forms of relations between the state, business, and labor, very little has been written on the substance of the reforms undertaken—and in particular, why, despite the common emergence of social bargaining writ large, we have seen such profound differences in the substantive outcomes emerging from social bargaining in the two countries, especially in the realm of collective bargaining.3 Indeed, the substance of changes in collective bargaining is striking given the two countries’ similar starting points. By the mid-1990s, governments in both countries had come to the conclusion that existing structures of collective bargaining were outdated. Rigid and highly fragmented occupational classifications codified in law were inhibiting reorganization of work processes that would increase productivity. The stated position of

156 | Sara Watson

both governments was that the social partners needed to be encouraged to widen the scope of collective bargaining. Not only should wages be negotiated at the firm level, but employers and unions should also be willing to bargain over questions related to work organization. Not surprisingly, given such a diagnosis, governments in both countries attempted to inject greater “flexibility” into their outdated collective-bargaining structures by introducing major reforms to labor law. Table 7.1. Key features of collective bargaining in Spain and Portugal, 1980s and 1990s Definition Sectoral Bargains with Extension Mechanisms

Sectoral bargaining predominates. State extends collective agreements to nonunion members. Unions and employers may choose at what level to bargain, but lower-level agreements may only improve on the contents of agreements made at a higher level.

Fragmentation of Bargaining Units

Involves many bargaining units, with the distribution of responsibilities between them left undetermined.

Strong Regulatory Role of State

Labor law limits collective bargaining over most aspects of work organization, such as occupational classifications and internal mobility.

The remainder of this section lays out the nature of changes proposed to each collective-bargaining system in recent years and what was at stake politically with these reforms. I then highlight how despite similar efforts by governments in both countries—both of which aimed at creating a more decentralized and flexible system of labor relations—employers in the two countries responded very differently to the new set of incentives facing them. Whereas Spanish employers have maintained tacit support for an intermediately centralized and relatively unarticulated system of collective bargaining and a pushed for only a limited expansion of the scope of collective bargaining, Portugal is seeing a trend toward “disorganized decentralization,’ in which there has been a shift toward more decentralized bargaining without any concomitant increase in coordinating capacity on the part of peak-level actors.4

Collective Bargaining in Iberia: The Diagnosis Throughout most of the twentieth century, the structures of collective bargaining in Iberia followed broadly similar trajectories. During their fortyyear authoritarian interregnums, Spain and Portugal were textbook cases

The Dynamics of Collective-bargaining Reform in Spain and Portugal | 157

of state-corporatist systems of labor relations.5 Not surprisingly, the onset of democracy brought dramatic changes to both systems, as the old corporatist vertical syndicates were dismantled and freedom of association—as well as freedom from obligatory association—were mandated. Broadly speaking, collective bargaining in the two countries has had three key features: the predominance of sectoral collective agreements; the fragmentation of bargaining structures; and limitations on the extent to which the regulation of internal labor markets was subject to collective negotiation. In both Spain and Portugal features of the legal framework governing labor relations added up to a system of labor regulation which, in the eyes of both the government and employers, was overly centralized and insufficiently capable of delivering microeconomic flexibility. Sectoral bargaining predominated, and although contract coverage was high, bargaining was highly fragmented and the division of labor between different levels of bargaining units was insufficiently specified. In addition, the continued existence of extensive state regulation of the labor market was allegedly blocking much-needed reforms in areas relating to microeconomic flexibility, such as the reregulation of occupational pay grades and working time. This, then, was the institutional framework that became increasingly contested and that governments in both countries targeted through major legislative initiatives after the mid-1990s. By removing the state from regulating large areas of labor relations, both governments hoped to encourage a wave of bargaining between employers and unions that would lead to a revitalization of the labor market.

Common Solutions, Divergent Outcomes The broad purpose of the labor market reforms promulgated by Iberian governments was to remove the state’s role in regulating large swaths of labor regulation—and, in so doing, to create a virtuous circle of economic dynamism in which the social partners (rather than the state) would now take the lead in negotiating new forms of microeconomic flexibility. That said, there are many ways in which the state can remove itself from the business of regulating labor; it may do so in a way that leaves existing inequalities between bargaining partners untouched, or it may try to level the playing field. Thus, before proceeding further, it is worth highlighting in general terms what was at stake for unions and employers with the new labor-relations regime being pushed by the Iberian governments. In essence, there were two issues in play; these are highlighted in Figure 7.1 (below). First, both the Spanish and Portuguese governments were attempting to decentralize bargaining away from the existing system of centralized collective bargaining, which privileged supra-firm (sectoral)

158 | Sara Watson

agreements, to a system dominated by what Wolfgang Streeck (1984) calls “microcorporatist’ bargaining. Microcorporatist bargaining refers to bargaining between unions and employers at the level of the firm rather than at the industry or national level. The system is still considered “corporatist” insofar as workers have collective representation via an enterprise union or works council,6 but by limiting the extent to which firm-level agreements are conditioned by central agreements, workers are encouraged to identify with the needs of their individual firms rather than the broader collectivity of labor. Second, the reforms also entailed a broadening of the purview of collective bargaining. This was to be achieved by revoking those extant laws that codified or limited bargaining over nonwage components of the employment relationship, such as work organization, functional flexibility, and the like. This tradition of having the state legislate these areas of the employment relationship (rather than allow the relevant parties to bargain over them directly) had emerged during the authoritarian era and had its origins in the regime’s efforts to limit class conflict. The underlying logic was that if the state guaranteed to workers a series of labor rights, then there was no need for trade unions; if the state was delivering to employers social peace and other policies that would deliver growth, then there was no need for employers to negotiate with workers. Thus, because these features of Iberian labor relations had previously been dictated by legislative fiat, the new labor-law reforms meant that if these issues were to be addressed, they must now be done so through collective bargaining. Level of Collective Bargaining

Content of Collective Bargaining

Supra-Firm Bargaining Sectoral bargaining predominated

Limited Scope of Collective Bargains Limited focus on salaries

Firm-Level Bargaining Greater freedom for firms to delink from sectoral bargains

Scope of Bargaining Widened New focus: functional and geographic mobility; working time, etc

Figure 7.1. Proposed changes to collective bargaining in Spain and Portugal

The Dynamics of Collective-bargaining Reform in Spain and Portugal | 159

These two reforms—which attempted both to reorganize the level at which collective bargaining takes place and also to expand the content of those bargains—were meant to go hand in hand. By removing the state from its role in regulating labor relations and by granting individual companies greater power to engage in bargaining, both the socialist government in Spain and the center-right government in Portugal hoped to promote microeconomic flexibility and, with it, enhanced economic competitiveness. It is noteworthy, however, that although the goal of the reforms was, in theory, to “strengthen” collective bargaining, the unions in both countries interpreted these measures as a dramatic effort to weaken labor rights in general and union confederations in particular. In the unions’ view, these reforms threatened their very organizational survival, because if employers chose to derogate from the existing sectoral agreements and to negotiate issues at the firm level, the higher-level union organizations would find themselves disconnected from individual workers at lower levels— and hence, irrelevant. This was of special concern because of the absence in both countries of a strong trade-union presence in the vast majority of small firms, which dominated both countries’ economic landscapes. Thus, unions argued, the governments’ reforms, although couched in neutral language that emphasized the need to encourage economic dynamism through more societal bargaining, in fact favored employers over unions. Surprisingly, despite similar efforts by governments in both countries, employers in Spain and Portugal responded very differently to the new set of incentives facing them. In Spain, employers chose not to act on the newfound powers granted to them by legislation and instead defended the status quo. With respect to the level of bargaining, Spanish employers chose not to exercise their right to bargain at the firm level. The vast majority of Spanish workers continue to be covered by multiemployer (industry/sectoral) agreements; indeed, the number of workers covered by company agreements has actually declined since the early 1990s. Nor has the content of bargaining changed dramatically. Rather, with respect to the broad swath of issues related to working conditions, Spain has witnessed remarkable stability. A report by Spain’s Ministry of Labor, for example, reports that the treatment of working time and the internal organization of the firm in matters relating to the use of technology and human resources recorded almost no changes in the 1996–2006 period (Comisión Consultiva Nacional de Convenios Colectivos 2008). In contrast, Portuguese employers immediately took advantage of the new legislation and launched a frontal attack on the existing system of collective bargaining.

160 | Sara Watson

The Political Economy of Collective Bargaining Reform: Multiple Rigidities and Political Exchange What accounts for the divergent strategies adopted by employers in the two countries? At first glance, the answer to this question is not straightforward. Union opposition to the changes proposed by governments, which were viewed as incursions on workers’ existing collective rights, is less surprising than the divergent positions taken by employers. Indeed, the most common understandings of employers found in the contemporary comparative political economy literature failed to predict the divergent strategies Iberian employers adopted. First, Spanish and Portuguese employers did not appear to have different “prestrategic” preferences when it came to collective-bargaining reform. On the contrary—prior to the reforms, they had long been calling for the further decentralization of bargaining and a concomitant widening of the subject matter over which they could bargain. Furthermore, accounts focusing on the relative strength of employers also fail to explain the difference. This is because, in conventional terms, Spanish employers were “stronger” than their Portuguese counterparts. Here, I follow Martin and Swank (2004) in conceptualizing employer strength as the centralization and concentration of representational power in national associations. Whereas Spanish employers are represented by a unitary employers’ confederation that represents 80 to 90 percent of Spanish firms, Portuguese employers are split between two major confederations whose representative capacity is unclear (Traxler, Blaschke, and Kittel 2001) and who often disagree on fundamental matters of policy making (Rhodes 2003). Clearly, then, the different choices made by Spanish and Portuguese employers cannot be blamed on the weakness of Spanish employers. Nor can employer-centered accounts focusing on the varying production needs of different types of firms account for the different strategies of Iberian employers. Scholars working in the “varieties of capitalism” tradition, for example, link employer support for labor market “rigidities”— including centralized wage bargaining—to their need for skilled labor forces. However, insofar as the vast majority of Iberian firms are small and medium enterprises competing on cost rather than quality (OECD 2004), we should have expected employers in both countries to support a liberal, decentralized model of collective bargaining.7 If the employer-centered literature is of little help in accounting for the divergent actions of employers, so too is recent scholarship on the politics of social pacts and “social bargaining.” This important literature has focused on the way in which macroeconomic pressures emanating from the European Monetary Union (EMU) contribute to bargaining recentral-

The Dynamics of Collective-bargaining Reform in Spain and Portugal | 161

ization via “social pacts” that address a broader range of issues, including deficit reduction and welfare state reform, than do traditional wage agreements (Pochet et al. 1997; Hassel 2003). The logic of this argument is that because the European Central Bank’s nonaccommodating monetary policy is incompatible with low unemployment in an intermediately centralized wage-setting system, the shift toward hard monetary policy attendant with European monetary unification contributes to interassociational consensus around centralized forms of wage determination. In a similar vein, the monographic literature on Iberian labor relations attributes the return of social bargaining in both countries to employers’ recognition that with increasing economic integration, social consensus and trust between employers and unions was even more important (Royo 2002, 2007). According to this argument, because future economic growth in a context of greater internationalization depends on the ability of the national system of labor relations to deliver collective goods, both Portugal and Spain have seen a shift toward more coordinated forms of interaction between unions and employers, exemplified by social pacts. And yet, neither set of explanations relating to social pacts can adequately account for the Iberian divergence. First, the combination of sectoral, relatively fragmented bargaining structures in the shadow of EMU clearly does not explain the different stances adopted by Spanish and Portuguese employers with respect to labor market reform. Both countries had highly fragmented and “unarticulated” bargaining structures, but this fact did not keep Portuguese employers from abandoning “consensus politics” and “social pacts” in the area of collective bargaining. Clearly, fragmentation alone fails to provide sufficient inducements to employers to maintain centralized “articulated” bargaining. Similarly, accounts based on the imperatives of globalization and countries’ position in the international division of labor also fail to explain the Iberian divergence, as the Portuguese economy is, if anything, more open than the Spanish economy, and yet this has not pushed employer-union relations with respect to collective bargaining in a more consensual direction. Moreover, such an approach cannot explain Spanish employers’ lack of concerted action in areas related to the content of bargaining. By contrast, the argument made here is that broad economic factors such as firms’ product market strategies and macroeconomic imperatives have played a less crucial role in shaping employer preferences in Iberia than the way in which the balance of rigidities and flexibilities within a given economy shaped the possibilities for bilateral forms of “political exchange” between employers and unions. As just discussed, employers in Portugal and Spain were initially relatively similar in their demands vis-àvis the need to reform collective-bargaining structures. Rather, it was the

162 | Sara Watson

broader labor market setting that they faced that differed, and it was this setting that accounted for the different strategies they adopted. Although firms may desire increased flexibility in matters relating to the supply side of the economy, this is only one dimension of flexibility; there are many others. The more a labor market is characterized by rigidities in multiple arenas, therefore, the more likely it is that one will see the emergence of reform coalitions between employers and unions centered around a strategy of “vice into virtue.” This phrase, coined by Jonah Levy (1999) with respect to the politics of welfare reform in Bismarckian systems of social protection, refers to a political strategy that targets policy inequities that are a source of economic inefficiency. By attenuating inequities (“vices”), reformers may be able to extract resources with which to pursue a variety of “virtuous” objectives. This chapter argues that such a dynamic is not limited to the politics of the welfare state but also may apply to industrial relations. In Spain, as we shall see, the many pathologies of the Spanish labor market made employers willing to limit collective-bargaining reforms in exchange for greater flexibility in terms of salary and external flexibility—that is, wage moderation and dismissal costs. In contrast, Portuguese employers, facing economic pressures for greater firm-level flexibility in the allocation of labor but operating in an otherwise liberalized labor market, had very little to gain from maintaining a centralized system of bargaining.

Trajectories of Collective Bargaining Reform in Iberia: Continuity Versus Change As the following pages will highlight, despite similar efforts by Iberian governments—both of which aimed at enabling employers to promote a more decentralized and flexible system of labor relations—employers in the two countries responded very differently to the new set of incentives facing them. Each country study in this section first lays out the socioeconomic situation that precipitated Iberian governments to legislate major changes to their collective-bargaining systems. I then highlight the extent of continuity in Spanish labor relations in contrast to the degree of change in Portugal and argue that the different outcomes are due largely to the different combinations of rigidities that existed in each country’s labor market at the time of the legislative reforms. These different sets of rigidities, broadly conceived, conditioned the possibilities for political exchange between employers and unions and the willingness of employers to compromise on the issue of collective-bargaining reform.

The Dynamics of Collective-bargaining Reform in Spain and Portugal | 163

Multiple Rigidities and Exchange: The Maintenance of the Status Quo in Spanish Collective Bargaining The first effort to seriously address these alleged shortcomings in Spanish collective bargaining in the postauthoritarian period came in 1994. By the early 1990s—after an economic upturn in the late 1980s in which overall employment levels increased although unemployment levels remained high—Spain was facing a recession of truly critical proportions. With unemployment at 24 percent, low overall employment rates, public debt, and high deficits, the socialist government decided that it was imperative to undertake a major structural reform of the labor market. According to the government, the purpose of their 1994 labor reform was twofold: first, to reduce several obvious institutional rigidities in collective bargaining; second, to strengthen the representational capacity of the social partners in order to promote the adaptability of the labor market. In concrete terms, this meant eliminating the old labor ordinances that had regulated many aspects of labor relations and requiring that what had been codified by law now be explicitly negotiated by the social partners. The government targeted this as a major issue for economic competitiveness because the existing narrow job demarcations inherent in the labor ordinances discouraged functional mobility. The government also promoted the decentralization of bargaining to the firm level by permitting opt-out clauses that allowed firms to modify wage increases that had been agreed to in sectoral wage settlements. In effect, this eliminated the principle of “general efficacy” (that is, the automatic extension of collective contracts to all workers in a given sector) from Spain’s Workers’ Statute. Finally, the legislation also gave employers greater control over functional and geographic mobility of their workers. Initially, the Spanish employers’ association, the CEOE, publicly pronounced itself satisfied with the legislation; its only complaint was that the reforms did not go far enough (El País, 18 November 1993). In particular, the CEOE felt that the government’s proposal was insufficient insofar as it failed to eliminate the automatic renewal (“ultra-actividad”) of the nonwage components of collective agreements. In contrast, the unions were outraged at the government’s proposal. Despite the assurances of party leaders that the reform would make Spain’s labor relations framework “more similar to that of Germany than that of Ghana” (El País, 5 February 1994), the unions were not reassured. In their view, the proposed reforms entailed a simultaneous cut in workers’ basic rights and a thinly veiled attempt to push back the frontiers of labor’s collective resources. Although Spanish unions had long agreed on the need to reform the existing system of collective bargaining, in their view this meant better “articulating” bar-

164 | Sara Watson

gaining by more clearly defining the division of labor across the hierarchy of bargaining units and strengthening the national confederations vis-àvis regional and provincial organizations. The 1994 reforma laboral, however, was another matter entirely, one whose effects threatened to empower employers over unions. Especially alarming was the effort to shift to a microcorporatist model of collective bargaining. By mandating that firms be allowed to opt out of suprafirm agreements and by legalizing the use of variable, productivity-linked wages at the company level, the government was indeed rearticulating collective bargaining but in a way that unions felt weakened the labor movement as a whole. By not including any legal provisions explicitly requiring employers to renegotiate the now-derogated labor ordinances, the reforms threatened to unleash a deregulatory spiral across the Spanish economy. Given this scenario, one might have expected the employers to press their advantage and to lead the deregulatory charge. Instead, however, they chose not to take advantage of either the opportunity to expand firmlevel bargaining or to force a renegotiation of important issues of internal flexibility. In the intervening years, little has changed in terms of the level or the content of collective bargaining. With respect to the level of bargaining, Spanish employers chose not to exercise their right to bargain at the firm level. As Figure 7.2 (below) highlights, the number of firm-level contracts signed has remained constant, and the vast majority of Spanish workers continue to be covered by multiemployer (sectoral) agreements. Indeed, the percentage of Spanish workers covered by firm-level contracts has actually decreased since the first attempted reform of the mid-1990s. Furthermore, although the 1994 legislation had in theory made it easier for firms facing economic hardship to opt out of collective agreements (descolgarse), the sectoral representatives of the employers’ association agreed in collective contracts to very strict standards for specifying the conditions of disconnecting from the broader collective agreement, thereby limiting any significant expansion of firm-level bargaining.8 Nor has the employers’ association pushed hard for firm-level productivity bargaining via efforts to “articulate” collective bargaining (Confederación Española de Organizaciones Empresariales 2008). If the level and structure of collective bargaining in Spain have remained relatively constant, so too has the content of that bargaining. Spanish employers, despite their complaints, have shown themselves willing to defend the status quo. Although the number of new clauses being included in collective contracts has increased, a fact that might suggest the emergence of “autonomous” collective bargaining between the social partners, when one examines the actual contents of such clauses, it is clear

The Dynamics of Collective-bargaining Reform in Spain and Portugal | 165

that very few flexibilizing changes have been introduced. The Comisión Consultiva Nacional de Convenios Colectivos, in its detailed analysis of Spain’s National Register of Collective Agreements, show that the distribution of working time has remained constant, while productivity-related wage incentives have increased only slightly (CCNCC 2008). In summary, despite the best efforts of the socialist government, very little related to fundamental issues of the wage structure or functional flexibility changed in the aftermath of the 1994 reforms. Figure 7.2. Percent of Spanish workers covered by firm-level contracts, 1990–2006

Why has there been so little substantive interest on the part of Spanish employers in leading a deregulatory charge with respect to collective bargaining and labor relations? As was mentioned earlier, the reticence of Spanish employers’ in pushing for deregulatory changes is not the result of inherent opposition to the substance of the legislation; the 1994 legislation actually attempted to implement reforms for which the employers’ association had long been calling.9 Rather, in order to understand their response, one needs to look beyond the realm of collective bargaining and consider other dimensions of flexibility valued by employers. In Spain, employers faced a labor market beset by a host of rigidities. Although they opposed the rigidities inherent in the collective-bargaining system, in the mid-1990s this was not their only, nor their largest, concern. Especially problematic from the perspective of Spanish employers were two other sets of rigidities: the absence of wage moderation among permanent workers and Spain’s very high dismissal costs. Employers were, therefore, quite willing to trade greater “flexibility” in these areas and, in exchange, limit a deregulation of collective bargaining. The political exchange between unions and employers emerged in two periods. After the 1994 labor reforms were announced, the unions initially

166 | Sara Watson

tried to threaten the government by organizing a general strike. When this failed to convince the Socialists to retire the offending legislation, the unions changed their tactics, now turning their attention to employers. Here, the union confederations let it be known that if the contents of the legislative reform were actually applied, not only would they bargain aggressively in the next round, but they would also do little to contain any localized uncoordinated conflicts that might arise (El País, 17 June 1994). If, however, employers were willing to maintain the existing system, the unions would offer significant wage moderation—on the order of two to three percentage points lower than their desired starting point.10 Thus, although the CEOE supported the government proposals, the possibility that they could obtain wage restraint absent a spiral of escalating conflicts led them to reconsider their position. In the late 1980s and early 1990s Spanish employers had not been terribly successful in convincing unions to moderate their wage demands. Even with unemployment at 18 to 24 percent, real wage increases nevertheless continued on an upward trend. Now, however, the unions were willing to moderate wages significantly while at the same time limiting social unrest; therefore, rather than moving ahead with bargaining decentralization or renegotiating the labor ordinances, the CEOE came to a tacit agreement with the unions. Employers would not take advantage of the new legal rights granted to them—in particular those allowing derogation from sectoral agreements and increased flexibility of working conditions.11 In exchange, the unions agreed to moderate wages and to keep a lid on social conflict (Bentolila and Jimeno 2003). If the promise of wage restraint and the threat of localized wage conflict served initially to keep employers in check, however, this in itself was insufficient to sustain a long-term equilibrium. Although the unions did indeed meet their commitment by limiting both social conflict and their wage demands, there nevertheless existed significant support among employers for the employers’ organization—the CEOE—to abandon its accommodating strategy with the unions (Molina Romo 2005). It was here that we saw the emergence of a second stage of political exchange in Spain. In order to keep the employers’ association from reneging on its informal bargain with the unions, the unions chose to sweeten the existing bargain of political exchange with a series of inducements to employers. The main arena in which this exchange occurred was with respect to the regulation of firing costs. Spain had partially liberalized temporary employment in 1984 while maintaining very high dismissal costs for workers on permanent contracts. Employers responded by taking advantage of the new laws, and by 1990 the proportion of the Spanish labor force on fixed-term contracts was over 30 percent. Workers hired under temporary contracts could be fired at a low cost relative to workers on indefinite contracts, and

The Dynamics of Collective-bargaining Reform in Spain and Portugal | 167

hence they bore the brunt of economic insecurity within the Spanish labor market (Güell and Petrongolo 2005). Although the unions had long been calling for reductions in this “precarious” form of employment, they had until this point been unwilling to acknowledge that there might be any relationship between high firing costs for workers on permanent contracts and the growing army of temporary workers. Throughout the 1980s and early 1990s, when employers demanded a liberalization of firing costs in order to promote employment creation, the unions refused out of hand to consider any of the proposals. Instead, they countered with demands that employers simply reduce their reliance on temporary contracts, absent any change in the legal framework. But, by the late 1990s, the unions were faced with the distinct possibility that employers might pull back from the existing gentlemen’s agreement with respect to the regulation of labor relations. Thus they decided to use the issue of dismissal costs as a way to lure employers into maintaining their accommodating position vis-à-vis collective bargaining. The 1997 bilateral agreement signed between unions and employers, known as the AIEE (Acuerdo Interconfederal para la Estabilidad del Empleo—Interconfederal Agreement on Employment Stability), represented the first time since 1980 that dismissal costs for permanent workers were lowered—and, moreover, they were lowered by some 25 percent. In exchange, the unions received two sets of policies. In the first agreement, the AICV (Acuerdo Interconfederal sobre la Cobertura de los Vacios—Interconfederal Agreement on the Coverage of Gaps), the existing gaps in coverage caused by the derogation of the old labor ordinances were closed. In a second, the AINC (Acuerdo Interconfederal para la Negociación Colectiva—Interconfederal Agreement on Collective Agreements), the social partners agreed to rationalize collective-bargaining structures in the manner favored by the unions (Molina Romo 2005). This breakthrough reform has since been followed by a series of bilateral agreements between the CEOE and unions in which they agreed to extend these agreements about bargaining articulation, external flexibility, and measures to promote employment creation. The backbone of these agreements, however, is that in sectors with a weak union presence, the existing structure and content of collective bargaining remain untouched. It remains to this day the third rail of Spanish labor politics.

The Portuguese Counterfactual: Employers’ Frontal Attack on Collective Bargaining As Ira Katznelson (1986) has eloquently noted, without a comparative vantage point, the “silences” as well as the “noises” in politics and policy making may appear natural, and thus unworthy of explanation. In Spain,

168 | Sara Watson

as we just saw, despite the friendly environment provided by the state, employers were willing to trade reforms in collective bargaining for wage moderation and rollbacks on firing regulation. In Portugal, in contrast, we will see that with employers, who already enjoyed significant flexibility when it came to wage-setting and firing costs, this type of political exchange failed to emerge. Employers wasted little time in taking advantage of the opportunities provided by government reforms to declare war on the existing system of collective bargaining. By the mid-1990s, proposals for amending the Portuguese system of collective bargaining had come to the forefront of the political agenda.12 Much of the pressure came from employers, whose public position was that they wanted to improve Portugal’s ability to compete in the global economy. In 2001, the new secretary of state for labor and vocational training, António Dornelas, noted that although the number of workers covered by collective bargaining remained high, the system had not adapted to the changing economic environment facing Portuguese firms. Issues such as productivity-related pay, variation in the organization of the workweek, and access to training also needed to be explicitly negotiated if Portugal was to find the proper balance between flexibility and security required by the new global economy (Dornelas 1999). The Portuguese Confederation of Industry (Confederação da Indústria Portuguesa, CIP) agreed with this reading of the problem, adding only that collective bargaining based on sector-level negotiations, as was currently the case, was an outdated model. Portugal needed more firm-level bargaining (International Labour Organisation 2001). The 2003 reform to Portugal’s labor code, passed by a center-right government, relied on somewhat different mechanisms than the Spanish reform, but it shared with Spain’s labor market reforms the end goal of decentralizing collective bargainiπng and injecting greater flexibility into labor relations. The reform operated in two key areas. First, the new labor code abolished a feature of labor law known in Portuguese as sobrevigência, which guaranteed the validity of collective agreements until their substitution by a new agreement signed by the same partners. The new reformed labor code now specified that collective agreements could expire under one of two conditions: if they were denounced by one of the signing parties (unions or employers) or if they were not formally renewed via a new collective agreement. Under the new regime, once the validity of a given collective agreement was denounced or expired, employees who had been covered by that contract maintained their respective rights and benefits as part of their individual work contracts. Workers hired after the expiration of the collective agreement, however, now came under the protection

The Dynamics of Collective-bargaining Reform in Spain and Portugal | 169

of the labor code, which guaranteed lower standards than the collective agreements (EIRO 2004). At the same time, it granted firm-level workers’ commissions the right to negotiate collective contracts, while severely circumscribing the legal rights granted to these workers’ commissions. The law placed considerable limitations on the participation rights of works’ councils by abolishing legal protections for their members (which had existed under the previous legal regime).13 This meant that employers wishing to negotiate firm-level flexibility could now bypass the national-level unions and deal directly with organizations over whom they had greater control. In sum, the new legislation resulted in the most sweeping changes to Portuguese collective bargaining since the transition to democracy. In contrast to Spain, however, where employers chose not to push their advantage, Portuguese employers wasted no time in declaring war on the extant system of collective bargaining. In the 2004 bargaining round, a large number of employers invalidated their existing collective agreements by taking advantage of their newfound right to formally denounce existing contracts. As a result, the number of agreements in effect fell to less than half of the average number in previous years, while the percentage of workers covered by renewed agreements fell to approximately 40 percent. This had the effect of enabling individual employers to push for sweeping changes in work organization and working time. In 2005 and 2006, collective bargaining recovered somewhat, but both the number of new agreements and the percentage of workers covered were still well below previous levels (UGT 2007). In the absence of sectoral agreements, unions report a clear increase in the proportion of company-level agreements within the collective-bargaining system (UGT 2006). Furthermore, despite the slow reemergence of collective negotiations in some sectors, bargaining remains at a standstill in several important industries, including ceramics, textiles, automotives, chemicals, metallurgy, and private health care (Campos and Naumann 2006). Although the outcome of the current war between employers and unions is still uncertain, there is broad agreement among Portuguese scholars that in the long run, the system is headed in the direction of further decentralization and that the hierarchical order between collective agreements and plant-level negotiations has been reversed. Moreover, although one is seeing the emergence of plant-level pacts aimed at expanding firms’ flexibility,14 it would be somewhat misleading to claim that this represents a genuine expansion of the scope of bargaining. To the extent that workers’ representative organs at the firm level have been weakened in Portugal, it is unclear how much—if any—meaningful bargaining is taking place.

170 | Sara Watson

Portuguese employers were much more focused on the benefits to be derived from bargaining decentralization—namely, the ability to restructure the organization of production and to restructure working time—than were their Spanish counterparts. This was because they faced fewer other rigidities in their labor market. Consider the two key areas in which Spanish unions were able to offer inducements to employers: macroeconomic wage restraint and the renegotiation of external flexibility. Neither of these avenues of negotiation was open to Portuguese unions. With respect to broad-based wage restraint, this was not an option because collective bargaining in Portugal has never had a large impact on the wage policies adopted by employers in the first place. For example, as the OECD has noted, although Portuguese unemployment increased between 1992 and 1996, the shift to nonaccommodating monetary policy that came with EMU did not have the same dramatic unemployment effects as in some other countries.15 Given this situation, macroeconomic wage inflation was not a major concern for Portuguese employers. This removed one potential resource available to unions. Second, Portuguese unions also had much less to offer employers in terms of other “vices” that might be remedied in exchange for a “virtuous” outcome in the collective-bargaining arena. In Spain, employers were sufficiently unhappy with the limitations on external flexibility that they were willing to trade concessions in this area for limitations on bargaining decentralization. In Portugal, however, this was not a viable basis for political exchange. As part of their reform of the labor code, the Portuguese government had also included measures liberalizing the cost of individual dismissals, which meant that this was also not available as a “resource” for the unions to use in bargaining with employers. Moreover, although legislation on individual dismissals had been quite rigid, Portuguese legislation on collective dismissals (layoffs) was not particularly stringent.16 Indeed, to the extent that wage flexibility was insufficient to meet the needs of Portuguese employers in the early 1990s, employers were already able to resort to the more favorable legislation on collective layoffs.17 The bottom line, then, was that Portuguese employers, facing economic pressures for greater firm-level flexibility in the allocation of labor but an otherwise relatively liberalized labor market, had little to gain from maintaining a centralized system of bargaining. Employers were thus delighted to oblige the government and attack the existing system. Labor relations are still somewhat in flux, as unions initially challenged the constitutionality of the legislation; however, many of the legal issues raised were resolved with the introduction of a new labor code in 2008, and 2009 saw a new wave of contract expirations (Neumann 2009). Communist union

The Dynamics of Collective-bargaining Reform in Spain and Portugal | 171

leaders predict that within five to ten years Portugal will have a much more decentralized—and disorganized—system of labor relations.

Conclusion In accounting for the divergent fates of industrial relations projects introduced by Iberian governments in the past fifteen years, this chapter has pointed to the fact that collective-bargaining institutions are not the only source of rigidity in a political economy. Employers, facing a range of other labor market rigidities, may be willing to exchange continued rigidities in one area for more flexibility in another. This relatively simple argument suggests that we take seriously Locke and Thelen’s (1995) admonishment to contextualize supposedly universal economic pressures for decentralization. Depending on the mixes of flexibility and rigidities in a given national economy, a common pressure for greater flexibility in one area may play itself out in very different ways. Furthermore, the evidence presented in this chapter also casts doubt on some recent work in comparative political economy on the changing dynamics of wage restraint. Traditionally, scholars have understood wage restraint as occurring via the logic of “political exchange.” The logic of the exchange was simple: unions delivered wage restraint contingent on the state expanding the social wage. According to Isabela Mares (2006), it is the exhaustion of this bargain that has resulted in higher European unemployment: Due to the “growth to the limits” of the European welfare state, governments can no longer deliver payoffs (in the form of expanded social policy benefits) to unions. As a result, unions have had little interest in moderating wages. Interestingly, however, evidence from the Spanish case, presented in this chapter, suggests that significant wage restraint may still be possible, even in the absence of government willingness to engage in exchange with unions. In Spain, wage moderation (and lower unemployment) emerged in the late 1990s only when the state removeditself from direct intervention in the labor market and refused to participate in the “market” for political exchange. This suggests that governments may be able to secure union cooperation in wage restraint not through political exchange, but rather by threatening the very organizational underpinnings of union strength. For unions, the threat of political and economic marginalization may be a functional substitute for political exchange with the state. This dynamic is not limited to Spain. The experience of the Netherlands since the early 1990s also suggests that wage restraint may be possible even when the state refuses to offer nonmarket financial compensation in

172 | Sara Watson

exchange for union agreement to limit wage militancy. In the Dutch case, the government indicated its refusal to engage in the Pizzornian market for political exchange when it began privatizing a series of social security benefits. In response, the unions turned to collective bargaining—that is, to negotiations with employers—in order to fill in the social-protection gaps that the government had created. Wage restraint in the Netherlands during the 1990s emerged out of a bilateral bargain between employers and unions in which the unions agreed to take the issue of wage increases off the bargaining table in exchange for employers agreeing to provide workers directly with social insurance measures (Cox 2001). The Dutch experience, then, like the Spanish one, implies that although the days of “classic” political exchange predicated on bargains between the state and unions may be over, European political economies are not doomed to an endless future of high unemployment. Rather, the possibilities for wage restraint may depend on the willingness of employers and unions to identify mutually beneficial opportunities for political exchange.

Notes 1. 2.

3. 4.

5.

6. 7.

8.

Peter Swenson’s (1991, 2002) and John Bowman’s (2002) works are notable exceptions. Internal and functional flexibility involve permitting firms to adapt differentiated working-time arrangements and requiring workers to perform different tasks. External flexibility is the ability of firms to shed labor at will (i.e., by making it easy to fire workers). Salary flexibility can operate both at the macro- and the microeconomic level, involving either economy-wide wage restraint or the use of productivity-linked pay incentives at the firm level. Sofia Pérez (2000) is one important exception in this general trend insofar as she examines the emergence of efforts to promote “articulated” bargaining in Spain. Franz Traxler (1996) contrasts this outcome with that of “coordinated decentralization,” in which peak associations are able to maintain coordination in the absence of centralized bargaining. The Organización Sindical Española (OSE, Spanish Trade Union Organization), set up in 1940 and structured around the corporatist mantra of “unity, totality, and hierarchy,” was similar to the Portuguese system of Sindicatos Nacionais in its intent to overcome class conflict through the national union of producers. The non-microcorporatist possibility would be that atomized workers bargain their situation individually with employers at the firm level. This point, as Culpepper very insightfully points out, raises the crucial question of interest representation: whose interests are being represented by the employers’ association? For our purposes, it is sufficient to note that Varieties of Capitalism fails to problematize interest representation. See Culpepper (2007). For example, a firm must show three consecutive years of losses and must apply for the right to delink within a very short window of time after the signing of the collective contract (typically two weeks to one month) (See OECD 2001: 69.)

The Dynamics of Collective-bargaining Reform in Spain and Portugal | 173

9.

10. 11. 12. 13.

14. 15.

16.

17.

It is important to note that Spanish employers did not wish the government to directly legislate these new flexible forms of work organization; rather, they favored the government’s strategy of devolving these decisions to collective bargaining. Employers were generally pleased with the legislation, largely because (although the alleged purpose of the new law was to encourage bargaining over these issues among the social partners) the new legislation did not require a renegotiation of the now-derogated labor laws. This enabled employers to promote a generalized deregulation of the labor market while at the same time maintaining the “autonomy” of collective bargaining. Interviews: Julian Ariza and Marcos Peña, 2007, and Antonio Ferrer, 2008. As Pérez (2000, 2002) has noted, employers also agreed to cooperate in clarifying the division of labor between different levels within the collective-bargaining structure. For an excellent analysis of failed efforts of governments during the 1990s to encourage the expansion of the scope of collective bargaining, see Lima and Naumann (2000). Their right to elect representatives to the governing boards of firms was limited to public companies, and the right of workers commissions to be involved in enterprise restructuring was removed. Finally, the amount of compensated time granted to works council members was severely curtailed. See Lima and Naumann (2006). These are aimed at controlling labor costs, increasing productivity gains by improving the flexibility of the production process and redistributing work among the employees. This in turn emerged out of the dynamics of Portugal’s transition, in which the antisystem communist left was very strong. As a result of its dominance of the labor movement, successive governments worked to weaken the role of unions in pay-setting institutions (see Watson 2006). Portuguese employers also had recourse to a very liberal regime governing temporary employment. In contrast to Spain, Portuguese firms were effectively free to shed labor without having to worry about court challenges on the grounds of unfair dismissal (see Addison and Teixeira 1999). For example, a high-ranking government official in the labor ministry reported that current OECD indicators for employment protection legislation are currently being revised and that, as a result, Portugal’s score will be lowered, largely because its collectivedismissals legislation is on par with the European average.

Bibliography Addison, John, and Paulo Teixeira. 1999. “Is Portugal Really So Artosclerotic? Results From a Cross-Country Analysis of Labor Adjustment.” Coimbra, Portugal: Working Paper, Faculdade de Economia, Universidade de Coimbra. Ariza, Julián. 2007. Interview with the author. Madrid: Summer. Bea González, Eva and Jesús Ferreiro Aparicio. 2002. “Estructura de la negociación colectiva e incrementos salariales en España.” Boletin Economico de ICE. No. 2735 (8–21 July). Bentolila, Samuel, and Juan F. Jimeno. 2003. “Spanish Unemployment: The End of the Wild Ride?” Madrid: Working Paper 2003-10, Fundacíon de Estudios de Economía Aplicada. Bowman, John. 2002. “Employers and the Persistence of Centralized Wage Setting: The Case of Norway.” Comparative Political Studies 35, no. 9 (November): 995–1026.

174 | Sara Watson

Lima, Maria da Paz Campos, and Reinhard Naumann. 2000. “Social Pacts in Portugal.” In Social Pacts in Europe, edited by Philippe Pochet, et al. Brussels: EITUC. ―――. 2003a. “Parliament Approves Labour Code.” European Industrial Relations Observatory Online, www.eurofound.europa.eu/eiro/2003/05/inbrief/ pt0305101n.htm ―――. 2003b. “Portugal: Fewer Agreements Reached in First half of 2003.” European Industrial Relations Observatory Online, www.eurofound.europa. eu/eiro/2003/09/feature/pt0309103f.htm ―――. 2006. “Portugal: Delay in Renewal of Collective Agreements.” European Industrial Relations Observatory Online, www.eurofound.europa. eu/eiro/2006/04/articles/pt0604019i.htm Comisión Consultiva Nacional de Convenios Colectivos. 2008. Análisis Económico de la Negociación Colectiva en España. Madrid: MTAS. Confederación Española de Organizaciones Empresariales. 2008. Balance de la Encuesta de CEOE sobre Negociación Colectiva, 2007. Madrid: CEOE. Cox, Robert Henry. 2001. “The Social Construction of an Imperative: Why Welfare Reform Happened in Denmark and the Netherlands but Not in Germany.” World Politics 53, no. 3: 463–98. Culpepper, Pepper. 2007. “Small States and Skill Specificity: Austria, Switzerland, and Inter-Employer Cleavages in Coordinated Capitalism.” Comparative Political Studies 40, no. 6 (June): 611–37. Dornelas, António. 1999. “As relações industriais em Portugal: É possivel mudar?” Sociedade e Trabalho 7: 19–28. European Industrial Relations Observatory. 2004. “Questionnaire for EIRO Comparative Study on Changes in the National Collective Bargaining Systems Since 1990—case of Spain.” Dublin: European Foundation for the Improvement of Living and Working Conditions. El País. 1993. “La CEOE reclama al Gobierno que aplique la reforma laboral, aunque considera insuficientes sus propuestas.” 18 November. ―――. 1994a. “Carlos Solchaga: ‘La reforma laboral nos acerca a Alemania y no a Ghana.’” 5 February. ―――. 1994b. “Huelga en hostelería en plena temporada alta si la patronal no negocia la ordenanza.” 17 June. Ferreira Aparicio, Jesús. 2003. “Políticas de rentas y reformas laborales en España,” Revista del Ministerio de Trabajo y Asuntos Sociales 46: 15-40. Ferrer, António. 2008. Interview with the author. Madrid: April. Güell, Maia, and Barbara Petrongolo. 2005. “How Binding are Legal Limits? Transitions from Temporary to Permanent Work in Spain.” Economics Working Papers 682, Department of Economics and Business. Barcelona: Universidad Pompeu Fabra, 2003, and revised July 2005. Hamann, Kerstin, and John Kelly. 2007. “Party Politics and the Re-emergence of Social Pacts in Western Europe,” Comparative Political Studies 40, no. 8 (August): 971–94. Hancké, Bob, and Martin Rhodes. 2005. “EMU and Labor Market Institutions.” Work and Occupations 32, no. 2: 196–228.

The Dynamics of Collective-bargaining Reform in Spain and Portugal | 175

Hassel, Anke. “The Politics of Social Pacts.” 2003. British Journal of Industrial Relations 41, no. 4 (December): 707–26. International Labour Organisation. 2001. “Main Features of Collective Bargaining in the First Half of 2001 Portugal.” Geneva: ILO. http://www.ilo. org/public/english/employment/skills/hrdr/init/por_6.htm Iversen, Torben. 1996. “Power, Flexibility and the Breakdown of Centralized Wage-Bargaining: Sweden and Denmark in Comparative Perspective.” Comparative Politics 28, no. 4 (July): 399–436. Katznelson, Ira. 1986. “Rethinking the Silences of Economic and Social Policy.” Political Science Quarterly 101, no. 2: 307–25. Levy, Jonah. 1999. “Vice Into Virtue: Progressive Politics and Welfare Reform in Continental Europe.” Politics & Society 27, no. 2 (June): 239–73. Locke, Richard, and Kathleen Thelen. 1995. “Apples and Oranges Revisited: Contextualized Comparisons and the Study of Labor Politics” Politics and Society 23, no. 3: 337–67. Mares, Isabela. 2003. The Politics of Risk: Business and Welfare State Development. New York: Cambridge University Press. ―――. 2006. Taxation, Wage Bargaining and Unemployment. New York: Cambridge University Press. Martin, Cathie Jo, and Duane Swank. 2004. “Does the Organization of Capital Matter? Employers and Active Labor Market Policies at the National and Firm Levels.” American Political Science Review 98, no. 4 (November): 593–611. Molina Romo, Óscar. 2005. “Political Exchange and Bargaining Reform in Italy and Spain.” European Journal of Industrial Relations 11, no. 1: 7–26. Neumann, Reinhard. 2009. “Portugal: Multinational Companies and Collective Bargaining.” European Industrial Relations Observatory Online, www.eurofound.europa.eu/eiro/ studies/tn0904049s/pt0904049q.htm OECD. 2001. Economic Surveys: Spain. Paris: OECD. ―――. 2004. Economic Surveys: Spain. Paris: OECD. Peña, Marcos. 2007. Interview with the author. Madrid: Summer. Pérez, Sofia. 2000. “From Decentralization to Reorganization: Explaining the Return to National Bargaining in Italy and Spain,” Comparative Politics 32, no. 4 (July): 437–58. ―――. 2002. “Monetary Union and Bargaining Institutions in the EU: Extrapolating from some member states’ experiences,” Comparative Political Studies 35, no. 10: 1198–227. Pochet, Philippe et al., ed. 1997. Social Pacts in Europe. Brussels: EITUC. Rhodes, Martin. 2003. “National ‘Pacts’ and EU Governance in Social Policy and the Labour Market.” In Governing Work and Welfare in the New Economy: European and American Experiments, edited by Jonathon Zeitlin and David M. Trubeck, 129–57. Oxford: Oxford University Press. Royo, Sebastian. 2007. A New Century of Corporatism? New York: Greenwood/ Praeger, 2002. ―――. 2007. “Varieties of Capitalism in Spain: Business and the Politics of Cooperation.” European Journal of Industrial Relations 13, no.1 (Spring): 47–65.

176 | Sara Watson

Streeck, Wolfgang. 1984. “Neo-Corporatist Industrial Relations and the Economic Crisis in West Germany.” In Order and Conflict in Contemporary Capitalism, edited by John Goldthorpe. Oxford: Clarendon Press. Swenson, Peter. 1991. “Bringing Capital Back In, or Social Democracy Reconsidered: Employer Power, Cross-Class Alliances, and the Centralization of Industrial Relations in Denmark and Sweden,” World Politics 43, no. 4 (July): 513-44. ―――. 2002. Capitalists Against Markets: The Making of Labor Markets and Welfare States in the United States and Sweden. New York: Oxford University Press. Traxler, Franz. 1996. “Collective Bargaining and Industrial Change: A Case of Disorganisation?” European Sociological Review 12, no. 3: 271–87. ―――. 2003. “Bargaining, State Regulation, and the Trajectories of Industrial Relations,” European Journal of Industrial Relations 9, no. 2: 141–61. Traxler, Franz, Sabine Blaschke, and Bernhard Kittel. 2001. National Labor Relations and in Internationalized Markets. Oxford: Oxford University Press. UGT. 2006. Balanço da Contratação Colectiva de Trabalho, 2006. Lisbon: União Geral dos Trabalhadores. ―――. 2007. Balanço da Contratação Colectiva de Trabalho, 2007. Lisbon: União Geral dos Trabalhadores. Wallerstein, Michael and Miriam Golden. 2000. “Postwar Wage-setting in the Nordic countries.” In Unions, Employers and Central Banks, edited byTorben Iversen et al., 107–37. New York: Cambridge University Press. Watson, Sara. 2006. “The Left Divided.” PhD thesis, Department of Political Science. Berkeley: UC-Berkeley. Xavier, Bernardo. 2005. “La Negociación Colectiva en Portugal” (unpublished manuscript).

Section III

DIVERGING INSTITUTIONAL LEGACIES, IDEAS, AND SOCIAL REFORM

(

Chapter 8

SOCIAL POLICY CHANGE “UNDER THE RADAR SCREEN” Health Care Reforms in Seven Small Countries

( Kieke G. H. Okma, Luca Crivelli, Toni Ashton, Iva Bolgani, Tsung-Mei Cheng, David Chinitz, Meng-Kin Lim, Hans Maarse, Rachel Meislin and Tim Tenbensel

Introduction This chapter analyzes the health reform experience in seven small and midsized industrial democracies during the last decades of the twentieth century: Chile, Israel, the Netherlands, New Zealand, Singapore, Switzerland, and Taiwan.1 The study seeks to contribute to knowledge on crossnational policy through structured multicountry research. In looking at the health reform experiences of seven quite different health care systems, it represents a “most different system comparison” (Marmor 1998) under a common analytical framework. At first sight, the seven countries selected for this comparative study do not have much in common. They are located on different continents and vary in size, population numbers, ethnicity, and historical backgrounds, as well as in their dominant cultural orientations and styles of social policy making. However, they also have some common features. They are all small to midsized industrialized democracies with open economies.2 They all aim to provide universal access to affordable and quality health care, broadening insurance coverage while restraining public expenditure. Over time, they have faced similar fiscal constraints, growing (and changing) demand for medical services and changing views of the role of the state in society. All seven have discussed a similar range of reform

180 | Okma, Crivelli, Ashton, Tenbensel, Cheng, Chinitz, Lim, Maarse, Bolgani, and Meislin

options, and all have enlarged access to health care services by expanding (public and private) health insurance. Moreover, these countries have actually implemented major reforms in the last two decades, rather than just discussing policy intentions (see Palmer and Short 1989 about the different meanings of the term “policy”). Reforming health care systems is not just a matter of passing new laws; it also requires governments to adjust the legislation when faced with unexpected and undesired outcomes. These countries all had to engage in this “postreform maintenance.” Finally, the cases selected fall somewhat “under the radar screen.” Even with the recent surge in interest in the health reform experiences of Switzerland and Holland, these countries are usually not included in comparative studies (e.g., Reinhardt 2004; Herzlinger and Parsa-Parsi 2004; Harris 2007; Naik 2007; Enthoven and Van de Ven 2007; Okma 2008). The most general finding of this study is that despite similar pressures for change in all these countries—and considering similar policy options— there has not been convergence across the seven health care systems. In each case, public discontent with existing arrangements, pressures to curb public expenditure while expanding access, a political willingness (and power) to act, and the availability of solutions that fit in to the national political context combined to open “windows of opportunity” for change (Kingdon 1984), yet the national, cultural, and institutional legacies shaped or restrained the direction and speed of reform. The conclusions of this study confirm the need to collaborate across countries and disciplines. No individual researcher can systematically study the changes—or lack of change—in, say, more than three or four countries in depth. Furthermore, the study shows the need to pay more attention to small and medium-sized countries, particular those outside of the “old” industrialized nations of Europe and North America. Most comparative research in the field focuses on big countries, with far less attention on the experience of small and medium-sized countries. Yet the vast majority of the world’s nations fall into this latter category, and their diverse experiences provide crucial insights into health policy. This chapter aims to take a step toward filling this gap.

Central Questions and Methodology This chapter addresses the following questions: Why have the seven countries of this study, facing similar pressures and with similar options for government action, chosen very different pathways to restructure their health care? What caused a “window of opportunity” for change (King-

Health Care Reforms in Seven Small Countries | 181

don 1984) in each country? What did they do? What was the position of the major stakeholders? And what happened after the implementation of those reforms? This study does not limit itself to a single theoretical framework, instead, it combines economic terms and classifications with a vocabulary borrowed from a range of political science theories to develop a comparative framework to study health politics in different countries. In so doing, it shows that a mix of country-specific structural features—institutional legacies, particular cultural orientations—and the position and behavior of key actors in a system play a part in shaping health care policies and policy outcomes. Those factors also filtered common ideas about health care reform in distinct ways. It is possible to describe any given health care system in terms of a country-specific mix of funding, contracting and payment, and modes of providing medical services (OECD 1992, 1994). First, there are four dominant funding sources for health care: general taxation, public and private health insurance, direct patient payments, and voluntary contributions. Second, there are three modes of contracting health care services: integrated services, long-term contracts, and reimbursement (OECD 1992). The “integrated model” combines funding and ownership. The best-known example of this is the British National Health Service (NHS), which provides health care for all, largely paid out of general taxation. Others use long-term contracts between health care providers and governments or other third-party payers (e.g., administrative agencies of social health insurance). The reimbursement model, common in private insurance, involves patients first paying their provider and then seeking reimbursement from their insurance agency. Third, on the provision side, the ownership and management of health services can be public, private (both for-profit and not-for-profit), or, most commonly, a mix of these. There are also country-specific mixes of formal and informal care, traditional and modern medicine, and medical and related social services. The combination of funding, contracting, and ownership models largely determine the allocation of financial risks and decision-making power over the main players in health care. For example, tax funding and government ownership make for strong government influence, whereas private funding combined with legally independent providers restricts the role of the state (even while governments do play an important regulatory role). Health care reform, as defined in this study, involves substantially shifting both decision-making power and financial risks: from central government to regional and local authorities, or in the opposite direction; from government to health insurers and providers, or vice versa; or from collective actors to individual choice, or in the opposite direction.

182 | Okma, Crivelli, Ashton, Tenbensel, Cheng, Chinitz, Lim, Maarse, Bolgani, and Meislin

The above economic terms describe basic features of health care systems but do not explain the causes or effects of policy change; however, certain combinations of those elements do permit more rapid or profound change than others. Equally important are the particular features of national social policy making. Social policy making reflects underlying values or “dominant cultural orientations” in society. Such values translate into particular styles of policy making or ways of interacting between governments and society that affect outcomes of health reform efforts. Douglas and Wildavsky (1982) identify three dominant cultural orientations: “hierarchical collectivism”; “competitive individualism”; and “sectarianism.” The first refers to the collectivist traditions in the social-democratic states in northwestern Europe. They have strong bureaucratic traditions to implement policy and base their fiscal and social policy on principles of solidarity and equality. In both Germany and the Netherlands social policy making involves “neocorporatist” interaction between state and stakeholders. Bureaucratic elites engage in semipermanent consultations with the leaders of peak organizations that represent major stakeholders (Lijphart 1968; Hill 1993). Israel and Switzerland also belong to this corporatist tradition that gives veto power to the main actors in the policy arena. This orientation also produces, in most cases, a fair degree of policy stability. In systems oriented around “competitive individualism” individual demand and competing providers form the basis for the allocation of goods and services, with a limited role for the state. The United States belongs to this model: a liberal welfare state with strong emphasis on individual rights and responsibilities, weak collectivism, and an outspoken streak of sectarianism. Finally, societies with a “sectarian” orientation are segmented into groups with strong ideological preferences, for example, religious sects. Singapore has elements of this model, although the strong government can impose major policy changes, which the population accepts when it perceives these changes as being in their best interest. Chile does not fit in easily to the above categorization: the dictatorial regime from 1973 to 1990 eliminated politically (and often physically) citizens’ voices. Corresponding to these general models are specific modes of social policy making: (semi) “pluralist,” (semi) “corporatist,” and “exclusionary” (Labra 2007). The hierarchical collectivist orientation corresponds with corporatist styles, while individualistic orientation follows the pluralist mode. The exclusionary mode involves a powerful central government able to exclude certain groups from decision making, for example under a military dictatorship or authoritarian regime.

Health Care Reforms in Seven Small Countries | 183

It would be an error, however, to take such orientations as general representations of particular countries or policy outcomes. As argued above, models have hybridized and there is no one-to-one relation between such ‘models’ and countries’ policy-making. Nonetheless, the underlying values and policy styles make certain policy outcomes more likely than others. Thus values affect policy; and reversely, policies may help to fortify or reshape values (Marmor, Okma, and Latham 2006). Theories of historical institutionalism and path dependency (e.g., Pierson 1994) emphasize that institutional legacies and popular support for existing policy arrangements create strong barriers to change. Institutions alone, though, are not determinative, and the countries in this study provide interesting cases of “windows of opportunity” for change (Kingdon 1984). Within the parameters of institutions and particular opportunities it is also important to trace the behavior of the actors in the policy arena: governments and other third-party payers, patients, providers, insurers, and others. Hirschman’s (1970) theory of “voice” and “exit” is useful in analyzing those actors’ behavior in the face of health reforms. Exit is the base for consumer sovereignty in the market: if consumers do not like the price-quality ratio of a given product, and if there is sufficient competition, they may look elsewhere, forcing companies (or governments) to improve the quality of their goods and services. The second mechanism— voice—often plays a larger role in the political arena, particularly in health care. For many citizens or patients, physical exit (from a region or a wellknown health care insurer or provider) is difficult. Similarly, small countries with long traditions of social health insurance, like the Netherlands, Switzerland, and Israel do not offer care insurers and providers much opportunity to break off long-lasting relationships with each other. Looking at how they exert voice provides a further lens into understanding the reform process.

Origins of Health Care–Funding Models Industrialized countries share basic underlying principles and—sometimes rhetorical—goals in health policy: universal (or near-universal) access to health services and health insurance, solidarity (understood as fairness) in sharing the financial burden of illness, and good-quality services (OECD 1992).3 As governments have a large and growing role in funding health care, they are increasingly concerned with efficiency and cost control. Many nations also regard patient satisfaction, patient choice, and professional autonomy of physicians as important goals.4

184 | Okma, Crivelli, Ashton, Tenbensel, Cheng, Chinitz, Lim, Maarse, Bolgani, and Meislin

Despite these similarities, countries vary in their funding and contracting arrangements. For example, the majority of health care funding comes from (earmarked) general taxation in Australia, New Zealand, the United Kingdom, Italy, Spain, and the Scandinavian countries. In Austria, Belgium, Germany, France, Luxembourg, and the Netherlands social health insurance and, to a lesser degree, private insurance provide the main source. Germany (since 1994) and the Netherlands (since 1968) have separate population-wide social insurance for long-term care. In 2006, the Netherlands replaced the dual public-private insurance system by a mandate for all legal residents to take out health insurance. In all countries, patients face copayments or have to pay for some services out of their own pockets. Most governments mitigate the effects of user fees by exempting certain groups or setting limits on how much a family has to pay in any given year. Cross-national variation in funding and contracting models follows country-specific developments, but two particular events in Europe played a crucial role. The first was Chancellor Bismarck’s introduction of social (health) insurance in Germany in 1883. The law required certain categories of industrial workers to sign up with a Krankenkasse, literally “fund of the sick,” often somewhat erroneously labeled as “sickness fund.” These sick funds remained (legally) independent but took on the public task of administering the social health insurance for their members. The second major innovation was Britain’s establishment of a National Health Service (NHS) in 1948 (Klein 2006; Timmins 1995. The NHS extended health care coverage to the entire population, nationalizing hospitals but leaving family physicians (GPs) as independent practitioners. Reforms in the 1980s and 1990s encouraged and finally mandated GPs to work in regional group practices. In the twentieth century, many other countries in Europe and elsewhere followed the German and British examples by implementing income-protection schemes for unemployment, disability and old age, as well as universal or employment-based health insurance. By the end of the century, most if not all countries’ health care funding had become hybridized, combining employment-based arrangements for certain categories of workers with population-wide and tax-based payment for other health care services.

Reassessment of the Welfare States Even as these models of expanding access spread across the globe, they came under scrutiny. The confluence of economic, demographic, and ide-

Health Care Reforms in Seven Small Countries | 185

ological factors in the 1970s began to challenge the popular notion that the welfare state offered a solution to social problems, instead portraying it as an economic burden that was causing economic stagnation (Timmins 1995). First, economic stagnation coupled with high and persistent unemployment meant that public revenue was declining while spending continued to grow. Second, with the end of the postwar baby boom, demographers began to revise their earlier demographic projections downward and future pension outlays upward. Third, actors on both the left and right of the political spectrum began to argue that state powers had become too wide and too intrusive in the lives of individuals. Governments reacted to these pressures with efforts to contain public spending and reduce the dominant role of the state (Ranade 1998). In health, policy makers responded by expanding “old” tools such as strict budgetary limits—delisting entitlements from public insurance—and increased direct patient payments. They also used “new” policy tools, including a shift toward competition, efforts to privatize health care and health insurance, increased freedom of the insured and patients to select health insurance plans and providers, new payment methods, and reduced state control over the funding and planning of health care services.

International Proliferation of Health Reform Ideas In so doing, national governments began to look abroad for solutions (Klein 1995). This search fueled the rapid growth of cross-national studies in the 1980s and 1990s. The majority of these studies, however, are descriptive country studies. They often lack a common vocabulary or focus and suffer from poorly defined terms (Okma 2008). For example, terms like “health reform,” “managed competition,” or “consumer-driven health care” are regularly used but rarely defined in any operational way. The studies also commonly assume that policy as stated in formal government documents or law is the same as policy actually implemented. As we will see, the ultimate outcome of reform often differs greatly from the original policy intentions and requires substantial “after reform maintenance.” The main focus of this study is on health insurance and medical care; however, the boundary between medical care and related social services (for example, long-term care for the elderly or handicapped) is not always clear and varies across countries. Moreover, this study focuses on health policy that seeks to improve the financial protection against risks and the organization of health care services, with less attention to efforts to improve the health of the population (for example, measures aimed to improve road safety, food quality, consumer protection, or changes in lifestyles).

186 | Okma, Crivelli, Ashton, Tenbensel, Cheng, Chinitz, Lim, Maarse, Bolgani, and Meislin

This chapter defines “health reform” as a major shift in decision-making power over the allocation of resources and in financial risks in health care funding, contracting, ownership, and administration or governance. Such shifts include, among others, the abolition (or reinstatement) of compulsory contracting with providers, changes in the authority over capital investments, expansion or contraction of entitlements of public health insurance, and (new) restrictions on medical decision making imposed by practice guidelines and other rules. Changes in decision-making power and financial risks can also shift from the national level to regional and local governments (or in the other direction), or from government to individual insurers, patients, and insured. These changes can further affect the organization of health care. For instance, reforms allowing health insurers to contract health services selectively have triggered mergers and other collaboration by health care providers keen to defend or extend their market positions. Despite the common pressures and similar reform models, the following sections show that the existing institutions, opportunity structure, and actors shaped the actual health reforms and postreform maintenance differently across cases.

Health Care Reforms in the Netherlands, Switzerland, Chile, Israel, New Zealand, Singapore, and Taiwan This section analyzes the historical developments and current arrangements for funding, contracting, and “governance” (ownership and administration) of health care in seven countries. In each case it looks at the origins of the health care reforms, the discussion of policy options, and almost universal “after-reform adjustments.” As in recent years, The Netherlands and Switzerland have become the focus of international commentary about “consumer-driven health care,” we will start with those two countries.

Health Care Reform in the Netherlands: Market-oriented Experimentation Within Public Parameters The Netherlands is a small, densely populated country in Western Europe, with a population of sixteen million in 2006. It has a stable democracy and a long tradition of consensual, “neocorporatist” policy making, where governments share the responsibility for social policy making with organized groups in society (De Swaan 1988; Lijphart 1968; Wilensky 2002).

Health Care Reforms in Seven Small Countries | 187

In 2007, total health spending in the Netherlands exceeded €50 billion (approximately US $70 billion—in 2006 one euro equaled about 1.40 US dollars), or about €3,100 (US $4,340) per person, per year. (By 2010, that amount had increased to over 70 billion or more than €4,300 per person). Of this total, about 47 percent came out of the basic health insurance covering medical care and prescription drugs, 42 percent from contributions for the long-term care insurance, Algemene Wet Bijzondere Ziektekosten (AWBZ), about 7 percent from patient copayments, and about 5 percent from tax subsidies (Ministry of Health, the Netherlands 2006). Health funding consisted of a mix of earmarked taxes, flat-rate premiums, user fees, and government subsidies. The Dutch health reform debate started in the early 1970s with efforts to centralize funding and administration with regionalized health care planning (Okma 1997). In the early 1980s, the combination of economic stagnation and high unemployment, fear of population aging, and eroding faith in government planning led to a change in the direction of Dutch welfare policies. After reducing the level and duration of unemployment and disability benefits, governments shifted their attention to health care (ibid. 1997). In 1987, an expert committee, the so-called Dekker Committee, proposed to reduce the government’s role in health care and strengthen competition and consumer choice (Commissie Dekker 1987). Opposition resurfaced several years into implementation, political support waned, and the strong veto powers of the main actors in the policy arena brought the reform process to a halt (Okma 1997). The 1994 “Purple Coalition” of the Labor Party Partij van de Arbeid (PvdA), liberal conservatives VVD and liberal Democrats D66 shelved the reforms but did not reverse the steps its predecessors took (Okma and De Roo 2009). In the early 2000s, health reform made its political comeback (De Roo 2002; Strategisch Akkoord 2002 2001). Rapidly rising health costs—mainly due to the political decision to invest heavily in expanding capacity in order to reduce waiting times and waitlists—the return of the Christian Democrats to power, and the availability of a reform model that seemed to reflect political preferences for a reduced role of government created a “window of opportunity” for change. The 2003 governing coalition took up the basic ingredients of the earlier Dekker proposals with an even stronger orientation toward market competition (Hoofdlijnenakkoord 2003). As the coalition had a comfortable majority, the reform bill passed Parliament in a surprisingly short and uncontested time in 2005. The return of the Labor Party in 2007 to the governing coalition did not affect the introduction of the new insurance (Regeerakkoord 2007). Moreover, the main players in the field—in particular health insurers and some large provid-

188 | Okma, Crivelli, Ashton, Tenbensel, Cheng, Chinitz, Lim, Maarse, Bolgani, and Meislin

ers—had already anticipated the introduction of the new scheme, expressing far less opposition than in the 1980s and 1990s. As a result, beginning in January 2006, a mandatory “basic health insurance” (or rather, mandate to take out private insurance) replaced the former public and private health-insurance schemes (Ministry of Health, the Netherlands 2005). All legal residents had to sign up with one of the competing health insurers to obtain basic coverage, and most Dutch citizens have done so. In 2009 they paid, on average, about €1,050 (US $1,680) per person, per year, as flat-rate premiums directly to their insurer. All insured also paid on average, about €1,000 for the long term care insurance. (Higher income earners pay more, up to a ceiling of €3,600). The insurers receive the other half of their income from earmarked taxes that employers pay into a central fund administered by the Tax Department. Health insurers can selectively contract with self-employed physicians and some other health care providers. Finally, a small but rapidly growing share of funding comes from general taxation, for example the premiums for insured younger than 18 year, research subsidies, and the administrative costs of the regulatory and administrative bodies for the health insurance. Patients face modest copayments, and low- to middle-income families can apply for a fiscal subsidy. By law, premiums for the basic coverage are community-rated, but insurers set their own premiums and cannot turn down applicants. The scheme combines features of both the former private and social insurance.5 The coverage more or less equals that of the former sick-fund scheme. Efforts to reduce the scope of entitlements from social insurance in the last three decades have largely failed; in fact, they read as a “catalogue of failure” (Maarse and Okma 2004). In 2006, the first year of the reforms, about one-fifth of the population changed their insurer. In the second year of the new insurance, less than 5 percent of Dutch insured switched (Smit and Mokveld 2007). A new phenomenon in the Dutch insurance market since the reforms is the rise of collective contracting by groups of patients—at least for groups that insurers are willing to accept. In 2010, about two thirds of the insured joined a collective contract offered by their employer. The reforms also had unanticipated consequences that evoked policy adjustments. By 2007 the number of uninsured or those defaulting on insurance was rising, creating a political problem. In response, the government announced that if any uninsured ended up in the hospital, he or she would have to pay the costs, take out insurance on the spot, and pay a fine as well. Then, after a study showed that (young) immigrants, single parents, and welfare recipients were overrepresented in the uninsured population, the government realized that this model would be hard to enforce (CBS 2007). To deal with this problem, the government adopted

Health Care Reforms in Seven Small Countries | 189

a proposal by local authorities to have the welfare offices administer the flat-rate charges for welfare recipients. In addition, the Ministry of Health took over the cost of debt collection for defaulters (previously the insurers’ responsibility) as long as the insurers agreed to keep the delinquents on their roll (Ministry of Health, the Netherlands 2007a, 2007b). Equally, we see changes in health care provision. Yet here, too, longstanding practices have shaped health reform efforts. Dutch hospitals have a centuries-long tradition of private, not-for-profit ownership and governance by self-appointed boards (De Swaan 1988; Okma 1997). The last two decades have seen a rapid process of mergers and takeovers that led to smaller numbers of bigger hospitals, as well as vertical and horizontal integration of other health services, in particular in long term care (Boot 1998; RVZ 2003). This has raised concerns (and sometimes even outright rebuke) from the National Competition Authority (NMa).6 Dutch insured are free to choose their general practitioner (GP) or hospital—but they may have to pay extra charges if these providers are not contracted by their health insurer. Until the mid-2000s, insurers were reluctant to break off long-standing contractual relations with providers, but some have started to use their power to contract selectively. Most GPs and some other health professionals work solo or in small group practices. Others are employed by hospitals or other facilities, and a minority work as self-employed practitioners. GPs receive a mix of payments: a fixed amount (“capitated payment”) for each patient; fees for a patient consultation and certain other activities; and subsidies for particular purposes (e.g., for buying computers). The largest share of hospital budgets is (still) based on historical costs, but the system is—slowly—shifting to case-based payments. At first, the government encouraged medical specialists and hospitals to develop case-based tariffs themselves, but this decentralized process led to over 30,000 tariffs (covering less than 20 percent of all hospital activity in 2007). The implementation of the “home-grown” and very complicated casebased hospital-payment model has slowed, and in 2008, the government announced a drastic simplification but as of 2010, that had not been implemented. One of the side effects of the increased emphasis on competition has been the erosion of the traditional corporatist bargaining model (Lijphart 1968; Okma 2002). For many decades associations of hospitals and other providers met with representatives of the public and private health insurers to negotiate contracts. After the NMa ruled that such collective bargaining, in fact, implied undue market protection, the providers had to abandon this practice and now have to seek contracts with insurers on an individual basis. In theory, every single general practitioner has to sit down with all the insurers to negotiate contracts. In practice, however, Dutch

190 | Okma, Crivelli, Ashton, Tenbensel, Cheng, Chinitz, Lim, Maarse, Bolgani, and Meislin

GPs have found a way around that individualized bargaining model by creating legal entities that serve as aggregate bargaining agencies. Importantly, the 2006 health insurance reform did not (fully) replace earlier models of government planning and control. The current system involves an intricate layering of competing, and sometimes conflicting, governance models; it reflects efforts to decentralize. At the same time, faced with budgetary pressures—or perceived (future) budgetary pressures—the government has not really reduced its involvement in quality control, information technology, the development of new payment methods and other areas (Scheerder 2005; Ministry of Health, the Netherlands 2004). While the policy rhetoric emphasizes market efficiency, less government, and more consumer choice, the state has actually expanded its role in monitoring and supervising health care. The system is less market oriented than some experts claim (Schut and van de Ven 2005; Enthoven and van de Ven 2007) or some foreign observers hope it to be (Naik 2007; Harris 2007; Okma 2008). The Dutch experience confirms the need for “postreform maintenance.” In several instances, the government (partially) reinstated entitlements it had delisted previously (like dental care or birth control) because of widespread opposition. Likewise, the response to the rising numbers of uninsured and defaulters in 2009 (see above) shows the need for governments to continue to innovate even after reforms are formally in place.

Health Care Reform in Switzerland: From Social Insurance to Federalist Arrangements Switzerland is a small landlocked country in central Europe, with a population of 7.8 million. It is a federal state composed of twenty-six smaller provinces or cantons. Its decentralized political system with institutions of direct democracy, long tradition of social security, and liberal economic culture provide Swiss citizens with an unusually high degree of “voice” and “exit” opportunities (Hirschman 1970). Swiss citizens can directly engage in the political decision process by referendum and popular vote. The country’s federal structure further allows for distinct regional models. Together, these factors have produced large regional variations in health care and health expenditure, as well as seemingly insurmountable barriers to nationwide reform (Crivelli, Domenighetti, and Filippini 2007). The first Swiss social health insurance law of 1911 established a statutory package of benefits administered by private nonprofit sick funds (CIVITAS 2002).7 In contrast to the French or German social insurance, individuals—and not employers—were (and are) responsible for obtaining

Health Care Reforms in Seven Small Countries | 191

insurance. By 1990, nearly 98 percent of the population had (voluntarily) purchased insurance. The sick funds faced financial difficulties throughout the middle of the twentieth century. Despite these pressures, the popular vote rejected several proposals by left-wing parties to shift to income-dependent contributions. In 1994, however, Parliament did pass the Revised Health Insurance Law, which aimed to strengthen solidarity, improve cost control, and promote fair competition among health insurers. The law mandates that all citizens take out health insurance. As in the Netherlands, the law defines the basic coverage that includes (unlimited) inpatient and outpatient treatment, and one-third of prescription drugs (also subject to a 10 percent copayment). In contrast to Holland it also includes care for the elderly and handicapped. Patients pay for all other non-covered drugs directly or seek supplementary coverage. Health insurers cannot make profits off the mandatory basic coverage but they can do so from their supplemental-coverage plans. As in Germany and The Netherlands, there has been a market concentration in health insurances. Today there are about 90 independent and competing sick funds, down from 1,100 in the mid–twentieth century. These funds must contract with all providers but can offer alternative “managed care” plans. Historically, the insurers had federal, regional, religious, or occupational bases (ibid.), but nowadays most operate on a national scale as regular commercial insurers. Since 1996, the federal legislation sets three basic requirements for health insurance: a benefit basket defined by law, the right to change health insurer yearly, and uniform insurance contracts. Swiss insured can change insurer or insurance plan every year. Insurers have to accept anyone seeking basic coverage and provide community-rated premiums within a given area. They can, though, offer discounts to young adults (aged nineteen to twenty-five), and they must set lower premiums for children (up to age eighteen). Insurers receive (or have to pay) extra funds to compensate for the overrepresentation (respectively under-representation) of demographic risk groups in their portfolio (Beck et al. 2003).8 To alleviate the impact of this regressive financing, the state gives subsidies to low-income households (amounting to 20.5 percent of total premiums in 1999 but only 17.5 percent in 2009). These subsidies have failed to grow apace with premiums. Insurers offering basic cover must register with the Federal Office of Public Health, which monitors the insurance market. Within this broad legal framework, the cantons are responsible for the implementation of the federal health-insurance legislation and for regulating and controlling health care provision.

192 | Okma, Crivelli, Ashton, Tenbensel, Cheng, Chinitz, Lim, Maarse, Bolgani, and Meislin

As a result, there is substantial variation in the system. Cantons differ significantly in their health-service regulations, public-spending strategies, supply of hospital and ambulatory care, and expenditure levels. This has resulted in large regional variations in financial burdens: health insurance for a family of four may claim only 5 percent of disposable family income in one canton but 16 percent of disposable family income in another (Balthasar et al. 2005). Despite these differences, Swiss citizens do not move much between cantons and for long time were not inclined to switch health plans (Colombo 2001; Frank and Lamiraud 2008). In 2004, only 10 percent had a managed care plan, and less than 3 percent of the policy holders switched insurers. “Partial exit” (a switch to another plan with the same insurer) mostly occurs in high-deductible plans (over 1,200 francs or US $720 per year). Illustrating the process of self-selection with respect to the choice of high deductibles, the death rate of those selecting the minimum deductible is twice as high as those who select the average amount (Geoffard, Gardiol, and Grandchamp 2006). Swiss insured can freely choose their health care provider, unless they have opted for a “managed competition” (or “managed care”) plan that restricts choice in exchange for lower premiums (Lehman and Zweifel 2004). As in other countries, such plans tend to attract younger, healthier, better-informed, and more mobile people (Beck et al. 2003; Strombom, Buchmueller, and Feldstein 2002). However the situation has changed significantly since 2004 because of the continuous growth of premiums. That has made health insurance more and more unaffordable for the middle class (those who are not eligible for subsidies). In 2009, more than one-third of the population decided to benefit from the discount associated with managed-care contracts, and the percentage of people switching between 2009 and 2010 to a cheaper health insurer exceeded 15 percent. These changes have also affected health care services. Swiss citizens receive their medical treatment in a wide range of settings, like hospitals, clinics, and ambulatory-care facilities. There is ample hospital capacity: over 310 hospitals that can accommodate relatively long stays. Medical specialists work in hospitals and private practice. Self-employed health professionals receive fee-for-service payments, while hospital budgets are based on a mix of direct government subsidies, per-diem or DRG-based payments (due to be fully implemented by 2012), and other payments (Crivelli and Bolgani 2010). Swiss health care is based on a liberal conception of health and medicine and allows for a high degree of freedom of choice of plan and provider. Proponents of “consumer-driven” health care (e.g., Herzlinger and ParsaParsi 2004) argue that the Swiss system combines universal coverage with

Health Care Reforms in Seven Small Countries | 193

effective cost control. Yet critics point out that if one compares the Swiss system to that of the United States, it is clear that extensive government regulation, not competition or high out-of-pocket payments, is what has kept costs down (Reinhardt 2004). The competitive model assumes that consumers are willing and able to use their options that will prompt insurers and providers to offer better and more patient-friendly care at lower costs. This assumption cannot be proven conclusively from the Swiss system, where there is little systematic information available about performance indicators like clinical quality, efficacy, and effectiveness of individual providers. Moreover, the proliferation of plans does not combine well with cantonal responsibility for the availability of health care. By law, cantons have to safeguard sufficient health care capacity, and this has led to both the creation of regional monopolies and fragmentation of the hospital system (Crivelli 2007). The health care system in Switzerland, then, demonstrates that even within a single country health care reform can work in different ways, reshaping insurance and provision in often unanticipated ways that may require additional measures by government to address unwelcome outcomes.

Health Care Reform in Chile: Changing Political Regimes and Health Policies Chile is a midsized industrial country in southeastern Latin America with a population of about sixteen million as of 2004. Chile was one of the first countries in Latin America with public pensions and health insurance (De Viado and Flores 1944). The 1924 social health insurance covered urban manual workers. Its entitlements included sickness and medical benefits, maternity benefits, health services for infants up to two years old, and benefits in case of invalidity, old age, and death. There were separate schemes for white-collar workers, which merged into the National Health Service for Employees, the Servicio Médico Nacional de Empleados (SERMENA) in 1942 (Labra 1995). In 1952, Chile began to implement its universal national health service, the Sistema Nacional de Salud (SNS), resembling the British National Health Service. Almost two decades later, in 1973, the Pinochet military regime overthrew Socialist president Allende. Heavily influenced by neoliberal ideology, the new regime favored a reduced role of the state, privatization, and consumer choice in health care (Jost 1999). It drastically reduced social spending and public health services, and in the early 1980s, it allowed those insured by the public scheme, the Fondo Nacional de Salud (FONASA) to opt out and seek coverage with one of the private insurers, the Instituciones de Salud Previsional (ISAPREs). As in other countries, this system,

194 | Okma, Crivelli, Ashton, Tenbensel, Cheng, Chinitz, Lim, Maarse, Bolgani, and Meislin

allowing the insured to opt in and opt out, led to a spiraling process of risk selection: the young and healthy went private, but the sick and elderly had to remain in (or return to) the public scheme, as they faced serious access barriers in the private market (Höfter 2006; Labra 2002; Barrientos 2002). After the return to democracy in 1990, the new government announced massive additional investments to improve the quality of public health services and to reduce waiting times. It did not do away with the dual health-insurance system but imposed extensive regulation on (private) health providers and insurers. For example, insurers have to offer a certain minimum coverage, a uniform premium structure with community-rated premiums, and accept dependents. Spouses were not permitted to split their coverage between the public and private insurance. In 1997, about 25% of the population had signed up with an ISAPRE, but that share had dropped to 16% in 2008 (WHO Bulletin 2008). The public scheme FONASA covered almost 80 percent of the population and 5 percent had access to other schemes (e.g., armed forces). The public health services (including primary care, immunization and other preventive services, eldercare, and mother and child care) cover 100 percent of the population. The ISAPREs mostly function as traditional for-profit insurers and offer coverage for health care and pay for sick leave. Most charge user fees, with copayments between 30 and 50 percent that are subject to legal caps in order to mitigate the financial burden for lower-income families. Some of the private insurers own health facilities or have preferred-provider arrangements with health care providers (WHO Bulletin 2008; Jost 1999). Chile has a long tradition of free choice of provider, and insurers are hesitant to impose too many restrictions on their insured or face the wrath of the medical association. The majority of hospitals, local clinics, and municipal primary-care centers are under public ownership and control. Long-term contracts between FONASA and hospitals provide the base for paying inpatient care. Hospitals receive prospective budgets (60 percent of budgets is based on historical costs), plus additional payments for specific activities. Private insurers provide about 15 percent of the hospital income; they negotiate contracts and fees with (preferred) providers. The public system, though, faces a lack of funding and shortages of qualified personnel. There are frequent strikes by health professionals dissatisfied with low pay and poor working conditions. (Jost 1999). In the late 1990s, the government announced a shift toward capitation payment of primary care and case-based (DRG-based) payment for hospital care (Jost 1999). But this turned out to be a complex process. In the mid-2000s, case-based payment constituted only about 10 percent of hospital budgets. Facing much opposition to the market-oriented changes,

Health Care Reforms in Seven Small Countries | 195

the government reaffirmed its strong commitment to maintaining a public health system as a viable alternative to the private sector. The Chilean experience illustrates that a change in political regime can create a “window of opportunity” for change. But dominant values as well as long-standing institutions restrain or slow what government can, in fact, implement. Even while the military regime encouraged private health insurance, it did not do away with all public health services. Likewise, the restoration of democracy in 1990 did not do away with private health insurance. Both the move toward and away from the market have followed a distinct path.

Health Care Reform in Israel: New Procedures for Determining Entitlements under Public Health Insurance Israel, a small country at the eastern shore of the Mediterranean with about seven million inhabitants, is another country with a long tradition of public health insurance. Labor unions established the first mutual sick funds in the first three decades of the twentieth century (even before the creation of the state of Israel), modeled after the German social health insurance. It gradually expanded to cover almost the entire population (Rosen, Goldwag, and Thompson 2003). By the late 1980s, 95 percent of the population had voluntarily enrolled with one of the four sick funds. Despite this nearly universal coverage, the system faced financial instability, public and provider dissatisfaction, hospital overcapacity, and fragmentation of services. As in the Netherlands, even the small number of uninsured caused concern among many politicians and members of the Israeli public. In 1995, Israel enacted the National Health Insurance (NHI) Law. The law mandated all residents (not only the employed) to register with a sick fund, introducing a hybrid between the German and UK funding models. The NHI is a population-wide social health insurance, administered by four major (competing but not-for-profit) sick funds. Israelis can choose the fund they want to register with and can change up to two times per year. In 1995, the first year they had this option, about 4 percent of the population switched funds, but in the years thereafter the rate of change slowed to about 1 percent. The largest fund, Clalit, covers 60 percent of the population. Insurers can set their own premiums for supplementary coverage, and membership is subject to underwriting. The National Health Insurance Regulator, a branch of the Ministry of Finance, regulates the private health insurance. (Rosen, Goldwag and Thompson 2003). Israel combines social insurance contributions, tax subsidies, and modest patient user fees to fund health care. In recent years, copayments for

196 | Okma, Crivelli, Ashton, Tenbensel, Cheng, Chinitz, Lim, Maarse, Bolgani, and Meislin

hospital stays and prescription drugs have increased, but there are many exemptions and caps on the total amount families have to pay each year. The mandatory social health insurance offers a wide range of entitlements, meaning supplemental health insurance covering private physicians and clinics. Complementary medicine plays an insignificant (albeit growing) role. The NHI was part of a three-pronged reform proposed by the statesponsored 1990 Netanyahu Commission. Resistance from hospital labor unions meant that the two other planks of the proposed reform—changing government hospitals to public trusts and the reorganizing the Ministry of Health—failed to materialize. However the NHI led to one radical change outside the envisioned reforms: the creation of an expert commission to assess whether to add certain health services to the social health insurance entitlements (Revital, Rosen, and Chinitz 1998). This committee faces many pressures from patients and lobby groups. Not surprisingly, the proposed additions, mainly pharmaceuticals, usually exceed the available funding. All insured are free to choose their primary care physician; these physicians are either employed in a local clinic of their health fund or are selfemployed with a contract with their fund. Access to nonemergency care generally requires a referral from the physician or preapproval from the fund. While Israelis can, and often do, exercise choice of hospital, referrals generally include direction to a specific provider. Hospitals receive capped budgets, though the funds typically reimburse 50 percent of budget overruns. Emergency care and outpatient clinics are paid on a feefor-service basis. Most physicians in hospitals are salaried, while general practitioners receive capitation payment for each patient on their list. The National Health Insurance Institute administers the funding of the NHI. It collects contributions via the tax system and allocates those over the sick funds. The Health Ministry (MoH) has overall responsibilities: it sets the rules, defines benefits, is involved in planning and allocation of budgets, sets hospital budgets, and imposes limits on public spending, as well as on the numbers of physicians. National professional associations of hospital physicians, nurses, and other providers negotiate salaries on behalf of their members. The collective bargaining and active participation of the main organized stakeholders resembles the neocorporatist style of social policy making of Western Europe. In general, this style gives providers of care a strong veto position. The fact that the Ministry of Health owns two-thirds of general hospital beds has created a conflict of interest and sometimes an inability to turn its attention from the day-to-day management of hospitals toward planning and regulation. By controlling hospital reimbursement rates, however, the ministry has been able to stabilize hospital expenditures. As of 2006, the

Health Care Reforms in Seven Small Countries | 197

sick funds were, by and large, working within balanced budgets. However the ministry has been less adept at regulating the quality of care. Lacking resources and with less than complete cooperation from physicians’ associations, it has not been able to create a framework for ongoing quality assurance in provision and insurance. Physicians’ associations and sick funds participate in benchmarking and other quality-assurance efforts, but they do not agree to public disclosure of measured results. Israel’s health policy–making thus combines strong government involvement with political timidity to enact radical change. Organized interests have strong veto positions that create barriers to rapid policy change. Still, the partially implemented reforms offer a more realistic and interesting comparative experience than “perfect” policy reform models that are not (fully) implemented. In practice, we see a very limited use of the “exit” option (seen as a crucial element of competition) by Israeli while another unintended aspect of the reform, the use of expert committees, are key. This experience shows again how specific reform ideas and practices play out differently in particular institutional contexts.

Health Care Reform in New Zealand: Shifting Emphasis from Community to Markets and Back Again New Zealand has a small population on a large land area (about the size of the UK) in the Pacific Ocean. Of the 4.2 million people, 70 percent live in urban areas. Since 1996, elections have been based on mixed proportional representation. The 1840 Treaty of Waitangi plays an important role in defining the rights and privileges of about six hundred thousand Māori, New Zealand’s indigenous population. The 1938 Social Security Act laid the foundation for the public health system. Resistance from medical professionals, many of whom did not want to become government employees, led to a dual health system, with state hospitals offering services free of charge and general practitioners charging patients for primary health services. In later decades, this dual system became problematic as the “free” hospitalization (versus primary care, where patients faced user fees) led to relatively high rates of hospitalization. In 2008, New Zealand spent NZ $17.7 billion (or US $11.2 billion), or 9.8 percent of the GDP on health care (Ministry of Health, New Zealand 2010). The main sources were public taxation (80 percent), patient payments (14 percent), and private insurance (5 percent). The public spending included general taxation, as well as the universal accident insurance. All permanent residents in New Zealand are automatically covered by the public system and receive most health care free of charge—including

198 | Okma, Crivelli, Ashton, Tenbensel, Cheng, Chinitz, Lim, Maarse, Bolgani, and Meislin

hospital care, maternity care, and many community-based services. User fees apply to visits to general practitioners, prescription drugs, and stays in long-term care facilities. Most GPs work in group practices with two or more other GPs, practice nurses, and other staff. Patients can choose their GP but need GP referral to receive hospital care (and usually have no choice of provider), unless they prefer to go outside the public system and pay for private health care out of pocket or via private health insurance. District Health Boards (DHBs) are responsible for administering the major share of all public funding and contracting with providers in their district. DHBs own hospitals and some community-based services, provide health services to their populations, and receive their budget from the Ministry of Health, with additional funding to provide for certain high-risk (and high-cost) groups. The DHBs purchase other health care from private providers. The economic recession of the mid-1970s hit New Zealand hard. Successive governments implemented major (market-oriented) public-sector reforms between the 1980s and 1990s. In the mid-1980s, the government transformed local hospital boards (which were seen as inefficient) into fourteen Area Health Boards (AHBs). The AHBs became responsible for public health services, hospital care, and some community-based services while the central government remained responsible for other services, including primary care and disability services. This restructuring aimed to encourage more of a population-based approach to the provision of health services. However the AHBs faced mounting criticism relating to their perceived inefficiencies and lack of accountability. A governmentappointed expert committee recommended separation of the functions of contracting from providing health care. But, internally divided over the matter, governments could not agree, and the AHBs remained in place. The conservative National Party that came into power in 1990 extended market reforms to the health care sector. This government—mostly informed by a task force of experts set up by the Treasury and with little input from the Ministry of Health—swiftly replaced the AHBs with commissioners who were to oversee the implementation of the separation of providing and purchasing functions. The new course commenced in July 1993. It reflected the influence of new institutional economics and a public-choice framework that sought to “rationalize” health care by creating contractual relationships between independent providers and the public purchasers. Four newly established Regional Health Authorities (RHAs) were to contract all the health services for their populations. The public hospitals, renamed Crown Health Enterprises (CHEs), became for-profit organizations, with the Ministry of Health and Treasury as shareholders. One part of the proposals—to give patients free choice of insurer—was

Health Care Reforms in Seven Small Countries | 199

never implemented, and RHAs faced no threat of “exit” by disgruntled patients. By the time of the general election in 1996, however, there was little evidence of major efficiency gains under the new model. Many CHEs were in deficit. The restructuring proved costly, both in terms of public money and in terms of rising dissatisfaction of health professionals and the New Zealand population. The 1996 elections brought the first coalition government to power. The junior partner of the coalition was much opposed to the competitive health care idea. It forced the coalition to adjust the policy rhetoric about competition, emphasizing the need for collaboration instead, as well as the need to focus more on population health issues. Late in 1999, a new coalition took office. The Labour Party—now as senior coalition partner— announced a radical departure from the market-driven reform course. It proposed a return to the model of community governance of health care, with renewed emphasis on primary care, population health, and efforts to reduce inequalities. The coalition set up twenty-one DHBs with locally elected board members. The budget of the DHBs not only includes hospital care, but also primary care and care for the elderly. DHBs provide some of these services themselves, but they also contract with private providers. The Ministry of Health has framed general goals and guidelines in the New Zealand Health Strategy and in the 2001 Primary Health Care Strategy. The latter led to financial incentives for GPs and other providers to work together in Primary Health Organizations (PHOs). Patients who sign up with PHOs receive a higher government subsidy and hence pay lower copayments. Within five years, there were more than eighty PHOs, and most GPs and over 95 percent of the population had signed up (Ministry of Health, New Zealand 2007). New Zealand has thus proven itself to be an eager policy innovator. The seriously depressed economic situation of the 1980s prompted the then government to initiate market-oriented economic reforms that a few years later, after a shift to a conservative government, also led to the reshaping of the health care system. Similarly, the combination of a changed election system (that led to coalition governments and thus more consensual policy making), mounting dissatisfaction over the poor results of the reforms, growing popular opposition against the very notion of markets in health care, and a return of Labour to power in 2000 created another “window” to undo some of the former policies and push for a much stronger orientation on public health.

200 | Okma, Crivelli, Ashton, Tenbensel, Cheng, Chinitz, Lim, Maarse, Bolgani, and Meislin

Health Care Reform in Singapore: Pragmatism, Not Ideology, as a Driving Force for Change Singapore is a tiny island city-state with a population of 5 million, one of the most densely populated countries in the world. It is a parliamentary democracy that gained self-rule from the British in 1959 and independence from Malaysia in 1965. The ruling People’s Action Party has been in power since 1959; hence, it has the rare advantage of being able to pursue its reform agenda without much opposition or undue interruption (providing an example of “exclusionary” policy making). In 2005, Singapore spent about S $7.6 billion (or US $5.1 billion) or 3.8 percent of its gross domestic product on health care. The main funding sources are employer benefits (35 percent), government subsidies (25 percent), and out-of-pocket payment (25 percent), with private insurance contributing 5 percent. In addition, there are three medical-savings schemes to help families pay their medical bills: Medisave, Medishield, and Medifund (for a more elaborate description, see Meng-Kin Lim 2010). In times of budget surpluses, the government adds to the savings schemes of lowincome families. Together, those savings accounts fund about 10 percent of total health care expenditure. Health reform began in early 1960s, when the political leadership had taken an abrupt right turn ideologically. It favored market mechanisms to allocate finite resources but, at the same time, compensation to protect lower income groups. For the first time, the new government introduced user fees (50 cents per visit to a government outpatient clinic). Furthermore, it decentralized primary care from the overcrowded General Hospital to outpatient clinics. In 1983, the government unveiled its National Health Plan, which included an ambitious hospital construction and expansion program. The hospital restructuring started in 1985 and took twenty years to roll out. The government considered various models to reduce government control over hospitals. At first, it announced full privatization. Facing widespread public unhappiness over this plan—and in a rare instance of reversal—it opted instead for “corporatization,” granting hospitals autonomy in fiduciary and operational matters (Phua 1991; Preker and Harding 2003). The facilities remained under the state-owned Health Corporation of Singapore (HCS). Within a few years, efficiency and service levels improved. But over time, competition created unwanted side effects, like the poaching of staff from other hospitals and missed opportunities to exploit economies of scale (e.g., in centralized drug purchasing or common standards for electronic medical records). Hospital expenditures rose sharply, and the government intervened in 2000. It regrouped the hospitals into

Health Care Reforms in Seven Small Countries | 201

two quasi-independent clusters—the National Healthcare Group and the Singapore Health Services (both with Ministry of Health–appointed boards). Simultaneously, the two clusters took control of all the government polyclinics for primary care. Shortly after, it introduced global budgeting for hospitals and DRG-based payments. Traditionally, Singapore patients have had free choice of provider. The private sector accounts for about 21 percent of inpatient beds and 80 percent of outpatient attendance. The government would like to see the private share of hospital beds increase to 30 percent and publishes hospital tariffs and selected quality indicators on its website to encourage consumer choice (Lim, Meng-Kim 2010). The reforms resulted in a dramatic rise in patient satisfaction and improvement in standards of care, and levels of service, as well as a drop in waiting times (Lim 2004b). It also led to a considerable growth in the number of foreign patients seeking medical treatment in Singapore. The government actively supported this trend as a welcome source of income for hospitals and other health care providers. The health care system of Singapore thus reflects an interesting mix of strong market orientation and individualism, with an acceptance of paternalistic government intervention. The absence of a tradition of well-organized stakeholders and opposition groups has contributed to the rapid adoption of top-down policies. The combination of individual savings accounts with employer subsidies and public (means-tested) subsidies targeted at low-income families makes for a quasi-private system under tight government control. Patients are accustomed to cost sharing. Compared to OECD countries, Singapore has been very successful in containing health care spending to below 5 percent of its GDP. As Singapore’s economy matures, economic growth will inevitably slow down, lessening the masking effect of the expanding GDP denominator. Moreover, Singapore has a very young population; the share of elderly (now only 8 percent of the population) is projected to rise to 25 percent by 2030, and this may push up health expenditure (Lim 2005). Paradoxically, just when governments elsewhere are mulling over cutbacks in health spending, Singapore plans to upgrade and expand its health facilities over the next ten years (Newsweek 2007). The government feels that to sustain Singapore’s dynamic economy, the population needs to grow by 2 million (largely through immigration). Singapore Medicine is further aiming to attract one million foreign patients by 2012, and it reckons this will generate S $3 billion and create 13 thousand new jobs (Choo 2002). Hence demand for quality health care services (and expenditure) cannot but rise. For the pragmatic government with a knack for

202 | Okma, Crivelli, Ashton, Tenbensel, Cheng, Chinitz, Lim, Maarse, Bolgani, and Meislin

turning necessity into virtue, that is not necessarily a bad thing. Singapore will likely continue to follow its own distinct path.

Health Care Reform in Taiwan: The Big Bang of National Health Insurance Taiwan consists of one major and several smaller islands, with a total land size of 36,000 square kilometers. It has one of the highest population densities in the world. Of the population of 22.8 million, the majority lives in urban areas, and less than 2 percent live in the mountainous areas and offshore islands. Taiwan’s total national health spending was 6.9 percent of its GDP in 2009. The National Health Insurance (NHI) accounted for 51.3 percent, out-of-pocket spending for 35.1 percent, government spending on public health, general administration and captal formation for 6.3 percent, and other (private not-for-profit) sources for about 6.4 percent of total health expenditure. The NHI is a single payer social insurance, covering the entire population, with income-based contributions. Insured, employers, and government all pay a share (Cheng 2003): employers contribute 35.5 percent, the insured contribute 38.5 percent, and the government contributes 26 percent to the NHI (Department of Health, Taiwan 2009). The contributions are levied on a per capita basis, up to a maximum of three dependents per insured household. Any additional dependents enjoy free coverage. The government pays the contributions for unemployed veterans and the poor. For over fifty years (1949 to 2000), Taiwan’s Nationalist government (Kuomintang or KMT), which had retreated to the island after its defeat by the Chinese communists in 1949, enjoyed the seat of political power. In 1995, the KMT-led government established the single-payer universal National Health Insurance (NHI) which overnight brought the then 41 percent of the population without health isnurance under the protective uembrella of the NHI. Policy makers began to consider health reform in the late 1980s. At that time there were more than ten separate health insurance schemes for particular population groups, for example, industrial workers, government employees, farmers, low-income households, etc. Altogether, these schemes covered about 59 percent of the population, leaving 41 percent without health insurance. Private health insurance did not exist (Cheng 2003). Several factors created a “window of opportunity” which led to the establishment of the NHI: several decades of high economic growth which created favorable economic conditions, rising popular demand for uni-

Health Care Reforms in Seven Small Countries | 203

versal coverage, a political challenge to Taiwan’s Nationalist government from the opposition party, the Democratic Progressive Party (DPP), and last but not least, the strong personal leadership of then president, Lee Teng-Hui (Cheng 2003). To prepare for the introduction of the NHI, government bureaucrats and scholars extensively studied health systems abroad, in particular in North America and Western Europe for lessons, and in the end opted for the single-payer health insurance scheme for the entire population. A single payer approach to national health insurance was thought to be the most efficient way to combine universal access and cost containment. The planning stage lasted seven and a half years, from 1986 to 1993, and parliament passed the NHI bill in July 1994. On 1 March 1995 went into full implemntation, an amazing five years ahead of the original 2000 implementation date (Cheng 2003). Overnight, the formerly uninsured (41 percent of Taiwan’s population) gained access to health care. Within a year, their utilization of medical services approached the same level as those who already had health insurance coverage in 1995 (Cheng 2003). Initially, the hasty rapid implementation of the NHI led to confusion and chaos, and critics expected failure, citing inadequate preparation. In retrospect, however, the early implementation may have been a blessing in disguise. Even though Taiwan’s economy was not directly hit by the Asian financial crisis of 1997-1998 and suffered less than many other Asian countries, economic growth nevertheless slowed considerably, from over 10 percent per year (1992 to 1995) to just over 4 percent in 2000 and to a negative 1.7 percent in 2001. Had Taiwan’s government waited to implement the NHI in 2000, the original start date, the NHI might not have come about as doubts of financial affordability and sustainability might have caused the government to postpone implementation, perhaps indefinitely. The public warmed to the program quickly. Nationwide surveys showed high satisfaction rates, in particular among residents in remote mountainous areas and offshore islands. In 2002, Taiwan’s supreme court ruled that no one in Taiwan could be denied care because of inability to pay (Cheng 2003). Clearly, Taiwan’s society considers access to health care a basic right for all. Enrollment in the NHI is mandatory. The single payer Bureau of National Health Insurance (BHNI) administers the NHI under the Department of Health (DoH). The BNHI has strikingly low administrative costs: 1.5 percent of its total budget in 2007 (Cheng 2007) and 1.4 percent in 2010 (Cheng 2010), largely due to the efficiency of its uniform administration and information system and the absence of expenses for litigation, marketing and advertising. The BNHI sets fees for all covered services in

204 | Okma, Crivelli, Ashton, Tenbensel, Cheng, Chinitz, Lim, Maarse, Bolgani, and Meislin

the NHI. The NHI’s benefits are comprehensive: they include inpatient and ambulatory care, laboratory testing, diagnostic imaging, prescription drugs, dental care (except orthodontics and prosthodontics), traditional Chinese medicine, day care for the mentally ill, limited home care, preventative care, and other services (Cheng 2003). Dialysis was added in 2005. Health care services are delivered through a predominantly private delivery system, and patients enjoy free choice of provider and therapy. Providers receive their revenues from three sources: predominantly feefor-service payments by the BNHI, patientuser fees and copayments, and the proceeds from the sale of goods and services not covered by the NHI (cosmetic surgery, single room in hospital, etc.) (Cheng 2003). While much of the academic debate on health policy—not only in the United States but also in many other countries—equates the word “choice” with “choice among private health insurers and health insurance products,” it is likely that ordinary people understand “choices in health care” to mean unrestricted choice of health care providers and medical treatments. This is one of the most important lesson to be drawn from Taiwan’s experience. We see that although a single-payer system, such as Taiwans’s, does not offer consumerchoice of private insurers or a wide variety of insurance products, it nevertheless provides consumers complete free choice of health care providers. A second key lesson is that a single payer system such as taiwan’s can easily control health care costs. Growth of Taiwan’s total national health spending before the establishment of the NHI averaged over 10 percent a year (14-16 percent a year). After the establishment of the NHI, growth in total health spending averaged in the low single-digit. For example, the average annual rate of growth for the period 2005-2009 was 4.4 percent. A third key lesson from Taiwan’s experience is that health information technology (HIT) is crucially important in the efficient administration of a health system. Taiwan’s government invested heavily in HIT in the very early stages of the NHI, and had benefited from its powerful IT system ever since (Cheng 2009). After looking abroad, Taiwan was able to incorporate elements from other health systems but developed its own distinct model. Taiwan’s experience also illustrates the need for more or less perpetual “after-reform reforms” as a crucial policy feature.

Health Care Reforms in Seven Small Countries | 205

Conclusions The main conclusion of this comparative study of the health reform experiences of seven small and midsized countries is that national values (or dominant cultural orientations), institutions, and interests all play an important role in the shaping and outcomes of national health policy (Klein and Marmor 2006). In each case, factors as diverse as fiscal and budgetary pressure, economic development, public opinion, and demographic and ideological change led to a reassessment of existing arrangements during the 1980s and 1990s. The search for solutions to these problems led to a rapid proliferation of health reform ideas that traveled the globe without much reflection on their applicability in national contexts. Only a few countries systematically studied experiences abroad; in most cases, countries “borrowed” reform notions from others without much reflection (Marmor, Freeman and Okma 2009). The health politics in the seven countries reflect country-specific cultural values (for example, an emphasis on solidarity versus individual consumer choice and family responsibility), institutions (for example, centralized governance versus decentralization and veto points in the system), and interests (for example, well-organized stakeholders that encourage, thwart, or slow-down reforms). This finding has a number of broader implications. A striking feature of the vast majority of comparative literature on health politics is its “Eurocentric” nature. Expanding the geographic range of countries studied not only widens the intellectual horizon, but it also helps to assess to what degree certain findings have universal bearings. All the countries in this study seek to provide good quality health care to their populations by expanding public or quasi-public health insurance or tax-based health services. Despite this common experience in reform goals, the health politics and policies of these seven countries have not converged into one common direction. The cases reveal a remarkable variety in reform activity, ranging from adjustments of the public-private mix of health insurance and strict regulation of private insurance in Chile, the continuation of the basic sickfund model in Israel, and new regionalized governance models in New Zealand, to quasi-privatized schemes in Singapore and the Netherlands, the implementation of a uniform nationwide single payer social health insurance in Taiwan, to the regulated private insurance within a regionally decentralized health system in Switzerland. While the policy problems, policy goals, and range of options considered were strikingly similar, the countries implemented change within the constraints of existing national institutions and political boundaries.

206 | Okma, Crivelli, Ashton, Tenbensel, Cheng, Chinitz, Lim, Maarse, Bolgani, and Meislin

The timing and speed of change of the health reforms varied as well. In countries with strong central government power and weak or absent organized interests, governments were able to implement major change rapidly. Others, facing strong stakeholder opposition, had to adjust, delay, or even abandon part or all of their reform efforts. In several cases, the introduction of market competition went hand in hand with increased government control. The reality of growing diversity and hybridization (or perhaps, rather, “complexification”) of health care arrangements also illustrates that countries or health care systems do not fit easily within common categorizations. At the level of specific programs and policies, this study did find more similarity in country experiences. For example, in several cases the efforts to change payment methods for medical care took (much) more time than originally envisioned. In most countries, governments softened the effects of market competition by imposing rules and restrictions on both health insurers and providers of care (for example, by mandating entitlements of private health insurance, forcing private insurance to accept everyone seeking insurance, imposing national fee schedules for health services and quality norms for both publicly and privately funded health care). Faced with popular opposition, governments regularly moderated the effects of patient copayments by exempting certain groups. These experiences show that one common feature in health politics is the universal need for “afterreform maintenance.”

Acknowledgements Special thanks to Meryl Schwartz for her extensive editorial assistance.

Notes 1.

2.

This chapter is largely based on an article for a special edition of the Journal of Comparative Policy Analysis (Okma, Crivelli et al. 2010). The study has been a truly international collaborative undertaking. The authors have all lived and worked in one or more of those countries and have brought together a unique degree of in-depth knowledge that allowed for more detailed findings than studies solely based on aggregate data of the OECD or similar international sources. Most international comparative studies take one or more of the world’s large countries as the main comparator: the United States, the United Kingdom, Canada, Germany, France, and sometimes other OECD member states. We use the term “small and medium sized” to indicate countries that clearly do not belong to that group. Harold Wilensky (2002) argues that rather than actual size in terms of population or geographic area, it is the complexity of administration that matters, and small countries can pro-

Health Care Reforms in Seven Small Countries | 207

3.

4.

5.

6.

7.

8.

vide a “laboratory for policy innovation.” Moreover, as we found in this study, small countries offer fewer “exit” opportunities than big ones. Policy elites, managers of large organizations, and representatives of organized interests regularly meet and know each other personally and have little opportunity to leave the policy arena. In the context of health insurance, the term “solidarity”–often taken as “equity”— can operate quite differently. It generally refers to the sharing of the financial burden of health insurance. In one version, equity means that everybody faces the same burden and has access to medical services when he or she needs them. Here, “the same burden” usually means the same share of family income; thus, a proportional tax is fully equal, while community-rated flat premiums are regressive, as higher income groups pay a lower share of their income than people with lower incomes. Another interpretation of “equity,” however, sees it as the absence of an insurmountable financial access barrier to health care. For example, all members of a German sick fund pay the same contribution rates, but the rates differ across funds. The community rating serves as an equalization mechanism among the members, not among the funds. In contrast (until 2006), members of sick funds in the Netherlands all paid the same income-dependent contribution, plus a modest flat-rate premium per person. The British NHS is largely funded through general taxation; the degree of equity thus depends on the redistributive working of the tax system. Each of these three funding mechanisms is perceived as fair or equitable by the general populations, even while the distributional effects are quite divergent. For most patients, in most countries, notions of “consumer choice” in health care refer to their ability to see a health care provider of their own choice. For health policy analysts, the term often refers to an individual’s possibility to select a health insurer or plan. Paradoxically, the increased choice of plan sometimes reduces the choice of provider: when insurers can selectively contract, their members may find that their particular plan has not contracted with their preferred provider. There is some dispute over the question whether it should be labeled public or private—a question that the European Court of Justice, the only authority that can decide on the matter, has not yet considered. Since national competition law, based on European law, prohibits market collusion, insurers and providers in the Dutch health care system had to abandon their long tradition of collective bargaining over fees and tariffs (in Germany, that traditional neocorporatist model is still in place). Nowadays, hospitals, general practitioners, and other independent providers must negotiate individually with each insurer; and health insurers no longer have to contract with every provider. Until the Revised Health Insurance Law that came into effect in 1996, health insurance was optional. Before 1994, six cantons made affiliation to a sickness fund compulsory for the whole population; twelve cantons made affiliation compulsory for special social groups, such as people with a low income and foreigners; and four cantons delegated the decision for a mandate to each municipality. The present system adjusts for thirty age and gender classes. Starting from 2012 the risk equalization formula will include also hospitalization in the previous year as proxy for high risk individuals.

Bibliography Balthasar, Andreas. 2001. Die Sozialpolitische Wirksamkeit der Prämienverbilligung in den Kantonen Monitoring 2000. Bern: Office Fédéral des Assurances Sociales.

208 | Okma, Crivelli, Ashton, Tenbensel, Cheng, Chinitz, Lim, Maarse, Bolgani, and Meislin

Barrientos, Armando. 2002. “Health Policy in Chile: The Return of the Public Sector.” Bulletin of Latin American Research 21, no. 3 (July): 442–59. Bartolomee, Yvette, and Hans Maarse. 2006. “Health Insurance Reform in the Netherlands.” Eurohealth 12, no. 2: 7–9. Beck, Konstantin, Stefan Spycher, Alberto Holly, and Lucien Gardiol. 2003. “Risk Adjustment in Switzerland.” Health Policy 65, no. 1 (July): 63–74. Bentur, Netta, Ayelet Berg, Revital Gross, and David Chinitz. 1998. “Health System Reform and the Elderly: The Case of Israel.” Journal of Aging and Social Policy 10, no. 2: 85–104. Bolgiani, Iva, Luca Crivelli, and Gianfranco Domenighetti. 2006. “The Role of Health Insurance in Regulating the Swiss Health Care System.” Revue Française des Affaires Sociales 60, nos. 2–3: 227–49. Boot, J. Maarten. 1998. “Schaalvergroting en Concentratie in het Nederlandse Ziekenhuiswezen.”[Increasing Scale of Operations and Market Concentration in Dutch Hospitals] Bestuurskunde 7, no. 1: 28–37. Centraal Bureau voor de Statistiek CBS. 2007. Kenmerken van Wanbetalers Zorgverzekeringswet. [Characteristics of Delinquents of the Health Insurance Law] Voorburg: Centraal Bureau voor de Statistiek (CBS). Cheng, Tsung-Mei. 2007. Interviews with officials of the Planning Division, Bureau of National Health Insurance, Department of Health, Taiwan. ―――. E-mail correspsondence with official of the Bureau of National Health Insurance, Department of Health, Taiwan, 2010. ―――. 2005. “Quality of Health Care in Taiwan in the 21st Century: Challenges and Opportunities.” The International Symposium on Achievements and Challenges of National Health Systems in Celebration of the 10th Anniversary of Taiwan’s National Health Insurance, Taipei, March. ―――. “Lessons From Taiwan’s Universal National Health Insurance: A Conversation With Taiwan’s Health Minister Ching-Chuan Yeh.” Health Affairs Vol. 28, no. 4 (July/August 2009): 1043-1044. ―――. 2003. “Taiwan’s New National Health Insurance Program: Genesis and Experience So Far.” Health Affairs 22, no. 3: 61–76. Chinitz, David. 1999. “The Basic Basket of Health Services Under National Health Insurance, Health Policy in Israel.” Social Security 54: 53–68. Choo, Vivien. 2002. “Southeast Asian Countries Vie With Each Other to Woo Foreign Patients.” The Lancet 360, no. 9338: 1004. CIVITAS. 2002. “The Swiss Healthcare System.” London. www.civitas.org.uk/ pdf/Switzerland.pdf (accessed 28 January 2009). Colombo, Francesca. 2001. Towards More Choice in Social Protection? Individual Choice of Insurer in Basic Mandatory Health Insurance in Switzerland, Occasional Papers No. 53. Paris: Organization for Economic Cooperation and Development. Commissie Dekker. 1987. Bereidheid tot Verandering. [Willingness to Change] The Hague: Distributiecentrum Overheidspublicaties. Crivelli, Luca and Iva Bolgani. 2010. “Healthcare Reforms in Switzerland.” In Six Countries, Six Reform Models: The Healthcare Reform Experience of Israel, The Netherlands, New Zealand, Singapore, Switzerland and Taiwan: Healthcare reform ‘Under

Health Care Reforms in Seven Small Countries | 209

The Radar Screen,’ edited by Kieke G.H. Okma and Luca Crivelli. Singapore: World Scientific Publishers. Crivelli, Luca. 2007. “Current Problems and Future Challenges in the Swiss Healthcare System.” In Patient Pathways: The Canton of Ticino Experience, edited by Angela Greco and Maria Cristina Vichi, 3–19. Milan: McGraw-Hill. Crivelli, Luca, Gianfranco Domenighetti, and Massimo Filippini. 2007.“Federalism Versus Social Citizenship: Investigating the Preference for Equity in Health Care.” In Handbook on the Economics of Happiness, edited by Pier L. Porta and Luigino Bruni, 487–511. Cheltenham: Edward Elgar. Crivelli Luca, Massimo Filippini, and Ilaria Mosca. 2006. “Federalism and Regional Health Care Expenditures: An Empirical Analysis for the Swiss Cantons.” Health Economics 15, no. 5: 535–41. Department of Health, Taiwan. 2007. Health Statistical Trends 2006. Taipei: Department of Health. Department of Health. 2009, Health Statistical Trends (in Chinese) Chapter VI [National Health Insurance Survey],”. Department of Health, Taiwan, updated 2011/03/08: Table VI-9. De Roo, Aad A. 1993. “De Zorgsector als Bedrijfstak in Wording.” [Healthcare transforms into Business Sector]Lecture. Tilburg: Catholic University Brabant, 12 September. ―――. 2002. “Naar een Nieuw Zorgstelsel?” [Towards a New Healthcare System?] Lecture. Tilburg: Catholic University Brabant, 14 May. De Swaan, Abram. 1988. In Care of the State: Health Care, Education and Welfare in Europe and the USA in the Modern Era. Oxford: Polity Press. De Viado, Manuel and Alejandro Flores. 1944. “Health Insurance in Chile.” Canadian Medical Association Journal 51, no. 6 (December): 564–70. Douglas, Mary, and Aaron B. Wildavsky. 1982. Risk and Culture: An Essay on the Selection of Technical and Environmental Dangers. Berkeley: University of California Press. Enthoven, Alain, and Wynand P. M. M. van de Ven. 2007. “Going Dutch: Managed-Competition Health Insurance in The Netherlands.” New England Journal of Medicine 357, no. 24 (13 December): 2421–23. Fondo Nacional de Salud (FONASA). 2007. Boletin Estadistico 2005–2006. Santiago: FONASA. Frank, Richard G. and Karine Lamiraud. 2008. “Choice, Price Competition and Complexity in Markets for Health Insurance.” Working Paper No. 13817, Washington DC: National Bureau of Economic Research. Geoffard, Pierre-Yves, Lucien Gardiol, and Chantal Grandchamp. 2006. “Separating Selection and Incentive Effects: An Econometric Study of Swiss Health Insurance Claims Data.” In Competitive Failures in Insurance Markets, edited by Pierre A. Chiappori and Christian Gollier, 81–96. Cambridge: MIT Press. Gross, Revital, Bruce Rosen, and David Chinitz. 1998. “Evaluating the Israeli Health Care Reform: Strategies, Challenges and Lessons.” Health Policy 45, no. 2 (August): 99–117. Harris, Gardiner. 2007. “Looking at Dutch and Swiss Health Systems.” The New York Times, 30 October.

210 | Okma, Crivelli, Ashton, Tenbensel, Cheng, Chinitz, Lim, Maarse, Bolgani, and Meislin

Herzlinger, Regina E., and Ramin Parsa-Parsi. 2004. “Consumer-Driven Health Care: Lessons from Switzerland.” Journal of the America Medical Association 292, no. 10 (8 September): 1213–20. Hill, Michael J. 1993. New Agendas in the Study of the Policy Process. New York/ London: Harvester Wheatsheaf. Hirschman, Albert O. 1970. Exit, Voice and Loyalty: Responses to Decline in Firms, Organizations, and States. Cambridge: Harvard University Press. Höfter, Ricardo H. 2006. “Private Health Insurance and Utilization of Health Services in Chile.” Applied Economics 38, no. 4: 423–39. Hsiao, William H. 1995. “Medical Savings Accounts: Lessons from Singapore.” Health Affairs 14, no. 2: 260–66. Jost, Timothy S. 1999. “Managed Care Regulation: Can We Learn From Others? The Chilean Experience.” Michigan Journal of Law Reform 32, no. 4: 863–98. Kingdon, John W. 1984. Agendas, Alternatives, and Public Policies. Boston: Little, Brown and Company. Klein, Rudolph. 1995. “Learning From Others: Shall the Last be the First Markets?” In Four Country Conference on Health Care Reforms and Health Care Policies in the United States, Canada, Germany and The Netherlands. The Hague: Ministry of Health, Welfare and Sport. ―――. 2006. The New Politics of the NHS: From Creation to Reinvention. Oxford: Radcliffe Publishing. Klein, Rudolph, and Theodore R. Marmor. 2006. “Reflections on Policy Analysis: Putting It Together Again.” In The Oxford Handbook of Public Policy, edited by Michael Moran, Martin Rein, and Robert E. Goodin, 892–912. Oxford: Oxford University Press. Labra, Maria E. 1995. “As Políticas de Saúde No Chile: Entre a Razão e a Força.” [Health Politics in Chile: Between Ratio and Force] In Sistemas de Saúde: Continuidades e Mudanças [Health Care Systems: Continuities and Changes] (Argentina, Brasil, Chile, Espanha, Estados Unidos, México, Québec), edited by Paulo M. Buss and Maria E. Labra. São Paulo: Hucitec; Rio de Janeiro: Editora Fiocruz. ―――. 2002. “La Reinvención Neoliberal de la Inequidad en Chile. El Caso de la Salud.” [the Neoliberal Reinvention of Inequality in Chile] Cadernos de Saúde Pública 18, no. 4: 1041–52. ―――. 2007. “Notas Críticas Sobre Estatificación Socioeconómica, Copagos y Acceso en los Servicios Públicos de Salud de Chile.” [Criticial Notes on Socioeconomic Stratification] Revista Española de Economía de la Salud 6, no. 1: 46–50. Laske-Aldershof, Trea, Erik Schut, Konstantin Beck, Amir Shmueli, and Carine Van de Voorde. 2004. “Consumer Mobility in Social Health Insurance Markets: A Five-Country Comparison.” Applied Health Economics and Health Policy 3, no. 4: 229–41. Lehmann, Hansjörg, and Peter Zweifel. 2004. “Innovation and Risk Selection in Regulated Social Health Insurance.” Journal of Health Economics 23, no. 5: 997–1012 Lijphart, Arend. 1968. Verzuiling, Pacificatie en Kentering in de Nederlandse Politiek. [Pacification and Turnaround in Dutch Politics] Amsterdam: J. H. De Bussy.

Health Care Reforms in Seven Small Countries | 211

Lim, Meng-Kin. 1998. “Health Care Systems in Transition II. Singapore, Part I. An Overview of the Health Care System in Singapore.” Journal of Public Health 20, no. 1: 16–22. ―――. 2004. “Quest for Quality Care and Patient Safety: The Case of Singapore.” Quality and Safety in Health Care 13, no. 1 (February): 71–75. ―――.2004. “Shifting the Burden of Health Care Finance: A Case Study of Public-Private Partnership in Singapore.” Health Policy 69, no. 1: 83–92. ―――. 2005. “Transforming Singapore Health Care: Public-Private Partnership.” Annals of the Academy of Medicine of Singapore 34, no. 7: 461–67. ―――.2010. “Health Care Reforms in Singapore.” In Six Countries, Six Reform Models: The Healthcare Reform Experience of Israel, The Netherlands, New Zealand, Singapore, Switzerland and Taiwan: Healthcare reform ‘Under The Radar Screen.’ Edited by Kieke G.H. Okma and Luca Crivelli. Singapore: World Scientific Publishers. Maarse, Hans and Kieke G. H. Okma. 2004. “The Privatization Paradox in Dutch Health Care.” In Privatization in European Health Care, edited by Hans Maarse. Maarssen: Elsevier Gezondheidszorg. Marmor, Theodore R. 1998. “Hype and Hyperbole: the Rhetoric and Realities of Managerial Reform in Health Care.” Journal of Health Services Research and Policy 3, no. 1: 62–64. Marmor, Theodore R., Kieke G. H. Okma, and Stephen R. Latham. 2006. “Comparative Perspectives on National Values, Institutions and Health Policies.” In Soziologie der Gesundheit, edited by ClausWendt and Christof Wolf, 383–405. Sonderheft: Kölner Zeitschrift für Soziologie und Sozialpsychology. Marmor, Theodore R., Richard Freeman, and Kieke G. H. Okma. 2005. “Comparative Perspectives and Policy Learning in the World of Health Care.” Journal of Comparative Policy Analysis 7, no. 4: 331–48. Ministry of Health, The Netherlands. 2007a. “Oplossing Wanbetalersproblementiek (Nominale) Zorgpremie” [Solution for Delinquent Payers of Health Insurance Premiums] Letter to Parliament, 12 November. The Hague: Kamerstuk. ―――. 2006. Rijksbegroting Volksgezondheid, Welzijn en Sport 2007. [Annual Budget Ministry of Health, Welfare and Sport] The Hague: Ministerie van Volksgezondheid, Welzijn, en Sport. ―――. 2007b. “Structurele Aanpak Wanbetalers.”[Structural Solution Delinquent Payers Health Insurance Premiums] Letter to Parliament, 23 May. The Hague: Kamerstuk. ―――. 2004. Vraag aan Bod. [A Question of Demand] Den Haag: Ministerie van Volksgezondheid, Welzijn, en Sport. ―――. 2005. Ziektekostenverzekeringen in Nederland. Het nieuwe zorgverzekeringsstelsel 2006. [Health Insurance in Holland. The New Health Insurance System] The Hague: Ministerie van Volksgezondheid, Welzijn en Sport. Ministry of Health, New Zealand. 2010. Health Expenditure Trends in New Zealand 1998–2008. Wellington: Ministry of Health. ―――. 2007. “Primary Health Care.”

212 | Okma, Crivelli, Ashton, Tenbensel, Cheng, Chinitz, Lim, Maarse, Bolgani, and Meislin

http://www.moh.govt.nz/moh.nsf/indexmh/phcs-pho (accessed 27 November 2008). ―――. 2007. “Primary Health Care. Implementation Programme 2006–2010.” Primary Workings—Issue 3, June 2007. http://www.moh.govt.nz/moh.nsf/indexmh/phcs-iwp-newsletter-jun07#phofunding (accessed 27 November 2008). Newsweek. 2007. “Singapore Medicine, Singapore—A Global Affair.” 5 October. Naik, G. 2007. “Dutch Treatment: In Holland Some See Model for U.S. Healthcare System Wall Street Journal 6 September. OECD. 2006. OECD Reviews of Health Systems: Switzerland. Paris: Organization for Economic Cooperation and Development. ―――. 1992. The Reform of Health Care: A Comparative Analysis of Seven OECD Countries. Health Reform Studies No. 2. Paris: Organization for Economic Cooperation and Development. ―――. 1994. The Reform of Health Care: A Review of Seventeen OECD Countries. Health Reform Studies No. 5. Paris: Organization for Economic Cooperation and Development. Okma, Kieke G. H. 2007. “Dutch Treatment: In Holland Some See Model for U.S. Health-care System.” Response to G. Naik’s article. Wall Street Journal website, 6 September. ―――. 2002. “Health Care and the Welfare State: Two Worlds of Welfare Drifting Apart?” In Social Security in Transition, edited by Jos Berghman et al., 229–38. Leiden: Kluwer Law International. ―――. 2008. “Recent Change in Dutch Health Insurance: Individual Mandate or Social Health Insurance?” Paper presented at the Annual Meeting of the National Academy of Social Insurance. Washington, D.C., 30–31 January. ―――. 1997. “Studies in Dutch Health Politics, Policies and Law.” PhD diss., University of Utrecht, Utrecht. Okma, Kieke G. H. and Aad A. De Roo. 2009. “The Netherlands: from Polder Model to Modern Management.” In Comparative Studies and the Politics of Modern Medical Care, edited by Theodore R. Marmor, Richard Freeman, and Kieke G. H. Okma, 120–52. New Haven: Yale University Press. Okma, Kieke G. H., Luca Crivelli, eds. 2010. Six Countries, Six Reform Models: The Healthcare Reform Experience of Israel, The Netherlands, New Zealand, Singapore, Switzerland and Taiwan: Healthcare reform ‘Under The Radar Screen.’ Singapore: World Scientific Publishers. Okma, Kieke G. H., and Machteld Ooijens. 2005. “De Patient Centraal. Zijn de Verenigde Staten ons Voorland?” The Patient at the Centre: The United States as Our Example?] Tijdschrift voor de Sociale Sector, 26–29. June. Palmer, George R. and Stephanie D. Short. 1989. Health Care and Public Policy—An Australian Analysis. Melbourne: The Macmillan Company of Australia Pty Ltd. Phua, Kai H. 1991. Privatization and Restructuring of Health Services in Singapore. Occasional Paper No. 5, Institute of Policy Studies. Singapore: Time Academic Press. Pierson, Paul. 1994. Dismantling the Welfare State? Reagan, Thatcher and the Politics of Retrenchment. Cambridge: Press Syndicate of the University of Cambridge.

Health Care Reforms in Seven Small Countries | 213

Preker, Alexander S. and April Harding. 2003. Innovations in Health Service Delivery: The Corporatization of Public Hospitals. Washington, D.C.: World Bank Publications. Ranade, Wendy, ed. 1998. Markets and Health Care: A Comparative Analysis. New York: Addison Wesley Longman. “Regeerakkoord 2007.” 2007. Samen werken, samen leven. Kabinetsformatie 2006. Brief van de Informateur [Governing Accord 2006. “Working Together, Living Together”. Letter of the Queen’s Advisor to Parliament]. Kamerstukken 2006-2007, 30891, no 4. 7 February. [Documents of the Second Chamber of Parliament]. Reinhardt, Uwe E. 2004. “The Swiss Health System: Regulated Competition Without Managed Care.” Journal of the American Medical Association 292, no. 10 (8 September): 1227–31. Rosen, Bruce, Rachel Goldwag, Sarah Thompson. 2003. Health Care Systems in Transition, Israel. Copenhagen: European Observatory of Health Care Systems. Raad voor de Volksgezondheid & Zorg RVZ. 2003. Marktconcentratie in de Ziekenhuiszorg. [Market Concentration in Hospital Care] Zoetermeer: Raad voor de Volksgezondheid & Zorg). Sapelli, Claudio. 2004. “Risk Segmentation and Equity in the Chilean Mandatory Health Insurance System.” Social Science and Medicine 58, no. 2 (January): 259–65. Scheerder, Robert. 2005. “Kostenbeleid 2005.” [Cost Control Policy 2005] Zorg & Financiering 193, no. 2: 8–28. Schut, Frederick T. and Wynand P. M. M. van de Ven. 2005. “Rationing and Competition in the Dutch Health-Care System.” Health Economics 14, no. S1: S59–S74. Smit, Marieke and Philip Mokveld. 2007. Verzekerdenmobiliteit en Keuzegedrag. [Mobility and Choice of Insured in the Dutch health Insurance Market] Zeist: Vektis. Strategisch Akkoord 2002. 2001. [Strategic Accord] Kamerstukken II. 2001–2002, 28375, no. 5. [Documents of The Second Chamber of Parliament] The Hague: SDU. Strombom, Bruce A., Thomas C. Buchmueller, and Paul J. Feldstein. 2002. “Switching Costs, Price Sensitivity and Health Plan Choice.” Journal of Health Economics 21, no. 1: 89–116. “Tandarts Terug in het Basispakket.” 2007. [Dental Care Back into the Basic Insurance Coverage] 2007. Press release May 29. www.gezondheidsplein.nl Tiebout, Charles. 1956. “A Pure Theory of Local Expenditures.” Journal of Political Economy 64, no. 5 (October): 416–24. Timmins, Nicholas. 1995. The Five Giants, A Biography of the Welfare State. London: Fontana Press. Van de Ven, Wynand P. M. M., and Frederick T. Schut. 2008. “Universal Mandatory Health Insurance In the Netherlands: A Model For The United States?” Health Affairs 27, no. 3: 771–81. Wilensky, Harold L. 2002. Rich Democracies, Political Economy, Public Policy, and Performance. Berkeley: University of California Press.

214 | Okma, Crivelli, Ashton, Tenbensel, Cheng, Chinitz, Lim, Maarse, Bolgani, and Meislin

WHO Bulletin. 2008. “Public tensions, private woes in Chile”. Bulletin of the World Health Organization, vol 86, no 11: 817-908. Woolhandler, Steffie, Terry Campbell, and David U. Himmelstein. 2003. “Costs of Health Care Administration in the United States and Canada.” New England Journal of Medicine 349, no. 8 (21 August): 768–75. World Bank. 2006. World Development Indicators 2006. Washington, DC: The World Bank. Yong Y. L. 1967. Speech by the Minister of Health at the opening of the WHO Seminar on Health Planning in Urban Development. Singapore, 21 November.

Chapter 9

HUMBOLDT HUMBLED? The Germanic University System in Comparative Perspective

( Ben W. Ansell

Introduction The 1980s were a period of American paeans to the Germanic economic model of high value-added engineering and manufacturing, with much credit given to the vocational system of training developed in Austria, Germany, and Switzerland. Germanic comparative advantage was seen as deriving from the “German skills machine” (Culpepper and Finegold 1999). However, as the economic fortunes of America and Central Europe reversed during the late 1990s and early 2000s, the rise of technologically driven American sectors driven by a new “creative class” led to a change in the attribution of success (Florida 2004). Now universities are seen as the engines of American growth, whereas Germanic universities are pronounced “rotten to the core” (Ash 1997). This is a somewhat surprising critique, since universities in the German-speaking world had been the envy of Europe in the late nineteenth and early twentieth centuries. The “Humboldtian” mission of intertwined teaching and research was, in fact, the inspiration behind the founding of Johns Hopkins University in 1876 (Fischer-Appelt 1996). Even in the 1950s and 1960s, following wartime destruction and de-Nazification, German universities were larger institutions than most of their European peers. But since then, the pace has slowed. While the Anglo-American and Scandinavian countries have reached a level where a majority of each age cohort attends higher education, Germany, Austria, and Switzerland remain

216 | Ben W. Ansell

pinned to an enrollment level of just one-third of the population—less than half the size of Swedish enrollment. Given the importance of human capital development to economic growth and innovation, many commentators attribute the recent economic stumbles of Germanic states to this lagging level of higher-education enrollment. Is Germanic education really, as Glotz (1996) put it, verrottet? The political debate during the 1990s and the last decade has identified a number of significant problems with the higher-education systems in Germany and Austria: low enrollment, lengthy study periods, overcrowded classrooms, underfunded institutions, and esoteric training undertaken by an aging, chauvinist professoriate. Put together, these complaints paint a rather doleful picture of German and Austrian universities. The Swiss situation is distinct, with continued low levels of enrollment but much higher perstudent funding than Germany or Austria. The low levels of enrollment in all three states are dramatic in comparative context. Table 9.1 shows data on the level of tertiary education attained by various age groups in the population at large across the OECD (OECD 2005).1 Table 9.1. Higher education attainment in the population, 2004 Age 25–64

25–34

35–44

45–54

55–64 22

Australia

33

41

32

32

Austria

15

20

17

13

8

Belgium

31

45

35

25

17 35

Canada

48

60

51

43

Czech Republic

11

13

12

10

8

Denmark

35

40

37

36

24

Finland

38

47

47

36

24

France

25

41

25

18

14

Germany

21

23

23

21

15

Greece

19

27

24

15

8

Hungary

18

22

20

16

12

Iceland

30

38

37

24

13

Ireland

29

44

29

21

15

Italy

12

17

13

11

6

Japan

35

54

44

29

14

Korea

25

48

25

9

5

Mexico

14

18

15

11

5

Netherlands

27

36

28

25

18

Humboldt Humbled? The Germanic University System in Comparative Perspective | 217

Age 25–64

25–34

35–44

45–54

55–64

New Zealand

28

31

28

29

22

Norway

34

46

37

30

20

Poland

17

28

16

13

11

Portugal

15

24

15

11

7

Spain

26

42

29

17

8

Sweden

38

47

38

36

29

Switzerland

19

23

22

18

12

Turkey

7

9

5

6

3

United Kingdom

29

35

29

26

21

United States

40

42

41

41

33

OECD average

25

34

27

22

15

Source: OECD Education at a Glance, 2006

The Germanic cases stand out. While the OECD average attainment among 25- to 64-year-olds is one-quarter of the population, Germany has only 21 percent with tertiary education, Switzerland 19 percent, and Austria lags behind with 15 percent, barely above Mexico. It is worth noting, though, that the Germanic experience is not sui generis. A number of other continental countries have even lower levels of tertiary education attainment, for example, Italy. Consequently, this kind of broad variation among states begs a comparative analysis that can explain these patterns and situate the Germanic cases. What explains this failure of the Germanic systems to attain a true “mass” level of enrollment? And why are Austrian and German, though not Swiss, universities underfunded? This chapter, building on earlier work by its author (Ansell 2008), situates the Germanic countries in comparative perspective, demonstrating how governments are subject to a “trilemma” in higher education that forces them to choose between three types of systems: “elite,” “mass public,” and “partially private.” I argue that, for both institutional and political reasons, Germany, Switzerland, and Austria have as yet been unable to break out of the elite model into a mass model, either through increased public funding (mass public) or the introduction of tuition fees (partially private). First, the veto power of subnational units of government provides a powerful institutional block to greater enrollment. Second, the long-run dominance of center-right Christian Democratic governments uninterested in mass higher education has stymied demands for expansion.

218 | Ben W. Ansell

I begin with an overview of the current alleged “crisis” in higher education in Germany, Switzerland, and Austria. I then situate the Germanic experience in a comparative context, outlining a typology of elite, mass public, and partially private higher-education systems. I argue that partisan preferences and institutional constraints determine the choice of system. I then turn to a case historical analysis of the German, Austrian, and Swiss experiences with higher education, showing how the institutional constraints of federalism and streaming of secondary education combined with unfavorable partisan coalitions have limited expansion to mass higher education in these countries. I conclude with some predictions for the future of higher education in the Germanic states.

Explaining the Germanic Experience in Comparative Perspective Unlike the Germanic states, many other countries have recently moved from an elite model of higher education to a mass enrollment system. I argue (Ansell 2008) that this move has involved making political choices within a “trilemma.” If a government chooses to expand enrollment, it must choose between either maintaining the rate of public subsidization while increasing the total public cost of provision or reducing the rate of public subsidization in order to maintain total public cost.2 These tradeoffs can be simply characterized by using the “ideal types” of elite, partially private, and mass public systems. With a low public cost, some states will have high enrollment at a lower level of subsidization (partially private), whereas others will have higher subsidization but lower enrollment (elite). With a mass level of enrollment, some states will have low subsidization and low public cost (partially private), while others will have high subsidization and high public cost (mass public). Finally, if higher education is fully subsidized, some states will have low enrollment at low public cost (elite) and others will have high enrollment at high cost (mass public). Table 9.2 sets out the distinction among the systems. Table 9.2. The trilemma in higher education Public Funding

Public/Private Funding

Low Enrollment

Mass Enrollment

Elite System (Germany) Low Public Cost

Mass Public System (Sweden) High Public Cost Partially Private System (England) Low Public Cost

Note: No advanced industrial country combines low enrollment with private funding. This configuration exists in many developing countries.

Humboldt Humbled? The Germanic University System in Comparative Perspective | 219

Germany, Switzerland, and Austria remain wedded to an elite model. This raises two questions: Why have they not transitioned to one of the mass models? And which mass model are they likely headed toward, if any? The case studies in the following three sections broach these questions directly. Before heading to the specific case histories, it is useful, however, to demonstrate the general cross-national pattern of a trilemma. I draw here on cross-sectional data on enrollment, subsidization, and public cost for nineteen OECD states. The gross enrollment rate in higher education in 2002 ranged from one- third of the university-age cohort in Austria and Germany to over 80 percent in New Zealand and Sweden. The degree of subsidization also varies widely. Some systems, like Denmark and Finland, are almost totally publicly funded, whereas others, like the United States and Australia, are less than 50 percent publicly funded, with the remainder coming from private sources. Finally, some highereducation systems take relatively little from the public purse—less than 1 percent of GDP in England and Italy—while others, like Finland and Sweden, cost over 1.5 percent of the GDP. Figure 9.1, taken from Ansell (2008), plots the empirical relationship between gross tertiary enrollment against the share of tertiary spending from private sources (that is, inverse subsidization) for nineteen OECD states in 2002.3 For each state I label both its three-letter country code and its level of overall public spending on higher education. Figure 9.1. Gross enrollment as a percentage of age cohort, private spending as a percentage of total spending, and public cost as a percentage of GDP, 2002 (from Ansell 2008)

220 | Ben W. Ansell

In bivariate terms there does not appear to be a strong relationship between the variables; however, once we examine the public spending variable, we see an interesting pattern. There are three distinct southwest-northeast lines in the figure. In the northwest corner are the two East Asian states with low levels of public funding. Along the main diagonal axis are a dozen states spending between 0.8 percent and 1.2 percent of GDP on publicly funded higher education. In the southwest corner are the elite states like Belgium, Austria, and Germany, with low enrollment and low private financing. At the northeast end are the United States, Australia, and New Zealand, with high enrollment and higher private spending but similar levels of public spending to the elite states. Finally, we have a line of states in the southeast corner: the Nordic mass public states, which spend at least 1.4 percent of GDP on tertiary education, with high enrollment and high subsidization. Austria and Germany sit right at the extreme bottom left of Figure 9.1. Among those states that spend between 0.8 percent and 1.2 percent of national income on public funding of higher education, they have both the lowest levels of enrollment and the lowest levels of private funding. Thus the Germanic states (along with Belgium) are the paradigmatic cases of elite higher-education systems. This low enrollment is not solely a function of having high levels of vocational training or high-end manufacturing, since the Scandinavian states sit at the bottom right. Austria and Germany could achieve Scandinavian or American levels of gross enrollment in two ways. Either they could increase their overall public spending to over 1.5 percent of GDP, or they could increase the share of private spending to around 50 percent of all higher education spending from the current level of 10 percent. These are large shifts in financing, and they do not come without political winners and losers. Unfortunately, there is no available data on private spending for Switzerland, hence it is not plotted—however, because of its low levels of enrollment and private spending, it would cluster with the elite states, albeit with a higher level of public spending (1.4 percent of GDP). Hence Switzerland provides us with an intriguing puzzle since it “breaks” the trilemma. The Swiss have increased public spending without increasing enrollment. Given their comparative position as elite states, and the Swiss anomaly, what might explain the particular choices the Germanic states have made about higher-education policy?

Partisan Preferences and the Trilemma Perhaps the key political factor related to higher education is its fiscal regressivity. Publicly provided higher education is typically received by the

Humboldt Humbled? The Germanic University System in Comparative Perspective | 221

wealthier members of society but is paid for by all taxpayers. Consequently, unless access is made independent of income, or financing is privatized, spending on higher education is fiscally regressive. Accordingly, we can derive some simple expectations about how voters with different incomes might feel about different higher-education systems (Ansell 2008). We would expect poorer voters to dislike spending more on higher education for a given level of enrollment, and for wealthier voters, who actually receive higher education, to support it. This reverses the standard view of partisan preferences over public spending associated with, for example, the “Power Resources” literature (Korpi 1983). As enrollment expands to a mass level, this regressivity becomes diluted and higher education begins to look more like standard redistributive public spending. Moving from increased spending at a fixed level of enrollment to increasing the level of enrollment itself, we obtain slightly different conclusions. The chief beneficiaries of increased enrollment are those who stand to receive higher education for the first time. If access is positively correlated with income, the new beneficiaries are likely to be fairly well-off, albeit not so well-off that they were already recipients. Consequently, we should expect the upper middle class to be the recipients of expansion in an elite system and the middle to lower-middle class to be beneficiaries of expansion once enrollment has reached a mass level. This suggests that preferences over expanded enrollment and over spending both depend on the existing level of enrollment. How are such preferences aggregated upward into policies? Parties with little support from the middle of the income distribution are likely to try to limit enrollment and retain an elite model, since only the middle class directly benefits from expansion. Of course, expanding or restricting enrollment may be out of the hands of national parties. As for spending, when enrollment is low, it is right-wing parties that favor increasing spending, as it benefits their wealthier constituents. If they do expand enrollment levels they will maintain public financing and consequently move toward a mass public system, as occurred in Sweden (Ansell 2008). Left-wing parties, conversely, will tend to neglect aggregate public spending on elite higher-education systems or even try and introduce tuition fees to reduce the burden of paying for the education of richer citizens. Enrollment expansion under the Left tends to lead to partially private systems, as in Australia and England. However, once enrollment has passed a particular threshold—around 40 percent—it is increasingly the leftwing parties that favor increased tertiary spending as their constituents become the new targets. Thus the partisan story behind the journey to a mass system (and consequently the transition from elite to mass public or partially private system) is quite distinct from partisan preferences once

222 | Ben W. Ansell

a truly mass system has been obtained. What are the implications for the Germanic states? Austria, Switzerland, and Germany currently have gross enrollment levels of around one-third of the population and thus are at the point at which the partisan politics of higher-education spending are becoming murky. While in the past, at lower levels of enrollment, it was the right-wing Germanic parties that favored increased spending, currently there is no clear pattern. If anyone, it is likely to be left-wing parties that favor increased funding as the Germanic states shift to a mass model. A further political option is a “grand coalition” of Left and Right parties—here we should expect enrollment expansion to occur since the middle class no longer splits its vote across two parties, but there are fewer clear implications on financing, given the contrasting views of Left and Right. In fact, this is the case we shall see most clearly with Germany and Austria—expansion of enrollment occurred during “grand coalitions,” but the funding issue was not resolved due to disagreement between parties—eventually leading to expansion without sufficient funds and a consequent decline in the quality of higher education. The Swiss, conversely, never had a socialist party dominant enough to break the general rightwing tone of government, and this resulted in little expansion but high per-student spending. One other policy that parties can control is the degree to which access to the university is dependent upon family income. Parties seeking to reduce this kind of dependence on income can subsidize entry by poor groups, either through class-based affirmative action or through targeted grants and loans. Another strategy is altering the structure of schooling to promote access of poorer students, perhaps by removing streaming. Since poorer voters do better in income-independent systems than in incomedependent ones, left-wing parties seek to make access more income independent and spend resources on grants and loans for poorer students. The reverse applies for right-wing parties. We will see a variety of examples of attempts at reducing income dependence by left-wing parties in Germany and Austria and an attempt at increasing income-dependence by the center-right alliance in Austria. Right-wing dominance at the federal level in Switzerland may explain why its funding of student support like grants and loans is some of the least generous in the OECD and why relatively few students take the Maturat university qualification exam. One final political factor must be noted, since it is particularly germane to these three states. The arguments above assume that parties have some control over higher-education policy; however it is not obvious that parties actually have the freedom to alter higher-education policy without institutional constraints. If veto players can totally block reform, the elite system will remain in place, as we shall see in the cases of Germany and

Humboldt Humbled? The Germanic University System in Comparative Perspective | 223

Austria, where the sub-national states - the Länder - have been powerful impediments to change, and especially in Switzerland, where cantonal veto power has been an especially strong obstacle to expansion.

The Trilemma and German Higher Education Since 1945 I now turn to the historical record to examine these propositions more closely. I begin in this section with Germany, before turning in the following sections to Austria and Switzerland.4 Germany has a storied tradition of higher education. The system of research universities established in nineteenth-century Prussia became a model for many of the world’s best universities, especially in America (Fischer-Appelt 1996). Indeed, in 1960 German higher education was as expansive as any in Europe: around 4 percent of each cohort of school-leavers entered the higher-education system. However Germany faced a set of institutional constraints that would make future expansion and reform of higher education much more difficult than in other large European states. First, Germany had sixteen higher-education systems rather than one, since the Basic Law provided the Länder with control over universities, creating a broad set of potential veto players. Coordination between the Länder was slow and piecemeal: Helmut Kohl referred to the council of Länder education ministers as “the most reactionary institution in the Federal Republic” (Hüfner 2003). Second, each year’s higher-education budget was negotiated in the Länder legislations as a lump sum rather than as per-student funding, as in England and Sweden (Konow 1996). This meant that expansion of student numbers would not necessarily produce increased funding. Third, the structure of German law, based on the Basic Law, meant that reform to the higher-education system would have to be achieved through constitutional mechanisms. Finally, Germany had a streamed schooling structure that limited university access to students from the academic Gymnasien (despite the occupying forces’ postwar attempt to integrate schools, see Nugent 2004). Although enrollment had expanded to 10 percent by 1970, universities struggled to meet this rising demand, with both Länder and universities extremely cautious about increased enrollment.5 In early 1969, the Christian Democrat (CDU-CSU) and Social Democrat (SPD) grand-coalition government passed an amendment to the Basic Law that stipulated a new role for the federal government in higher education, particularly in terms of funding university expansions, for which the federal government would pay 50 percent. The grand coalition permitted legislative expansion of higher education since the middle class—the chief beneficiary of expan-

224 | Ben W. Ansell

sion—was no longer split across parties. However partisan preferences sharpened following the Social Democrats’ victory in late 1969. This leftwing SPD government, which would control German government until 1982, introduced a variety of higher-education reforms reflecting their particular ideological concerns. Chief among these SPD reforms was the introduction of the BAFöG (Bundesausbildungsförderungsgesetz) system of student financing. This provided student grants aimed at low-income students, with two-thirds of funding coming from the federal government (Onestini 2002). This was followed in 1974 with the introduction of student loans to help working class children access higher education. The countervailing position of the CDU-CSU became clear in 1983, one year after entering office, when they changed the BAFöG system into an interest-free loan instead of a grant (Lingens 1998). The CDU-CSU were content to let inflation reduce the number of eligible recipients from 44.6 percent in 1972 to just 12.6 percent by the time the Social Democrats returned to power in 1998 (Hüfner 2003). Gerhard Schröder’s government reacted to this starving of the BAFöG system by increasing the generosity of grants and eligibility significantly in 2002. In terms of increasing the progressivity of access to university, partisan patterns played out in Germany with the Left pushing grants and access for the working class and the Right limiting such support. In terms of overall expanded access, however, neither was willing to encourage the transition to a mass system, partly because of the institutional constraint of overriding the Länder and partly because of the SPD’s disinterest in encouraging middle class access (Ostermann 2002). The SPD government instituted a numerus clausus national quota system for enrollment in 1972 and a lottery system in 1976, which increased the probability of working class children receiving one of the limited number of places (Nugent 2004). Thus, the SPD kept overall enrollment low, thereby avoiding benefiting the middle class constituents of the conservative opposition while still promoting working class access. Yet the size of the overall system continued to stagnate once the conservative CDU-CSU-FDP coalition came to power in 1982. Between 1985 and 1995, gross enrollment increased from 18 to 25 percent only. Underlying this stagnation was one of the Kohl government’s final acts of legislation: the 1998 amendment to the Higher Education Framework Act. This amendment established quotas on the number of students entering medicine, dentistry, architecture, business management, or psychology. These restrictions allowed the CDU/CSU to limit entry into a number of the professions best represented in their constituency. The return of the SPD to power further retarded expansion: by 2000 the rate of attendance

Humboldt Humbled? The Germanic University System in Comparative Perspective | 225

had risen only to 27 percent, almost twenty points lower than England and Sweden (Hüfner 2003). Why has German enrollment remained low? Three potential explanations emerge. First, the streamed nature of German secondary education remains in place, thus limiting the number of students eligible for university. Baker and Lenhardt (2008, p.55) argue that “at the core of the resistance to the expansion of higher education is the extreme selectivity of German secondary education.” Second, the Länder have had control over the admissions process and appear to have been unwilling to increase funding or enrollment, thereby acting as veto players. Third, the SPD has been unwilling to countenance expansion of a system where few working class students gain access (in the 1990s, only 12 percent of university students had working-class parents), while the CDU/CSU have for their part enrolled the lion’s share of their support (Windolf 1997). In 1997 there were widespread national protests over the underfunding of higher education; however, the SPD was unable to break the endemic “Reformstau” (reform deadlock), and many Germans now believe a crisis point has been reached. There is little consensus on whether a partially private or a mass-public path is the way forward. The SPD split in 1996 in reaction to their education spokesman’s suggestion of implementing fees for higher education. They also appeared to reject a partially private system by passing a bill in 2002 that prevented tuition fees (Führ and Tapia 1997). The CDU/CSU have made recent rhetorical nods toward fees. Nonetheless, they are unlikely to be able to enact these fees, since they would directly hurt their own constituency of the upper middle class, who are the vast majority of university students. Ironically, a grand coalition, as in the late 1960s, could be the savior of expansion. Or, since federal reform is constrained, change may come from the veto players themselves—the Länder, which managed to overturn the SPD ban on fees in the Federal Constitutional Court in 2005. The state of Nordrhein-Westfalen introduced limited tuition fees in 2006 to much student opposition. However, these fees are extraordinarily low by contemporary standards, around €500 per semester, and they are more likely to impede access for credit-constrained working-class students than they are likely to change the funding of higher education so as to lessen the fiscal burden on the working class. Put simply, minimal fees have a negative net impact for poorer citizens, a pattern we shall see once more in the Austrian case. It is not surprising that it has been Länder-dominated, conservative governments that have introduced fees, since they are likely to bias access toward upper-middleclass students at fairly minimal cost to their constituency. For the most part, though, the German system remains in stasis: quotas, streaming, and the Länder have blocked transition to mass higher education.

226 | Ben W. Ansell

The Trilemma and Austrian Higher Education Since 1945 While the history of the German higher-education system since 1945 conforms to the effects of partisanship and federalism on higher education discussed in Section Two, Austria appears more anomalous. In Austria, the partisan pattern of tuition fees seen in England, New Zealand, and Australia—with the Left imposing fees and the Right opposing fees—is reversed. It was the left-wing Sozialdemokratische Partei Österreichs (SPÖ) that removed existing fees in the 1970s, and the (far) right-wing coalition between the Österreichische Volkspartei (ÖVP) and the Freiheitliche Partei Österreichs (FPÖ) that reintroduced them in 2001. Furthermore, the major expansionary period of Austrian higher education in the late 1960s and early 1970s happened under the left-wing SPÖ, even though most of the initial beneficiaries of this expansion were the typical constituents of the right. Finally, the federal government, rather than the Länder, has been responsible for higher education, meaning that the institutional impediments to reform that put an end to change in Germany were less important in the Austrian case. So how are we to understand Austria in the context of the theoretical claims and empirical findings of Section Two? In fact, on closer examination, the pattern of policy making in Austria followed many of the partisan and institutional trends experienced in the broader OECD. The specific pattern of left-wing parties emphasizing reducing the income dependence of access to university and right-wing parties being less inclined toward such measures takes particular hold in Austria. Moreover, though the Austrian Länder are not as strong veto players as their German equivalents, the broader context of labor market institutions favoring upper-secondary vocational training has also constrained expansion. In this section, I follow the two main “reform cycles” in Austrian higher education—the intrusion of the federal government into the autonomous Humboldtian system in the 1960s and early 1970s and reforms marking a retreat of the state in the early 2000s (Pechar 2005). As with most other European countries, the Austrian education system in the early postwar era was elite in both enrollment and attitude. Austrian universities cherished their autonomy and were largely run by the professoriate, with little input from students or junior faculty “below” or rectors or bureaucrats “above.” Fewer than 5 percent of each age cohort attended university, and those that did were almost exclusively graduates of the Gymnasium system, which selected students through exams at age ten. Universities were state funded but had tuition fees that acted as a barrier to low-income students. Furthermore, because vocational training was concentrated at the secondary level in specially streamed schools, there

Humboldt Humbled? The Germanic University System in Comparative Perspective | 227

was no system of technical colleges alongside the universities. Developing a mass higher-education system would thus mean a broad array of reforms: to the power of the professors, to the streamed secondary-school system, to the funding mechanism, and to the overall enrollment capacity of the universities. As with Germany, it took a “grand coalition” of the Left and Right in the 1960s to move the higher-education system haltingly in this direction. Following the 1962 election, the SPÖ and ÖVP formed a coalition government that controlled policy making until 1966. During this period the first steps toward a mass higher-education system were taken. In 1963, the Austrian system of student aid was established to permit broader enrollment beyond the upper middle class. Then, in 1966, a major higher-education restructuring bill was passed. The “Allgemeines Hochschulstudiengesetz,” or “General University Studies Act,” established new guidelines for increased enrollment and reorganized the country’s curricular and degree-granting structure (Beerkens 2003). As with the German reforms of 1969, it took a “grand coalition” to pass these major changes to university financing and organization—reflecting the increased power of the “middle orders” branching across the two parties. The act of 1966 remained the guiding principle for Austrian higher education until the next “grand coalition” of the 1990s. But 1966 did not mark the end of the first “reform cycle” in Austrian higher education. Although there was little activity in higher-education policy during the period of sole ÖVP rule between 1966 and 1970, the return to power of the Social Democrats in 1970 led to a series of reforms that mimicked the objectives of the German SPD, emphasizing increasing the access of working class children. The SPÖ’s focus was on reducing the income dependence of the Austrian university system, and this necessitated dealing with the root problem of school streaming at the secondary level. While the split between academic Gymnasien and vocational Hauptschule remains to this day, the Social Democrats were able in 1971 to abolish the entrance-exam system that biased the Gymnasien toward accepting the children of the upper middle class and therefore tilted university intake to these same students. In 1972, the SPÖ followed their attempt to introduce income independence at the secondary level by doing so at the tertiary level through the abolition of tuition fees. Why did the SPÖ remove fees if this potentially constituted a fiscally regressive policy? By 1972, Austrian tuition fees were fairly meager, since they had never been indexed to inflation. Consequently, abolishing them did not require a large compensatory increase in public support. Thus, the fees disincentivized access by the working class without reducing the burden of taxation. More critical for the SPÖ was ensuring that overall ex-

228 | Ben W. Ansell

penditure on higher education did not become a major budgetary burden that might impinge on more favored public policies. Consequently, into the mid 1980s the SPÖ was content to allow per-student public funding to decline, and they attempted to keep enrollment fairly limited. Their interest was in keeping the burden of taxation low for the working class, who were by and large still excluded from higher education. When they did engage in higher-education reform, it was focused on removing barriers to access for working-class students. The long period of Social Democratic dominance thus marked a lull in higher-education policy in Austria, even as the economy shifted toward the service sector and demand for mass university education—and as other European states began to “massify” their university systems. Though the second “grand coalition” began in 1986, it was not until 1992 that policy making for higher education began again in earnest. Once more, coalition government permitted a further shift toward a mass higher-education system whose beneficiaries would be the broad middle class rather than just the children of professionals and the elite. During the SPÖ-ÖVP coalition, which lasted until 2000, a flurry of structural reforms were passed. In 1992, the coalition passed the Study Support Act, updating the 1963 act mentioned above, which moved toward direct support of students, granting up to the equivalent of €7,800 per annum. In 1993, they established the Fachhochschule, a form of technical college, which provided a more vocationally oriented alternative to universities. These colleges have been highly successful and have aided in expanding Austrian enrollment significantly, accounting for almost 10 percent of enrollments. The 1993 University Organization Act further marked a retreat of the federal government from control of the universities, granting them significant autonomy while financially empowering rectors. In 1997, the coalition passed a revision to the 1966 General Universities Studies Act, which further reformed the structural organization of the degree system, with an amendment in 1999 introducing standard bachelor’s and master’s degrees. Thus, as in the previous period of “grand coalition,” large-scale structural reforms—presumably the underpinning of any transition to a full mass higher-education system—were precipitated by cross-party cooperation. While the mid-1990s saw major shifts in Austrian higher-education policy, these shifts were only a prelude to a further shift in the first decade of the twenty-first century, following the seismic political victory of Jörg Haider’s FPÖ in 2000, which ushered in a right-wing coalition with the ÖVP. The shift to the right was marked also in higher-education policy, with the introduction of a set of “new managerial” reforms, inspired by

Humboldt Humbled? The Germanic University System in Comparative Perspective | 229

the New Public Management literature popular among neoliberals in the 1990s. In 2002, the Organizational Act was passed, removing universities from state control and establishing them as “legal persons under public law,” along the lines of the British model, and providing them with lumpsum budgets under control of the rector, removing them from the broader federal budget. This structural reorganization was accompanied by the 2001 reintroduction of tuition fees. Given the argument I made in Section Two, which suggested right-wing parties should favor public funding of higher education, how can we reconcile these findings? While superficially the laws of 2001 and 2002 may seem anomalous given my broader argument about partisan preferences, they are on closer examination less surprising. First, regarding tuition fees, these were set at such a low level (€363 per semester) that they barely changed the overall tax burden of higher-education funding. Consequently, poorer citizens who never attended university still paid taxes to support those who did attend. In contrast, in the United Kingdom, New Labour set fees at a maximum of £3,000 per annum in 2003, around six times the Austrian level. Unlike the UK, the Austrian fees were to be paid up front and were not adjusted for socioeconomic status—consequently they amounted to a regressive fee without noticeably reducing the overall tax burden of funding the public share of higher education. This fits with the pattern of left-wing governments introducing sizeable fees that replace over one-third of public spending (England, Australia, New Zealand) and right-wing governments introducing much lower fees that mostly limit enrollment rather than introduce progressivity (Austria, some German Länder; see Marcucci and Johnstone 2007). Second, in terms of the removal of universities from the federal budget and making them legally independent bodies, this shored up the highereducation system against the potential tide of mass enrollment. The 2002 law meant that the federal government would have less control over admissions, which satisfied the Right’s aim to keep enrollment fairly limited, while also potentially preventing the kinds of budgetary constraints that reduced higher-education spending under Social Democrat control in the 1970s. Thus, the FPÖ-ÖVP reforms achieved the aims of maintaining, and enhancing, the income-dependent nature of access to higher education and shielding a largely publicly funded elite system from budgetary fluctuation and encroaching federal policy. Thus, the anomalous nature of these reforms appears rather less peculiar. Overall, the Austrian case does closely resemble the patterns suggested in Section Two. Since 1960, the Left has been more interested than the Right in reducing the income dependence of access; major structural expansions have generally occurred under “grand coalitions,” which typically favor

230 | Ben W. Ansell

expanded enrollment; the Left has trimmed back overall public funding for what remains an elite system. Even the anomaly of the Right introducing tuition fees can be explained by reference to the particular form of fee used: a low up-front fee that acted more as a deterrent to access of poorer students than as a way of “progressivizing” the funding of higher education. Consequently, Austria retains an elite higher-education system that despite superficial appearances has not moved much toward a mass “partially private” system. The slow expansion toward a third of students participating in higher education means that the Social Democrats are gradually becoming more favorable toward funding the system, with the Right attempting to maintain the elite system as is—hence the 2002 granting of legal “independence” to universities. As yet, a major breakthrough toward a mass system is not on the horizon.

Switzerland: A “Pure” Elite System? Like Germany and Austria, Switzerland retains an elite higher-education system with limited enrollment levels. Even as German enrollment began to expand gradually in the 1970s, Swiss enrollment remained flat until the 1990s. By contrast, however, the Swiss higher-education system is not underfunded—its level of per-student spending is one of the highest internationally. Why did this pattern of extreme elitism, unique in Europe, occur? And how can we explain it using the theoretical logic laid out in Section Two? Switzerland appears to have experienced a “perfect storm” of rightleaning governance, at both the federal and cantonal level, and a large number of veto players—both the cantons and the existence of popular referenda. Right-leaning government meant little concerted political interest in expansion to a mass system and a focus on increasing funding for this restricted enrollment. The cantonal veto players and the existence of referenda have made it almost impossible to establish new universities, which might enable expansion, or to increase the federal government’s role in university planning. These two factors combined with another factor common to Germany and Austria—the existence of a vibrant “dual system” that supports vocational education and channels students away from higher education. Excluding the ancient University of Basel, the Swiss higher-education system dates back to the mid– to late nineteenth century, during which seven of the ten cantonal universities and the two federal institutes of technology were founded (Heidenheimer 1997). Like their German and Austrian counterparts, Swiss universities had a strong reputation during the

Humboldt Humbled? The Germanic University System in Comparative Perspective | 231

nineteenth century. However, in the last hundred years, only two new universities have been founded, both since 1990 (Strauf, Scherer, and Bieger 2007). The Swiss higher-education system, at least at the university level, simply did not expand beyond its 1900 baseline of ten higher-education institutes. This institutional stasis explains the very slow growth in enrollment over the last century in Switzerland. At the end of the nineteenth century, Switzerland’s enrollment rate was, at 1 percent, slightly higher than the European norm, and it continued to grow at the average regional rate until the 1950s, when, like most other industrial countries, around 3.5 percent of the age cohort attended university (Heidenheimer 1997). After this point, however, unlike most European states, even Germany, Switzerland failed to expand to incorporate the growing baby boom generation. Between 1960 and 1965, around 4 percent of Swiss citizens graduated university, compared to a European average of 7 percent. By 1975, when 13 percent of Europeans were attending university, less than 8 percent of Swiss students did so. This pattern has continued to today, with just over a third of students participating in higher education at some point (the same as Austria and Germany), much less than the OECD mean of 53 percent. Thus, while other states experienced a takeoff in enrollment at some point post-1960, the Swiss experience has been halting and gradual. Nevertheless, spending on higher education has been exceptionally high by European standards. In 2002, the average annual public expenditure per student in higher education in Switzerland was US $23,714 compared to an OECD mean of US $10,655 (OECD 2005). Germany and Austria had levels just over the mean—US $12,448 and US $10,999 respectively. What explains this massive difference? It is not merely a function of the high level of personal income in Switzerland, since per-student spending amounts to 73 percent of GDP per capita in Switzerland, compared to an average of 43 percent. Switzerland consequently spends 1.4 percent of its national income on higher education, almost all of it publicly financed. This places it just below the Nordic countries and considerably above the Continental norm. Such spending has come not from increased enrollment, but from increased per-student spending: of the 49 percent growth in overall spending between 1995 and 2002 in Switzerland, 80 percent came from increased per-student funding rather than enrollment (OECD 2005). Consequently, Switzerland appears to have “jumped off ” the trilemma—in that it has increased public spending without becoming a mass system. Instead, it has doubled down on a “pure” elite system. What is the historical story behind this trend, and how might we understand it analytically? The key historical developments underlying the peculiar structure of Swiss higher education lie in the mid– to late nineteenth century, during

232 | Ben W. Ansell

which time the modern Swiss political economy developed. The constitution of 1848, while creating a newly empowered confederation, also left the cantons with significant political power and continued control over education (Hega 1999). It would take a series of extensions of federal power for the central government even to take any lead at all in education, chiefly in the creation of two federal institutes of technology. Nonetheless, since the eight existing cantonal universities provided the bulk of higher education, power in higher education rested at a very localized level. Furthermore, financing students at universities outside their home canton rests on a complicated intercantonal scheme of pooled funding, outside of the federal government’s control (Werner and Shah 2006). The second historical legacy from the nineteenth century was the development of the “dual system” of vocational apprenticeship, during the period of Swiss industrialization. This provided a viable track to stable and decent wages for the majority of Swiss students (Hanhart and Bossio 1998). Indeed, even today, the relative returns of university versus vocational education in Switzerland are much lower than even in Germany (OECD 2003). An influx of foreign workers in the 1920s helped solidify this system, as Swiss students were able to access white-collar jobs denied to the large foreign population, even without higher education. Consequently, Swiss demand for higher education has remained low because of the dual system, just as cantonal politics have restricted supply. The importance of cantonal politics to restrictions on expansion is apparent in an examination of the two referenda on higher-education reform put to the Swiss electorate in the 1970s. In 1973 the first referendum was on the federal government’s role in higher education and, particularly, in the founding of new universities to push expansion. Not coincidentally, this push occurred as the Social Democratic Party, as in many countries the most favorable to expansion, controlled the Swiss Federal Council (the Swiss federal government) and several cantonal education positions. A constitutional revision was necessary to expand the federal government’s role in higher education, and this required the winning of a referendum and the support of the majority of cantons. The first task proved viable but not the second, thus undermining the revision. This 1973 attempt was followed a few years later by near unanimous agreement in the National Council of Switzerland (the Swiss Parliament’s lower house) to increase the federal government’s role in funding increased enrollment directly, but this too was defeated resoundingly in the referendum of 1978 (Heidenheimer 1997). Tellingly, it was defeated by a coalition of right-wing voters, skilled workers, and farmers. The structure of the Swiss labor market, which benefited the vocationally trained, combined with conservative antipathy, condemned the reform.

Humboldt Humbled? The Germanic University System in Comparative Perspective | 233

The broadly defined conservatism of the Swiss polity has continued to keep expansion minimal. Conservative politics have meant that student financial support is low, even while per-student spending on education is high. Few working class children attend Swiss university, and the paltry student-support funding, some of the least generous in Europe, maintains this inequity in access (OECD 2003). The highly streamed structure of the upper secondary schooling system reinforces this pattern—only those who attend Gymnasien—a mere 20 percent of Swiss students—can receive the Maturat qualification for university. Cantons retain well over 50 percent of the responsibility for funding higher education, granting cantons the political power over the system that comes with financing and a veto over expansion. The localized level of decision-making also gives the cantonal universities more power in dictating the terms of their own governance and competition. This brake on expansion has enabled the cantons to keep per- student spending high, since numbers remain low and research expenditures by the federal government and private firms are also extremely high by international standards. This is not an unhappy outcome for the students and professors of these institutions, but there have been increased concerns, partly from Swiss businesses and partly from international observers, that Switzerland is facing a higher skills shortage. Consequently, a new round of reform has been engaged since 1995. A “binary” system of higher education was implemented that year with the full institution of advanced vocational schools (universities of applied sciences) that would produce vocational degrees that conform to international standards of higher education (three years at institutions recognized as tertiary level, as stipulated in EU Directive 89/48). The federal government has been able to take direct control over the financing and management of these schools, much as the Austrian government was able to with Fachhochschulen. This was followed by concerted federal attempts to rethink power-sharing with the cantons, the creation of two new cantonal universities, and single academic diplomas. These two new universities, in Lucerne and Italian Switzerland, both charge limited tuition fees, amounting to just over 10 percent of their income (all other universities charge less than 5 percent), possibly suggesting an institutional change in financing (Werner and Shah 2006). Swiss higher education still remains limited in enrollment, though decently funded. It is not clear at present whether the greater federal display of muscle will manifest itself as a concerted push for a mass higher-education system or whether tuition fees of any sizeable amount will emerge on the political agenda.

234 | Ben W. Ansell

Conclusion Superficially, it appears that since the introduction of tuition fees in Austria and some German Länder the Germanic higher-education system is transitioning from an elite to a mass partially private system along AngloAmerican lines. Yet these fees have hardly altered the essentially regressive nature of higher-education funding in both countries and, if anything, have worsened the problem of upper-middle-class bias in access. Furthermore, there has been no move toward tuition fees at all in Switzerland. Despite the appearance of reform, the Germanic higher-education system remains the most limited in enrollment in Europe and is basically dependent on the public purse. It is possible that a future “grand coalition” in Germany may push for expansion of higher education to the middle classes, since they are the chief beneficiary from cross-party government. Even then, federal intentions may founder, as they typically have, on subnational intransigence from the Länder. The Austrian situation is more promising since the central government has more leeway in higher education policy. However, even here the universities themselves have lobbied extensively and successfully against expansion. The political turn to the left in Austria does not have clear implications, because the higher-education system has not yet grown to the extent that the chief beneficiaries of further expansion would be the constituents of the Social Democrats. The right-wing parties appear most interested in bulwarking the system against the encroachment of poorer students. Thus, in the Austrian case, the obstacles to reform may be partisan rather than institutional in nature. Swiss expansion appears, conversely, to be institutionally determined. The creation of the Università della Svizzera Italiana in 1996 and the University of Lucerne in 2000 marked the first expansion of cantonal universities for a century and may herald a change in cantonal attitudes toward expansion. All three states share one further constraint. Although I only touched on vocational training at the secondary level briefly, it clearly provides an acceptable substitute to university education for many middle class children in Germany, Switzerland, and Austria. Indeed, it is the part of the education system held most responsible for the postwar miracle in German economic growth. Yet vocational training itself is not a clear substitute for university education, since the Scandinavian countries and Holland all have well-established apprenticeship systems, as well as large university systems. If competition from lower-wage economies threatens Germanic competitiveness in high-end manufacturing, the high levels of vocational training currently provided in Germany and Austria may lead to a glut of intermediate skills and a deficit of university-level skills. Without an

Humboldt Humbled? The Germanic University System in Comparative Perspective | 235

existing mass higher-education system ready to absorb potentially unemployable vocational apprentices, the already unsatisfactory levels of unemployment in these states could worsen sharply. But higher-education expansion also implies a sharply widening college wage premium and the threat of Anglo-American inequality. Mass higher education, then, may be a mixed blessing; though it may be a functional response to globalization, it will not dispel bitter fights over education funding.

Notes 1 2

3 4 5

.

This measure of enrollment provides us with the stock of higher education, whereas the net entry rate into higher education, which I use elsewhere, measures the flow. . A further assumption underpins the trilemma—that the “quality” of the public good remains constant; that is, per-person spending remains constant. In fact, the Swiss case belies this, as we shall see. . This data is taken from OECD (2005). . The section on Germany builds on Ansell (2008). . The association of university presidents, the Westdeutsche Rektorkonferenz, was against building even a single new university (see Nugent 2004).

Bibliography Ansell, Ben W. 2008. “University Challenges: Explaining Institutional Change in Higher Education.” World Politics 60, no. 2 (January): 189–230. Ash, Mitchell, ed. 1997. German Universities Past and Future: Crisis or Renewal? Policy and Institutions. Providence, RI: Berghahn Books. Baker, David, and Gero Lenhardt. 2008. “The Institutional Crisis of the German Research University.” Higher Education Policy 21, no. 1 (March): 49–64. Beerkens, Eric. 2003. “Higher Education in Austria: Country Report.” Working Paper, Center for Higher Education Policy Studies. Culpepper, Pepper, and David Finegold, eds. 1999. The German Skills Machine: Sustaining Comparative Advantage in a Global Economy, Policy and Institutions. New York: Berghahn Books. Fischer-Appelt, Peter. 1996. “The University: Past, Present, and Future.” In Universities in the Twenty-First Century, edited by Steven Muller and Heidi L. Whitesell, 3–14. New York: Berghahn Books. Florida, Richard L. 2004. The Rise of the Creative Class. New York: Basic Books. Führ, Christoph, and Iván Tapia. 1997. The German Education System Since 1945. Bonn: Inter Nationes. Glotz, Peter. 1996. Im Kern verrottet?Fünf vor zwölf an Deutschlands Universitäten. Stuttgart: Deutsche Verlags-Anstalt.

236 | Ben W. Ansell

Hanhart, Siegfried, and Sandra Bossio. 1998. “Costs and Benefits of Dual Apprenticeship: Lessons from the Swiss System.” International Labour Review 137, no. 4: 483–500. Hega, Gunther M. 1999. Consensus Democracy? Swiss Education Policy Between Federalism and Subsidiarity. New York: P. Lang. Heidenheimer, Arnold J. 1997. Disparate Ladders: Why School and University Policies Differ in Germany, Japan, and Switzerland. New Brunswick: Transaction Publishers. Hüfner, Martin. 2003. “Governance and Funding of Higher Education in Germany.” Higher Education in Europe 28, no. 2: 145–63. Konow, Gerhard. 1996. “Planning, Financing and Accountability of German Universities.” In Universities in the Twenty-first Century, edited by Steven Muller and Heidi L. Whitesell, 49–57. New York: Berghahn Books. Korpi, Walter. 1983. The Democratic Class Struggle. London: Routledge and K. Paul. Lingens, Hans G. 1998. German Higher Education: Issues and Challenges. Phi Delta Kappa International Studies in Education. Bloomington: Phi Delta Kappa Educational Foundation. Marcucci, Pamela, and D. Bruce Johnstone. 2007. “Tuition Policies in Comparative Perspective: Theoretical and Political Rationales.” Working Paper, SUNY Buffalo. Nugent, Michael A. 2004. The Transformation of the Student Career: University Study in Germany, the Netherlands, and Sweden, Routledge-Falmer Studies in Higher Education. New York: Routledge-Falmer. OECD. 2003. Tertiary Education in Switzerland. Paris: OECD. ―――. 2005. Education at a Glance 2005. Paris: OECD. ―――. 2006. Education at a Glance 2006. Paris: OECD. Onestini, Cesare. 2002. Federalism and Länder Autonomy: The Higher Education Policy Network in the Federal Republic of Germany, 1948–1998. New York: Routledge-Falmer. Ostermann, Hanna. 2002. “‘Rotten at the core?’ The Higher Education Debate in Germany.” German Politics 11, no. 1 (April): 43–60. Pechar, Hans. 2005. “Backlash or Modernization? Two Reform Cycles in Austrian Higher Education.” In Reform and Change in Higher Education, edited by Åse Gornitzka, Maurice Kogan, and Alberto Amaral, 269–86. Dordrecht: Springer. Strauf, Simone, Roland Scherer, and Thomas Bieger. 2007. “The Role of Universities for Regional Labour Markets: The Example of Central Switzerland.” Working Paper, University of St Gallen. Werner, Jan, and Anwar Shah. 2006. “Financing of Education: Some Experiences from Ten European Countries.” Working Paper, Institute of Local Public Finance, Langen. Windolf, Paul. 1997. Expansion and Structural Change: Higher Education in Germany, the United States and Japan, 1870–1990. Boulder, CO: Westview.

Chapter 10

BEYOND THE WELFARE STATE Consumer Protection and Risk Perceptions in the European Union and Austria

( Paulette Kurzer

In the last twenty years, numerous publications have addressed the issue of risk perception and risk assessment. This literature on risk explores the apparent paradox of citizens who reside in advanced industrialized countries—and therefore enjoy unparalleled prosperity—yet express great anxiety about remote and distant hazards that are believed to threaten society and individual well-being. One of the questions raised in this literature is how citizens assess risk factors with regard to economic forces, technology, or the global system and why some issues become highly politicized while others remain ignored. This chapter looks critically at the case of Austria, which squarely belongs to the elite club of wealthy countries. Despite its membership in an “elite” group of countries, phases of intense anxiety have marked Austrian society. Global warming; terrorism; the pollution of waterways, soil, and air; contaminated foodstuffs; infectious diseases; high-voltage electricity; genetically modified organisms; plus hosts of other trends provoke much hand-wringing and widespread media discussions. Threats to the good life appear to lurk everywhere as the widening web of the mobility of people, products, and services generates new challenges, many of which do not fall directly within the regulatory scope of state agencies and cannot be governed by legislation. In addition, the very achievements of technological advances—whether they be biomedical, informational, material, or financial—generate unintended outcomes that seem to harbor new and unknown harms, which are also hard to anticipate and impossible to contain. In Austria, the first event that drew attention to the emergence of new threats to the quality of life took place in the 1970s. Deep and widespread

238 | Paulette Kurzer

concern mobilized voters to oppose the building of nuclear power plants, and a subsequent popular referendum on nuclear power resulted in the Nuclear Power Prohibition Act, in 1978 (Getzner 2003). Afterward, Austria pushed for the adoption of the precautionary principle as a guide to regulatory decision making. This approach obliges governing authorities to step in when potentially dangerous effects deriving from a phenomenon, product, or process have been identified. Elected officials are supposed to take steps even if scientific evaluation does not specify any particular risk factors with sufficient certainty because of insufficient data or inconclusive evidence. The adoption of this principle puts the burden on policy officials to take action even when it is unclear whether action is actually useful or necessary. The European Commission, the executive center of the European Union, endorses this principle in recognition of citizens’ ambivalence about the benefits of unproven technological innovations (Ahteensuu 2007; Graham and Hsia 2002). Nevertheless, the link between perceived or existing risk and public intervention is never direct, since, once scientists agree, the political process ultimately determines whether this consensus warrants new public-policy measures and regulatory action (Nathanson 2007). In the end, experts’ credibility is mediated by a host of variables related to the cultural legacies of a country, the power balance among political actors with divergent agendas, mobilization of interest groups, coverage by popular media, and, possibly, the timing of the controversy. Perhaps this helps account for why European countries rally against different types of risk and propose different solutions accordingly. Since perceptions of harms vary across Europe, individual member states adopt divergent health and environmental agendas not always in alignment with the rest of the EU. Austria is a “special” case and demonstrates how small member states can nonetheless fend off external pressures to conform to pan-European rules. Its approach to environmental and health policies diverges from mainstream Europe, while its actions (or inaction) at times influence the outcome of European legislation. This chapter focuses on tobacco control and bioengineered seeds/crops, which do not exactly fit the traditional definition of social policy. Nevertheless, tobacco control and biotech crops/food elucidate an aspect of small-state political dynamics that is easily overlooked in the standard literature on European integration and its impact on social conventions and cultural norms. Table 10.1 compares the position of Austria with that of the EU and the United States. As shown, whereas Austria is unconcerned about smoking, it is intensely worried about bioengineered crops and foods (made from GMO—genetically modified organisms).

Consumer Protection and Risk Perceptions in the European Union and Austria | 239

Table 10.1. Health issues: official level of concern Expressed by

Smoking

GMO

Austria—Government, media, experts, society

Minimally

Extremely

EU—Commission, Parliament, Euro groups

Extremely

Moderately

US—Federal government, society, media, experts

Extremely

Minimally

In Table 10.1, the EU category refers to the position of the European Commission and European Parliament and may not fully represent the view of each EU member state, which is why I prefer to stay with the official legislation adopted by the European Commission. The US category refers to federal agencies and expert discussion within the scientific community, which seems to shape public debate on these issues. The Austrian position is a distillation of the formal stance of different federal government coalitions, civic movements, nongovernmental organizations, and the scientific community. Using Table 10.1 as a starting point for our discussion, it is interesting to note that the EU and US disagree on the risks posed by GMO, while Austria, in turn, diverges from both the US and EU with regard to the depth or scope of both tobacco and genetically modified organisms. The purpose of this chapter is to explain why Austrians ignore the scientifically proven risk factor of smoking yet exhibit strong opposition to GMO food and crops, whose risks are unproven . The answer to this puzzle sheds light on the persistence of different political dynamics across the EU and the ability of small states to defend the privileged position of influential economic interests. In Austria, a politically astute cigarette industry persuaded politicians to hold off on anti-smoking measures, and an influential organic farming sector convinced these same politicians to impose a strict ban on biotech foods. In addition, owing to embedded national norms and values, the rhetorical statements of these organized interests resonated with voters and opinion leaders and strengthened their political advantages, which explains why they prevailed in spite of the fact that Austria’s unique position on smoking and agricultural biotechnology brought on conflicts with the European Commission. The chapter is divided into three sections. It first traces the formation of a pan-European tobacco-control program and a Community–wide regulatory framework to vet requests to release GMO into the food chain and environment. The next section discusses why Austria opposed measures seeking to reduce smoking rates while it pushed for a comprehensive ban on bioengineered crops. In the third section, I review present-day developments, especially in light of the continuing efforts to delegitimize smoking and the ongoing pressures to relax the authorization procedures to expedite the release of genetically modified (GM) crops into the European market.

240 | Paulette Kurzer

The Expansion of EU Competences in Health and Environment In Europe, in the late 1980s, popular awareness of the hazards of smoking was slight, and smoking rates were relatively high in many countries. Several European governments had introduced a few laws to lower the smoking rate or reduce the appeal of smoking by setting restrictions on advertisements, keeping excise taxes high, imposing age restrictions, and financing public-health announcements. However, enforcement was often weak because both European smokers and nonsmokers had not yet jumped on the anti-smoking bandwagon. This is shown in the last column of Table 10.2, which compares the smoking rates in the US and selected European countries. As can be seen, smoking prevalence rates were still high in many European countries in the 1980s in contrast to the US, where aggressive campaigns by nongovernmental agencies and state authorities had already brought down tobacco consumption (Brandt 2007; Bayer and Colgrove 2004). Table 10.2. Smoking rates, men/women in 2005 and 1985–87 (percent of population in selected countries). Ranked from high to low as of 2005. Country

Men 2005

Women 2005

Men* 1985–87

Women* 1985–87 n/a

Estonia

42

20

n/a

Austria

41

40

40

21

Bulgaria

41

23

n/a

n/a

Romania

39

18

n/a

n/a

Hungary

39

30

n/a

n/a

Poland

38

23

n/a

n/a

Slovakia

36

15

n/a

n/a

Spain

32

27

53

23

Germany

32

22

33

18

Netherlands

31

27

43

33

France

30

23

44

20

Czech Republic

30

20

n/a

n/a

Italy

31

15

46

18

UK

29

28

36

32

Denmark

28

24

51

42

Finland

25

18

32

17

Sweden

14

18

33

28

US

25

20

33

27

* The figures are not fully comparable because different age cutoffs are used, i.e., fifteen or sixteen years of age. Also, countries employ different modes of collecting the data. (Source: European Commission 1996; World Health Organization 2007)

Consumer Protection and Risk Perceptions in the European Union and Austria | 241

In that sense, it is not surprising that the European Commission embraced the idea of slashing smoking rates throughout the European Union in order to reduce cancer mortality since it could play a positive role by stepping into a national policy void. In addition, the European Commission could argue that some of the health and consumer-protection legislation related to smoking had a direct bearing on the European single market. In the years following the 1986 approval by the EU council of ministers to finance the war against cancer, the Commission drafted technical tobacco directives related to health warnings, tar and nicotine, excise taxes, and tar yields, seven of which were adopted between 1989 and 1992. Of all the measures, nothing provoked as much controversy as the Tobacco Advertising Directive (TAD) (98/43/EC), which called for a ban on both direct and indirect advertising of tobacco products and was first presented to the council of ministers in mid-1991. The directive was introduced on the basis of Article 100a of the EC Treaty, which aimed to remove barriers to the completion of the internal single market. After many negotiations, the council of ministers finally voted on the directive in December 1997, but the European Court of Justice subsequently overturned it. A modified version was finally adopted in 2003 (Duina and Kurzer 2004; Gilmore and McKee 2004). Since then, most EU member states have come around to the idea that smoking constitutes a health hazard. The legislative trajectory of genetically modified crops and food ingredients followed a different logic and contrasts sharply with the case of tobacco. Commercial development of GM or biotech crops took off in the mid-1990s, creating various dilemmas for the European Commission since it carried responsibility for the single market, and it had to decide on the import, cultivation, or sale of GM products. Officials at the European Commission had to draft new rules owing to pressures from below, which percolated up through representatives of national governments who were being pushed in opposite directions by a heterogeneous collection of environmentalists, biotech scientists, and agro-pharmaceutical firms. Both producer and consumer groups demanded a quick resolution to the dilemma of how to approve GM seeds and GM food ingredients, although they insisted on totally different approaches, which complicated the task of designing a workable regulatory framework. The first step the Commission took was to halt all authorizations of GMOs until new rules had been established. Unofficially, the EU operated with a moratorium on all authorizations for the approval of GM imports and cultivation between 1998 and 2004. This initial de facto moratorium was not lifted until the Commission passed a new directive (legislation) to keep track of biotech crops and food ingredients in the EU market. In 2004, after several years of debate, the EU adopted a revised Deliberate Release Directive 2001/18/ EC, as well as two additional regulations with regard to traceability and

242 | Paulette Kurzer

labeling of genetically modified organisms in both food and animal feed (Regulation 1830/2003 EC and Regulation 1829/2003 EC). After the European Parliament passed the new measures, the new authorization framework went into effect in April 2004 (Bernauer 2003; Falkner 2007; Kurzer and Cooper 2007a). Importers and users of biotech foods and crops must now request approval for the release of their products in the EU. The request for approval first goes to the European Food Safety Authority (EFSA), which provides a scientific evaluation of whether the product threatens human health or the environment (Buonanno 2006). Since the precautionary principle is not considered a scientific rule, the EFSA refuses to apply it to its evaluation of GMO (Friends of the Earth 2004). Therefore, it has approved every request for GMO authorization on the grounds that no existing evidence indicates serious risks (Randall 2006; Skogstad 2005; Tiberghien 2009). Since 2004, consideration of approval requests has resumed, and a few have been approved (Pollack and Shaffer 2009; Seifert 2006). Nevertheless, because the issue remains so controversial, no GMO approval has emerged from a vote in the council of ministers or the relevant standing committee with a qualified majority either in favor or against it. Instead, in each case, the Commission, exercising its prerogative to decide in the absence of a qualified majority, has decided in favor of the GMO in question. The EU has approved about twenty GM substances for import (Bounds 2008). Yet hardly any products have come onto the market, as farmers, retailers, and food processors are afraid of a consumer boycott of food products carrying GM ingredients. Thus, to summarize briefly, tobacco control has become part and parcel of the EU policy arsenal, constituting an important component of the panEuropean efforts to address the determinants of good health, but the GMO issue remains unresolved even though an increasing number of separate voices believe that Europe should move forward and invest in agricultural biotechnology.

Austria—Marching to Its Own Beat By 2003, the overwhelming majority of national leaders in the EU had concluded that smoking was unhealthy. In 1989, only three countries had tobacco-advertising bans: Finland, Italy, and Portugal. By 2003, twothirds of the EU (ten out of fifteen member states) had adopted legislation to ban tobacco advertising: Italy (1963), Finland (1977), Portugal (1982), France (1991), Sweden (1994), Belgium (1997), Netherlands (2002), Denmark (2002), the UK (2002), and Ireland (2003). By 2004, more than 70 per-

Consumer Protection and Risk Perceptions in the European Union and Austria | 243

cent (eighteen out of twenty-five) of EU countries had adopted legislation to ban tobacco advertising, the additional eight being Lithuania (1995), Slovenia (1996), Estonia (1997), Poland (1999), Hungary (2001), Slovakia (2001), Cyprus (2002), and the Czech Republic (2003) (Joossens, Raw, and Godfrey 2004). Thus health authorities in many countries had decided to implement some form of tobacco control and, therefore, welcomed panEuropean measures. The shift in attitudes in the EU mirrors global attention to containing the hazards of smoking in the OECD area (Studlar 2004). Nonetheless, it is interesting to note that a few countries are missing from the above list. Austria and Germany are remarkable outliers for two reasons: Most of the existing domestic tobacco-control legislation in these countries came about thanks to EU regulations and was imposed from above by EU rules. This accounts for why Austria and Germany lag in anti-smoking measures, since they only began to implement them once the EU had passed new directives. Second, officials from both countries objected to EU anti-smoking measures and sought to thwart the rise of pan-European rules for the longest period of time. Austria and Germany voted against the directive in 1997, though the majority overruled their votes. Germany, with the support of Austria, brought the directive to the European Court of Justice (ECJ), where it was overturned. The council of ministers passed a new version in 2003, which Germany also challenged and brought to the ECJ. This time, the court sided with the Commission, and in December 2006 it dismissed Germany’s challenge against the Tobacco Advertising Directive. The figures in Tables 10.2 and 10.3 speak for themselves because they show that Austria has among the highest smoking rate in the European Union and the lowest commitment to enacting anti-smoking measures. While the smoking rate of Austrian men has remained unchanged and that of women has increased, in the rest of Europe, the smoking rate of men has dropped while the rate of women has reached a plateau. In Austria, somewhat strangely, smoking rates actually increased between 1985 and 2005. Obviously, high rates of smoking stem from a relative lack of interest on the part of the official establishment and civic groups in promoting strong anti-smoking messages and in adopting measures to discourage smoking. Table 10.3 confirms that Austria scored the lowest of all EU countries in 2007 in terms of different types of government action to reduce smoking. In previous years, it was among the greatest laggards, but in 2007 it ended up at the very bottom of the chart. The score on tobacco control measures combines six policies universally recognized as effective tools for reducing smoking rates and preventing underage smoking: higher cigarette prices; bans/restrictions on smoking in public and workplaces; better consumer

244 | Paulette Kurzer

information; comprehensive bans on the advertising and promotion of all tobacco products; health warning labels on cigarette boxes; and treatment to help dependent smokers stop (World Bank 2003). Austria has implemented only a few of these measures, and those measures that they have put into place were mostly implemented because of the ratification of EU regulations.

Table 10.3. Ranking of selected European countries: tobacco-control measures (2007)

Price

Smoking ban in public places

Spending on public info campaigns

Advertizing ban

Health warnings

Treatment

Total

30

21

15

11

6

10

93

Rank

Country

1

UK

2

Ireland

23

21

3

12

6

9

74

6

Sweden

19

15

1

13

6

7

61

7

France

21

12

3

11

6

6

59

8=

Finland

17

12

2

13

7

7

58

8=

Belgium

16

13

3

12

9

5

58

10

Italy

17

17

1

10

6

6

57

12

Spain

12

15

5

12

6

5

55

13

Bulgaria

22

8

0

12

6

6

54

14=

Netherlands

14

9

4

12

6

5

50

14=

Poland

14

12

0

12

6

6

50

17

Slovakia

17

8

0

11

6

6

48

18

Switzerland

14

6

10

4

6

7

47

20

Denmark

16

3

3

10

6

7

45

22

Hungary

14

6

n/a

10

6

7

43

24

Latvia

9

12

4

9

6

1

41

25=

Czech Republic

13

6

0

10

6

5

40

25=

Slovenia

12

6

0

12

6

4

40

27

Germany

19

2

0

5

6

5

37

28

Greece

15

7

0

4

6

4

36

30

Austria

13

4

0

9

6

3

35

Source: Joosens and Raw 2007

Consumer Protection and Risk Perceptions in the European Union and Austria | 245

When we look at these figures, we must ask why a society would minimize the considerable health-related harms caused by smoking. By now, thousands of studies in numerous languages published in a variety of venues all point out that smoking shortens life expectancy and compromises quality of life. According to Austrian health ministry estimates, about 14 thousand deaths every year are related to smoking in a country of 8.3 million people—or as many as one in six deaths, compared to one in seven in Germany or one in nine in France. With such statistics well known, it cannot be that the Austrian health authorities, its biomedical community, and Austrian society at large are unaware of the enormous societal burdens smoking causes. One probable reason for the failure of Austrian officials and societal groups to promote the anti-smoking message aggressively is Austria’s special tie to Germany. Table 10.3 points out that Germany and Austria follow strikingly similar paths in terms of anti-smoking policy, and we must assume that this is more than a coincidence. Moreover, it is well known that both countries share many cultural values, reflecting common language and strong economic and financial links, although Austria is obviously the junior partner in this pro-tobacco coalition. Germany’s attitudes about smoking have been shaped by the experience of Nazism, a regime that took an intense interest in public health. As a reaction to the abuse of public-health policies for overtly political and racist objectives, German academia has shunned the study of public health aside from vaccinations and school health. Research on lifestyle “diseases,” such as addiction, are underemphasized and understudied, although smoking is even less researched than alcohol or drug abuse. German public health as a medical subspecialty is also underdeveloped. Although German academia has paid more attention to public health in the last ten years, German specialists publish their research in German-language journals, which only continues to isolate the German-speaking research community from the mainstream global circles of tobacco control (Grüning, Strünck, and Gilmore 2008; John 2001; Proctor 1999). Research debates in Austrian academia mirror those of the larger German establishment. The largest Austrian journal dedicated to the study of addiction—the Wiener Zeitschrift fur Suchtforschung—is tied to the German-speaking biomedical community and is not listed in the archive maintained by Medline, which is part of the English-language biomedical research field. Since Austrian researchers address a German-speaking audience, they continue to be influenced by German-language publications and remain divorced from the broader English-language publishing world (Thyrian and John 2006). The result is that societal activists engaged

246 | Paulette Kurzer

in anti-smoking lobbying cannot tap local health experts to enhance their credibility and neutralize the dubious claims of the pro-tobacco coalition. Moreover, at the other end of the spectrum, the Austrian pro-tobacco bloc is well organized and possesses some hidden weapons that it revives at regular intervals to deploy against tobacco-control voices. Like their German counterparts, the Austrian smoking lobby reiterates the Nazi abuse of public-health policy to tarnish the hesitant efforts of a few brave Austrian souls to curb smoking and reduce smoking-related mortality rates. By linking Hitler’s policies to the contemporary debate on lifestyle choices and their health consequences, the tobacco industry and its allies manage to portray anti-smoking measures as “fascism,” thereby silencing those who may have been inclined to press for more aggressive action against smoking (Grüning, Strünck, and Gilmore 2008; Proctor 1999). Austrian pro-tobacco voices have played the “Nazi card” to the fullest and label anti-smoking activists “nico-Nazis” in spite of the fact that neither Hitler nor the Nazi regime ever pursued anti-tobacco policies in Austria (Bachinger and McKee 2007; Schneider and Glantz 2008). In addition, the Austrian pro-tobacco groups have another powerful weapon at their disposal: The Austrian and German cigarette industries are intertwined and have closely coordinated their strategies to neutralize the move toward marketing restrictions. In 1784 Austrian imperial authorities established a tobacco monopoly, and Austria Tabak produced tobacco and snuff in Vienna. The Nazi regime put the monopoly under the Ministry of Finance in 1939, where it stayed until its privatization in 1997, when it was sold to a British multinational conglomerate. In 2002, Austria Tabak became a wholly owned subsidiary of Gallaher, maker of Benson & Hedges and Silk Cut cigarettes (Bachinger and McKee 2007). However, when Austria Tabak was a state monopoly under the direction of the finance ministry, it operated like a private company and exerted considerable influence over the ministry, since finance officials were keen to collect excise taxes from cigarette sales. Furthermore, during World War II it was decided to incorporate the Austrian tobacco company into the German association of cigarette manufacturers. After 1945, Austria Tabak remained a member of the German Association of Cigarette Industries, and it participated in various schemes to thwart the emergence of a European anti-smoking campaign. The association repeatedly went on the offensive to discredit research that drew attention to the harm passive smoking causes, and it challenged studies on the impact of tar levels on the incidence of lung cancer (Hirschhorn 2000; Grüning, Gilmore, and McKee 2006). Thus, the Austrian monopoly behaved like a private company and joined the German association of private cigarette companies, which gave it additional ammunition to fight

Consumer Protection and Risk Perceptions in the European Union and Austria | 247

against the rising wave of tobacco-control attitudes and rules, mostly coming from Brussels. Owing to their institutional and financial resources, Austrian opponents of tobacco control managed to subvert the measures to curb smoking. However, what further strengthened the position of the pro-tobacco lobby was the tight financial and commercial links between the German and Austrian media. German investors own or control a large proportion of the Austrian media market and profit from tobacco-advertising revenues. German-owned publishers and media companies dominate the Austrian print media in spite of efforts by Austrian politicians to protect the Austrian identity of the media sector (Signitzer 1978). Currently, different German media groups control some of the largest daily newspapers and popular Austrian weekly news magazines. While German media companies (together with cigarette companies) lobbied German officials to hold off on a ban on tobacco advertising, German politicians approached their Austrian counterparts to desist from legislating restrictions on tobacco advertisements. Their reasoning was, of course, that if Germany remained open to tobacco advertising, Austria ought to do likewise and not disrupt the business affairs of German media conglomerates (Joossens 2004; Simpson 2005; Wiener Zeitung 2004). Thus, a combination of cultural attitudes and economic calculations conspired to weaken greatly the Austrian official commitment to reduce smoking and address the risks tobacco poses for smokers and bystanders. Cultural attitudes lined up with the rhetoric of the pro-tobacco forces, because many Austrians are unfavorably disposed toward the idea of state intervention in lifestyle choices and restrictions on smoking. While pro-tobacco forces exploited the cultural inhibitions against active and interventionist public-health steps, policy toward bioengineered seeds and food ingredients has developed much differently. The official Austrian position remains fully committed to strict rules against GMO on the grounds that its risks to the environment, health, and food chain are unknown and, therefore, dangerous. The Austrian assessment of agricultural biotechnology was restrictive from the beginning, and many observers dismissed its position toward bioengineering as extreme and unrealistic (Torgesen 2002). However, its position has remained virtually unchanged and can no longer be dismissed as quirky and impractical. Table 10.4 summarizes Austrian public attitudes toward biotechnology and biotech foods. As shown, Austrian public opinion expresses a high degree of skepticism toward biotech foods, and Austrian politicians vote in EU ministerial committees in accordance with the “will of the people.”

248 | Paulette Kurzer

Table 10.4. Attitudes toward agricultural biotechnology, (percent) Austrian respondents Year

Question or statement

Agree

1996

Biotechnology in production of food should be encouraged.

Disagree Don’t know

18 (44)*

n/a

18 (11)*

2001

I do not want this type of food.

75 (71)*

n/a

n/a

2001

I think food based on GMO is dangerous.

64 (56)*

n/a

n/a

2005

Food made from GMO is dangerous.

70 (53)*

14

5

2005

Worried about GM food or drinks

69 (58)*

n/a

n/a

* EU-15 average. Source: Eurobarometer 1996, 2000, 2005; Bonny 2003

Thus, smoking—with definite risks—remains lightly regulated, while GMO—with remote and unknown risks—is strictly controlled. How to explain this paradox? The first answer is obvious to most people, because many of us engage in this kind of illogical reasoning. People tend to be more accepting of controllable risks, such as smoking, than involuntary risks, such as the contamination of the food chain or ecological degradation. Smoking is an acceptable risk because we “decide” to take that risk and can also decide to reduce the risk by quitting smoking. Exposure to biotech crops or food is beyond our control and reach. The “dread factor” is much higher in situations beyond our control that are imposed on us unwillingly, while acceptance of potentially harmful consequences is lower (Slovic 2000). Another way of capturing the different risk calculations between smoking and agricultural biotechnology is to consider individual instead of societal risks. Governments typically regulate the production of societal risks. Decision making about societal risks is presumed to be transparent, proportional, and nondiscriminatory, as well as based upon (commonly accepted) adequate grounds. Individual risks are considered out of bounds for government intervention in liberal democracies. In the past, smoking was considered an individual risk, but no longer, since many governments recognize passive smoke as a health threat. However, this sort of explanation would only be helpful if it described more or less all European societies. But most member states tend to take a different position in that they rank the risks of smoking higher than the harm caused by agricultural biotechnology, as smoking has been defined as a societal risk, which justifies in turn mandatory restrictions and oversight (Brandt 2007). Social psychology is, therefore, not of much help, because it does not account for Austria’s attitudes and policy approach. Rather, here, too, a mixture of economic interests coupled with the activities of civic action groups and particular patterns of incentives took advantage of existing

Consumer Protection and Risk Perceptions in the European Union and Austria | 249

cultural values and societal preferences to publicize the dangers of this new technology and convince the population to resist European trends. It could be that Austrians especially abhor large-scale, expert-directed technological systems that carry menacing images, which the environmental movement fully exploited (Torgersen 2002; Wagner, Kronberger, and Seifert 2002). Under the influence of the environmental movement many citizens came to regard green biotechnology as analogous to nuclear energy, with both having incalculable risks and minimal consumer benefits. Chernobyl proved doomsayers right in 1986, which later increased the risk perception of GM food (Torgersen et al. 2001). The memory of the successful fight against nuclear power spilled over into mass protest against green biotechnology. Bioengineering was presented by its opponents as yet another innovation, together with nuclear energy, that gambles with nature and may produce unforeseen outcomes. Presently, most Austrians do not like nuclear energy and do not believe that it is a helpful tool to reduce global warming, reduce dependency on imported fuel, or lower fuel prices. Their skepticism has only been reinforced by the leaking radiation from the Fukushima nuclear power plant in Japan. Austrian voters continue to be among the least receptive to nuclear energy or lower energy prices. They sense that the risks outweigh any potential benefit (Eurobarometer 2007). Building on earlier strategies, the environmental movement convinced 21 percent of the population to sign a people’s petition (1.23 million signatures) to ban GM food in spring 1997 (Torgersen et al. 2001). The Austrian government subsequently banned GM plants and food and required that products with GMO be correspondingly labeled. Austrian retailers also pulled products containing GM matter off their shelves. Austrian governments sided with the anti-GMO movement because coalition governments had already quietly opted for another kind of agriculture, and GM technology did not mesh with the new agri-environmental direction (Kurzer and Cooper 2007a, 2007b). Agricultural policy took on a strong ecological dimension, with organic farming—which can be profitable in a small cultivated area—becoming the survival strategy of low-intensity Alpine farmers (Lesjak 2008; Dabbert, Haring, and Zanoli 2004). The Ministry of Agriculture and existing corporatist agricultural organizations ignored the first wave of organic farming in the early 1980s, but the ministry switched direction in the late 1980s under the leadership of the People’s Party and established a special department to extend direct assistance to this farm sector in the form of general subsidies, marketing strategy, certification, and issuance of eco-labels. By the early 1990s, the coalition governments led by Social Democratic or People’s Parties adopted an “eco-social market” policy to secure the livelihood of

250 | Paulette Kurzer

a shrinking farming sector. Subsequently, the organic farming association expanded its membership rapidly and grew from two hundred to twenty thousand farmers (Hofer 2000: 159). To encourage organic farming and make this sector profitable, official campaigns celebrated locally grown food starting in 1994 and required labeling to indicate the product’s origin, its quality, and independent (i.e., family-owned) control. Public messages reiterated that organic farming equaled superior food quality, which was part of Austria’s national identity and represented a better option than EU productivist, polluting agrobusiness (Hofer 2000; Sassatelli and Scott 2001). The rapid expansion of organic farming fostered Austrian pride in its food, and Austrian consumers believed that they enjoyed higher food standards than European consumers in other countries. This faith in the superiority of Austrian-grown products came into play when the debate on GMO began. Opponents of GMO pointed out that organic farming and good food quality were part of Austria’s national identity (Wagner, Kronberger, and Seifert 2002). In the early 1990s, owing to the political consensus on how to address declining farm incomes and preserve delicate landscapes, Austrian farming centered on encouraging value added food production. Moreover, while farming is a business in the United States and under the control of producers, Austrians treasure the rural countryside, which is a source of tourist revenue and individual pleasure. Accordingly, Austrians feel they have a right to co-decide what kind of agriculture should be practiced in a multifunctional landscape that is an extension of the collective identity (Torgensen 2002). In conclusion, agricultural biotechnology evokes strong currents of hostility and anxiety because of the overall skepticism toward big technology projects and the special role assigned to farming and food production in Austria. Civic action groups exploited the public dislike of large- scale technology projects with modest benefits for consumers to rally opposition to biotechnology and bioengineering. Social-movement entrepreneurs established a relationship between biotechnology and the potential risks it posed for small, organic “natural” farming and Austria’s delicate Alpine ecology. Evidently, many Austrians recognized the risks and adamantly opposed the presence of bioengineered crops, feed, or food in Austria.

Evolution of EU Policy Regimes and Their Impact on Austria Advanced industrial countries take different approaches to managing personal and collective risks and to balancing the rights of the public

Consumer Protection and Risk Perceptions in the European Union and Austria | 251

good against the rights of individual freedom. Austria provides an interesting case study for illuminating the tension between personal risk (i.e., smoking and collective risk, biotech crops) and between national priorities and European developments. For cultural and economic reasons, the anti-smoking campaign did not resonate with Austrian society and public elites, but civic forces immediately jumped on the anti-GM bandwagon. In the latter case, the arrival of biotech foods into the European market helped revive the dormant anti-nuclear movement, which resurrected the previous tactics and outreach activities that had been so successful during the period when nuclear energy was on the agenda and applied them to the new issue of bioengineered crops and foods. The movement linked agricultural biotechnology to the threat to Austria’s food-production regime and highlighted the superior attributes of Austrian farming against the destructive regime prevalent in the rest of Europe. It also played on inchoate fears about the harm done to the environment and human health. The combination of unknown risk with real threats to the survival of organic farming coalesced into a powerful message that convinced many Austrians to oppose agricultural biotechnology. But societal voices had a harder time embedding anti-smoking messages into an existing framework with widespread resonance among Austrian citizens, because it could not tie tobacco control to the preservation or defense of the “public good.” In addition, there was no economic bloc that favored anti-smoking regulations in the way that an economic bloc supported a strict ban on biotech crops. The institutional configuration of economic interests was quite distinct in each case, since powerful economic voices stringently opposed tobacco control while another set of economic actors—farming, tourism, food retailers—opposed biotechnology. In this final section, I will examine whether the arrival of EU legislation, enacted over the last fifteen years, has influenced in any way the attitudes of Austrian officials and society. Have EU rules made any impact on Austrian attitudes and policy preferences concerning the relative risk factors of smoking versus biotech foods? The answer is twofold. Austria is ever so slowly moving toward mainstream tobacco-control legislation, but it holds steadfast to its position of “absolutely not” with regard to agricultural biotechnology. Austrian society is increasingly willing to contemplate restrictions on smoking. Two developments have contributed to the change of heart. First, European legislation has forced Austrian health authorities to introduce new laws to comply with EU requirements on health warnings, advertising restrictions, packaging, sales prices, promotions, etc., resulting in the greater salience of the debate on smoking and health measures. It has given the small anti-smoking coalition greater clout since it is now firmly

252 | Paulette Kurzer

aligned with wider European trends. The Austrian Council of Health and the Chamber of Doctors (Österreichische Ärztekammer—ÖÄK) have fully come around to the idea that smoking should be prohibited in public spaces (Austrian Council on Smoking and Health 2007; Hofmarcher 2009; ÖÄK 2006, 2007). Second, Austrian consumers have been increasingly exposed to developments in neighboring countries (Germany!), and tourists from many countries (including the United States and Canada) expect smoke-free areas in restaurants and other public spaces. The refusal to restrict smoking is not winning widespread accolades, as American and European visitors are accustomed to sitting in smoke-free restaurants and cafes. Moreover, the Austrian national media has picked up the public discourse on anti-smoking strategies in other European countries (for example, Norway, Sweden, France, Ireland, Italy, and the UK) and redirected the thrust of the debate on the societal burden and financial costs of treating tobaccorelated diseases. A proliferation of new laws enacted in neighboring countries underscores that indoor pollution is not a purely American obsession and is taken seriously by other European leaders (Grüning and Gilmore 2007; Krause 2008). Not surprisingly, the Austrian Health Reform legislation of 2005 contained a list of tobacco-control measures and introduced a ban on smoking in all public places (although many exceptions to this ban exist), a general ban on advertisements for tobacco products with the exception of billboards and advertisement in cinemas, and the obligatory signposting of places where smoking is forbidden. New EU laws dictate the advertising ban, while the ban on smoking in public places reflects new thinking about the hazards of smoking (Matouschek 2007; Austrian Federal Ministry of Health 2010). However, it is important to point out that Austria is still hesitant to issue a strong anti-smoking message. On 1 January 2009, new smoke-free legislation went into effect. The “grand coalition” of Social Democrats and the conservative People’s Party agreed on a messy compromise in December 2008 that required all catering establishments of 50 square meters or more (861 square feet) to build separate smoking rooms if they choose to serve smokers, while all restaurants smaller than 50 square meters may decide whether to be smoke free or not. The legislative mess is created by the fact that this is only a partial smoking ban and two-thirds of pubs, restaurants, and cafes have said that they will continue to allow smoking. In July 2010, the Greens, the Freedom Party (FPÖ), and the Alliance for the Future of Austria (BZÖ) showed rare unity by appealing to the Social Democratic health minister, Alois Stöger, to scrap the existing ruling and hold a referendum. The call for a referendum was prompted by a similar

Consumer Protection and Risk Perceptions in the European Union and Austria | 253

referendum in neighboring Bavaria, in Germany, where the majority of citizens supported a total smoking ban in locals and eateries. The health minister rejected appeals for a referendum, claiming that most Austrians are happy with the current law, which went into effect in January 2009 (Austrian Times 2010). In a 2009 study, 63 percent of Austrians were satisfied with implementation of the new anti-smoking law and 33 percent wanted a total ban on smoking in bars and restaurants (Austrian Times 2009). With time—and a change of government or health minister—the Austrian anti-smoking coalition will most likely prevail, although the same cannot be said about the prospects for Austria’s feeble pro-biotech lobby. In spite of numerous EFSA reports, assuring that no discernible risks to health and environment have been identified, Austrian officials have voted down all requests for approval to release GMO products. This rejection in spite of positive scientific evaluation by the EFSA has contributed to a stalemate between the EU and the rest of the world. The World Trade Organization ruled in September 2006 that the EU moratorium on the authorization of GM products between 1999 and 2004 broke world trade rules. During the last few years, the European Commission has approved the import of GM corn seed, but Austria has continued to impose a ban, in particular on the import of GM corn (Seifert 2007). The European Commission has asked other member states to reprimand Austria and force it to withdraw the ban. Because of the single market, a ban in one country impedes the free movement of goods throughout the EU, which is why the Commission desperately wants to solve the stalemate. When the vote took place in October 2007, many national representatives abstained, and only a few countries upheld the right of the Commission to force Austria to lift its ban. Fifteen out of twenty-seven national ministers voted to let Austria keep its ban, while only the UK, Estonia, the Netherlands, and Sweden voted in favor of the Commission’s proposal to lift Austria’s ban. Eight countries abstained. In early 2009, the Commission again tabled a vote to force Austria to lift its ban against the cultivation of GM corn (MON 810 and T25). Again, the vote was defeated, as twenty-two representatives of the member states rejected the Commission’s efforts to force the planting of these two varieties of GM corn (only the UK, Finland, the Netherlands, and Sweden sided with the Commission). Remarkably, this time the majority supportive of Austria’s position was even larger than during the first two votes in 2005 and 2007. Strengthened by this show of support, in June 2009 Austria suggested that the Council of Environmental Ministers tweak EU Directive 18/2001 (on the deliberate release into the environment of genetically modified organisms) and Regulation 1829/2003 (on genetically modified food and feed) so that member states could decide whether they want to

254 | Paulette Kurzer

plant biotech crops in their territory or not and would not have to accept the risk assessments of the European Food Safety Authority (Checkbiotech.org 2009). Many political leaders are in favor of a more “honest” solution by allowing Austria (and other member states) to take political preferences into consideration. In spring 2010, Commission officials endorsed the idea that member states can conduct their own risk assessment on the basis of new scientific knowledge. If the member state has detailed grounds for considering that a GMO constitutes a risk to human health or the environment, it can actively ban the potentially harmful GMO until scientific assessment has proven it safe. Paradoxically, the European executive is backtracking from EFSA assessments by emphasizing the right of member states to identify different thresholds of risk tolerance for different kinds of activities. The aim is to unblock the paralysis in GM crop approvals by giving those countries that want to grow them the freedom to do so, while also sanctioning the current “GM-free” stance of several member states (EurActiv 2010). Thus, here is potentially an intriguing example of how the determination of one small member state not to yield to EU procedures and international pressures is resulting in handing EU powers back to the member states and giving up on safeguarding a uniform European farming policy. This victory seems to confirm some of the findings in the rest of this volume. Even extensive market-conforming reforms and pressures for liberalization and deregulation will not always succeed in transforming key national policies.

Conclusion Popular anxiety persuaded the European Commission to adopt the precautionary principle formally, which compels the EU to take action to prevent unknown harms or risks. Since the scientific community tends to hold reservations about the applicability of the precautionary principle, most technical studies do not include an assessment of potential risks with remote consequences. In addition, whereas Europeans endorse the precautionary principle, they disagree about which developments constitute a genuine harm. The case of Austria is illustrative of how scientific risk assessment is subject to particular political dynamics and cultural norms. While it rejects the assessment of the safety and risks of GMO by European authorities, it discards the dangers posed by smoking. The explanation for this paradox is traced to a combination of cultural values and economic incentives. Smoking is an individual choice, and the

Consumer Protection and Risk Perceptions in the European Union and Austria | 255

sale of cigarettes is strongly promoted by a politically connected industry with allies in the federal government and abroad. Cultural inhibitions complicated the formation of a vigorous anti-smoking movement, while pro-tobacco forces exploited to the fullest the presence of lingering distrust against public-health campaigns. Agricultural biotechnology produced a very different configuration of values and interests, because economic actors sought a ban on bioengineered seeds and because public attitudes rejected biotech foods as harboring potential risks with zero benefits. Here, the alignment of economic and cultural interests created a powerful and influential advocacy coalition, pressing political leaders to continue to reject agricultural biotechnology and its products. Because of sharp divisions inside the EU, the European Commission has at times struggled to find a middle ground that would still point to progress in normalizing tobacco control and the commercial availability of biotech foods and yet be sensitive to the approach and attitudes of different member states. Presently, the other member states widely accept tobacco control, and influential voices have emerged in Austria that argue that the lack of progress in tobacco control is a national disgrace. Thus, eventually, as opinion leaders increasingly lobby for more assertive anti-smoking legislation, Austria will likely fall in line with the rest of Europe. Because of the single market, the Commission is keen to break the deadlock on biotechnology. At first, EU officials figured that stringent authorization rules and elaborate decision making procedures would reassure national officials that the EU had taken all precautionary and deliberate measures to evaluate potential risk factors. However, the complex decision making structure itself has not silenced critics of genetic engineering; instead, it has become a major source of friction, Increasingly, it would appear that the cleanest solution to the stalemate is to allow member states to opt out of EU scientific-risk assessments and enable politicians in the member states to respond to negative attitudes and public opinion. This unprecedented conclusion was reached after Austria (and a few other member states) absolutely refused to budge on approving GM crops. Commission officials are now committed to delegate responsibility for GM crop policies to national governments, with the result that Austria has prevailed in the end. The fate and eventual substance of the proposed regulation to allow EU countries to restrict or ban GM crops on grounds other than their demonstrated risks to the environment or health is unknown. However, if amendments to Directive 2001/18 (on the deliberate release into the environment of GMOs) comes into force, it seems clear that current bans on GMOs in some countries will continue. This would be a victory for Austria because the Commission formally acknowledges that ethical, moral, and cultural grounds could be introduced as potential rea-

256 | Paulette Kurzer

sons for member states to implement “proportionate” and “non-discriminatory” restrictions on GM crops ( European Parliament 2011; Kurzer and Skogstad, 2012) In conclusion, the two case studies in this chapter lend support to the argument that common beliefs about certain types of risks thwart the adoption of pan-European ideas and suggestions. Stakeholders—the cigarette industry, advertising, and the media sector—were able to prevent a major shift in tobacco policy, thanks to their privileged position in a consensual political system and due to prevailing ideas about public-health intervention. The symbolic and economic importance of organic farming created a different kind of logic, in part because stakeholders could link biotechnology to the widely loathed nuclear energy sector.

Bibliography AFP (Agence France Presse). 2008. “Partial Smoking Ban in Austria: Same Difference, Say Critics.” 29 December. Ahteensuu, Marko. 2007. “The Precautionary Principle and the Risks of Modern Agri-Biotechnology.” In Genetic Democracy Philosophical Perspectives, edited byVeikko Launis and Juha Räikkä, 75–92. New York: Springer. Austrian Council on Smoking and Health. 2007. “Austria Will Meet the European Standard in the Battle Against Smoking.” http://www.aerzteinitiative.at/GusiGastein0710e.pdf (accessed 7 October 2007). Austrian Federal Ministry of Health. 2010. “Verstöße gegen den NichtraucherInnenschutz in der Gastronomie – Erlass.” (BMG - III/B/6 (Drogen und Suchtmittel) 21 June. http://www.bmg.gv.at/cms/home/attachments/9/9/8/ CH1041/CMS1157723700705/erlass_des_bmg_-_verstoesse_gegen_die_ nichtraucherschutzbestimmungen _fuer _die _gastronomie_%282%29.pdf (accessed July 5, 2011). Austrian Times. 2010. “Opposition Parties Want Smoking Referendum.” 6 July. http://austriantimes.at/news/General_News /2010-07-06/24790/ Opposition_ parties_want_smoking_referendum (accessed 1 August 2010) ―――. 2009. “Almost Two-thirds Satisfied with Anti-smoking Law.” 19 June. http://austriantimes.at/index.php?id=14128 (accessed 1 August 2010). Bachinger, Eleonore and Martin McKee. 2007. “Tobacco Policies in Austria During the Third Reich.” The International Journal of Tuberculosis and Lung Disease 11, no. 9 (September): 1033–37. Bayer, Ronald, and James Colgrove. 2004. “Children and Bystanders First: The Ethics and Politics of Tobacco Control in the United States.” In Unfiltered: Conflicts Over Tobacco Policy and Public Health, edited byEric A. Feldman and Ronald Bayer, 8–37. Cambridge, MA: Harvard University Press. Bernauer, Thomas. 2003. Genes, Trade, and Regulation: The Seeds of Conflict in Food Biotechnology. Princeton: Princeton University Press.

Consumer Protection and Risk Perceptions in the European Union and Austria | 257

Bonny, Sylvie. 2003. “Why Are Most Europeans Opposed to GMOs? Factors Explaining Rejection in France and Europe.” Electronic Journal of Biotechnology 6, no. 10. Bornhäuser, Annette, William McCarthy, and Stanton Glantz. 2006. “German Tobacco Industry’s Successful Efforts to Maintain Scientific and Political Respectability to Prevent Regulation of Secondhand Smoke.” Tobacco Control 15, no. 2 (April). Bounds, Andrew. 2008. “Fresh fight looms over Europe GM crops.” Financial Times. June 24. Brandt, Allan M. 2007. The Cigarette Century: The Rise, Fall, and Deadly Persistence of the Product that Defined America. New York: Basic Books. Buonanno, Laurie. 2006. “The Creation of the European Food Safety Authority.” In What’s the Beef? The Contested Governance in European Food Safety, edited by Christopher Ansell and David Vogel, 259–78. Cambridge, MA: MIT Press. Chaffin, Joshua. 2009. “Europe Ban on Modified Maize Upheld.” Financial Times. 3 March: 5. Checkbiotech.org. 2009. “AustriaPushes for GMOOpt - out Clause.” 15 July. http://checkbiotech.org/node/26674 (accessed 25 July 2009). Cooper, Alice, and Paulette Kurzer. 2003. “Rauch Ohne Feuer: Why Germany Lags in Tobacco Control.” German Politics and Society 21, no. 3 (Fall): 24-47. Dabbert, Stephan , Anna Maria Haring, and Raffaele Zanoli. 2004. Organic Farming: Policies and Prospects. New York: Zed. Deutsche Welle. 2006. “Much of Europe Bans Smoking; Germany Keeps Puffing Away.” 19 February. Duina, Francesco, and Paulette Kurzer. 2004. “Smoke in your eyes: the struggle over tobacco control in the European Union.” Journal of European Public Policy 11, no. 1 (February): 57-77. EurActiv.com. 2010. “EU to Overhaul GM Crop Approval System.” 4 June 2010. http://www.euractiv.com/en/cap/eu-overhaul-gm-crop-approval-systemnews-494896 (accessed June 14 Eurobarometer. 2007. Europeans and Nuclear Safety, 66.2, 271. Brussels: Eurobarometer. ―――. 2008. Europeans, Agriculture and the Common Agricultural Policy : 68.2. Brussels: Eurobarometer. ―――. 1997. The Europeans and Modern Biotechnology in 1996. 46.1 (EB 1996) Luxembourg: European Commission. ―――. 2000. The Europeans and Biotechnology. 52.1. Brussels: European Commission. European Commission. 1996. Tobacco Consumption 1970 – 1994 in the Member States of the European Union and in Norway and Iceland. Brussels. European Parliament. 2011. “EU countries should be able to ban GMOs on environmental grounds-MEPs.” 13 April. http://www.europarl.europa.eu/en/ headlines/ content/ 20110408STO17335/ html/EU-countries-should-be-able-toban-GMOs-on-environmental-grounds-MEPs (accessed July 5, 2011). Gaskell, George et. al. 2006. Europeans and Biotechnology in 2005: Patterns and Trends. (EuroBarometer 64.3). Brussels: European Commission.

258 | Paulette Kurzer

Falkner, Robert. 2007. “The Political Economy of ‘Normative Power’ Europe: EU Environmental Leadership in International Biotechnology Regulation.” Journal of European Public Policy 14, no. 4 (June): 507–26. Frankenberg, Gunter. 2004. “Between Paternalism and Voluntarism: Tobacco Consumption and Tobacco Control in Germany.” In Unfiltered: Conflicts Over Tobacco Policy and Public Health, edited by Eric A. Feldman and Ronald Bayer, 161–89. Cambridge, MA: Harvard University Press. Friends of the Earth. 2004. Throwing Caution in the Wind: A Review of the European Food Safety Authority and its Work on Genetically Modified Foods and Crops. FOE: November. http://www.foeeurope.org/GMOs/publications/EFSAreport.pdf (accessed July 5, 2011). Getzner, Michael. 2003. Nuclear Policies in Central Europe New York: Peter Lang. Gilmore, Anna, and Martin McKee. 2004. “Tobacco-control Policy in the European Union.” In Unfiltered: Conflicts Over Tobacco Policy and Public Health, edited by Eric A. Feldman and Ronald Bayer, 219–54. Cambridge, MA: Harvard University Press. Graham, John D., and Susan Hsia. 2002. “Europe’s Precautionary Principle: Promise and Pitfalls.” Journal of Risk Research 5, no. 4 (October): 371–39. Grüning, Thilo, and Anna Gilmore. 2007. “Germany: Tobacco Industry Still Dictates Policy.” Tobacco Control 16, no. 1 (February): 2. Grüning, Thilo, Anna Gilmore, and Martin McKee. 2006. “Tobacco Industry Influence on Science and Scientists in Germany.” American Journal of Public Health 96, no. 1 (January): 20–32. Grüning, Thilo, Christoph Strünck, and Anna Gilmore. 2008. “Puffing Away? Explaining the Politics of Tobacco Control in Germany.” German Politics 17, no. 2 (June): 140–64. Hirschhorn, Norbert. 2000. “Shameful science: four decades of the German tobacco industry’s hidden research on smoking and health.” Tobacco Control 9, no. 2 (June): 242–47. Hofer, Karin. 2000. “Labelling of Organic Food Products.” In The Voluntary Approach to Environmental Policy: Joint Environmental Policy-making in Europe. Edited by Arthur Mol, Volkmar Lauber, and Duncan Liefferink, 156–91. New York: Oxford University Press. Hofmarcher, Maria. 2009. “Austrian Health Fund born.” Health Policy Monitor (14). http://www.hpm.org/en/Surveys/IHS_-_Austria /14/ Austrian_Health_ Fund_born.html (accessed July 5, 2011). John, Ulrich. 2001. “Addiction Research in Germany Makes Progress—However, It Has a Long Way to Go to Contribute Significantly to the International Community of Research and Care.” Addiction 96, no. 5 (May): 679–81. Joossens, Luk. 2004. Effective Tobacco Control Policies in 28 European Countries. Brussels: European Network for Smoking Prevention. http://www.ensp.org/ files/effectivefinal.pdf ―――, and Martin Raw. 2007. Progress in Tobacco Control in 30 European Countries 2005–2007. Bern: Swiss Cancer League. ―――, Martin Raw, and Fiona Godfrey. 2004. “The Development of European Union Tobacco Control Policy.” In Tobacco or Health in the European Union: Past,

Consumer Protection and Risk Perceptions in the European Union and Austria | 259

Present, Future, edited by The ASPECT Consortium, 99–138. Luxembourg: Office for Official Publications of the European Communities. Krause, Ernst-Günther. 2008. “Ausnahme vom Rauchverbot für geschlossene Gesellschaften streichen.” http://www.nichtraucherschutz.de/pressemitteilungen/pressemitteilungen_2008/ausnahme_vom_rauchverbot_f-ue-r_geschlossene_gesellschaften_streichen.html (accessed 17 February 2008). Kurzer, Paulette, and Alice Cooper. 2007. “What’s for Dinner? European Farming and Food Traditions Confront American Biotechnology.” Comparative Political Studies 40, no. 9 (September): 1035–58. ―――. 2007. “Consumer Activism, EU Institutions and Global Markets: The Struggle Over Biotech Foods.” Journal of Public Policy 27, no. 2 (August): 175–98. ―――, and Grace Skogstad. 2012. “The EU’s Governance of Plant Biotechnology Risk Regulation: Still Contested, Still Distinct.” Michael Howlett, ed. Next Generation Genomics in North American, Europe, and Asia. New York: Routledge Genetics and Society Series. Lesjak, Heli Annika. 2008. “Explaining Organic Farming Through Past Policies: Comparing Support Policies of the EU, Austria and Finland.” Journal of Cleaner Production 16, no. 1 (January): 1–11. Matouschek, Bernd. 2007. “Novellierungsvorschlag des Tabakgesetzes medizinisch und moralisch gesehen hoch problematisch.” Medizinische Universität Wien. http://www.ots.at/meldung.php?schluessel=OTS_20071009_OTS0243 (accessed 28 October 2007). Nathanson, Constance. 2007. Disease Prevention as Social Change: The State, Society, and Public Health in the U.S., France, Great Britain, and Canada. New York: Russell Sage. ÖÄK. (Österreichische Ärztekammer). 2006. “Nicht-Rauchen: Österreichische Ärztekammer unterstützt WHO-Aktion.” 26 May. http://www.aerztekammer. at/print.php? type=module&aid =xhtml&id =000000000020060519094107 (accessed 3 October 2009). ―――. 2007. “Weltnichtrauchertag : Plattform Nichtraucherschutzkritisiert für Gastronomie geplante Regelung.” 29 May. Available at http://www. aerztekammer.at/?aid=xhtml&id= 0000000000 20070529101606&id= 000000000020070529101606 &noedit=true&type=module (accessed 6 October 2009). Pollack, Mark and Shaffer, Gregory. 2009. When Cooperation Fails: The International Law and Politics of Genetically Modified Foods. New York: Oxford University Press. Proctor, Robert. 1999. The Nazi War on Cancer. Princeton: Princeton University Press. Randall, Ed. 2006. “Not That Soft or Informal: A Response to Eberlein and Grande’s Account of Regulatory Governance in the EU with Special Reference to the European Food Safety Authority (EFSA) .” Journal of European Public Policy 13: 402-419. Sassatelli, Roberta, and Alan Scott. 2001. “Novel Food, New Markets, and Trust Regimes.” European Societies 3: 213–44.

260 | Paulette Kurzer

Schneider, Nick K., and Stanton A. Glantz.2008. “‘Nicotine Nazis Strike Again’: A Brief Analysis of the Use of Nazi Rhetoric in Attacking Tobacco Control Advocacy.” Tobacco Control 17, no. 5 (October): 291–96. Seifert, Franz. 2006. “Regional GM Opposition as Multilevel Challenge?: The Case of Upper Austria.” Tailoring Biotechnologies 2: 11–36 Signitzer, Benno. 1978. “Austria: Media Dependence.” Journal of Communication 28, no. 3: 79–82. Simpson, David. 2005. “Austria: s=Small but Deadly.” Tobacco Control 14, no. 1 (February): 3–4. Skogstad, Grace. 2005. “Contested Political Authority, Risk Society, and the Transatlantic Divide in the Regulation of Genetic Engineering.” In Complex Sovereignty: Reconstituting Political Authority in the Twenty-first Century, edited by Edgar Grande and Louis W. Pauly, 238–60. Toronto: University of Toronto Press. Slovic, Paul. 2000. The Perception of Risk. Sterling, VA: Earthscan Publications. Studlar, Donley T. 2004. “Tobacco Control Policy Instruments in a Shrinking World: How Much Policy Learning?” In International Public Policy and Management: Policy Learning beyond Regional, Cultural, and Political Boundaries, edited by David Levi-Faur and Eran Vigoda-Gadot, 189–20. Boca Raton, FL: CRC Press. Thyrian, Jochen René, and Ulrich John. 2006. “Measuring Activities in Tobacco Control Across the EU: The MAToC.” Substance Abuse Treatment, Prevention, and Policy 1 (18 April): 9 Tiberghien, Yves. 2009. “Competitive Governance and the Quest for Legitimacy in the EU: The Battle over the Regulation of GMOs since the Mid-1990s.” Journal of European Integration 31: 389–407. Torgersen, Helge et al. 2001. “Austria: Narrowing the Gap with Europe.” In Biotechnology 1996–2000: The Years of Controversy, edited by George Gaskell and Martin Bauer, 131–44. London: Science Museum. Torgersen, Helge. 2002. “Austria and the Transatlantic Agricultural Biotechnology Divide.” Science Communications 24: 173–83. Wagner, Wolfgang, Nicole Kronberger, and Franz Seifert. 2002. “Collective symbolic coping with new technology: Knowledge, images and public discourse.” British Journal of Social Psychology 41: 323–343. Wiener Zeitung. 2004. “Smoking Ban—The Austrian Way.” 2 November. World Bank. 2003. Tobacco Control at a Glance. Washington DC: World Bank. World Health Organization/Europe. 2007. European Tobacco Control Report. Denmark.

CONCLUSION Ideas and Social Reform

( Robert Henry Cox

The reform histories of Europe’s smaller countries are much like those of other European countries. All have keenly felt the pressures of global economic adjustment. All have tried to respond by cutting back on social programs. All have attempted to balance reforms with a desire to preserve the essential elements of their welfare states. Some have been more successful than others. In many ways, as we noted in the introduction to this volume, the experiences of the smaller democracies are instructive. Because they are small, they are more sensitive to the global pressures than are the larger European countries. For this reason, the smaller countries are often leaders in experimenting with reforms and frequently are the proving grounds for many experiments that the larger countries later embrace. This makes them akin to the proverbial canaries in the coal mine, showing other countries where it is or is not safe to tread in responding to the pressures of globalization. Surveying the experiences of these countries offers three lessons for reformers. First, that reform is often facilitated when there is a shared narrative about the challenges and the direction reform should take. In this environment, a few clever ideas can serve as “coordination points.” Second, that the lateral diffusion of ideas across Europe offer important lessons for would-be reformers. Despite two decades of effort by European Union officials to foster policy harmonization, very little trickles down from Brussels into the reform processes, but policy makers do seem eager to borrow and share ideas from their counterparts in neighboring countries. Finally, that ideas themselves often outrun reality. In other words, actual reforms frequently fall short of the rhetorical goals. This reminds us that politics still matters, that policy making is still marked by compro-

262 | Robert Henry Cox

mises and bargains, and that new constellations of groups and interests sometimes emerge to shape recent reforms.

Ideas as Coordination Points We define a coordination point as a location where political actors share similar understandings of policy problems and challenges, embrace fairly similar values and priorities, and are able to find political space to bridge the other differences that separate them. Ideas are central to the construction of coordination points, as they show the relationship between a proposed course of action, an understanding of the nature of the policy challenges, and the shared values held among all policy actors. These ideas create a coordination point when key figures use them to build coalitions (Béland 2009). One idea that is prominent in many of this book’s country studies, reflecting the prevalence of this concept in Europe today, is the idea of a social investment state (Giddens 1999). This idea emerged out of the criticism of welfare states, specifically the suggestion that programs of social security were too passive and did not do enough to make people active participants in society. The advocates of a social investment state argue that social programs fail if they only provide income support. The state should not simply protect peoples’ incomes from the risks of unemployment, injury or old age, which were the objectives of the old welfare states. In addition, a social investment state must work at putting people back into the labor market. This requires investment in human capital, thus the hallmarks of the social investment state are education programs that provide people with useful workplace skills and active labor market policies, which help prepare people to reenter the labor market after periods of inactivity, brought on by layoffs, injuries, or industrial obsolescence. The idea illustrates how coordination points work. First, the idea of the social investment state identified a problem with the old model of the welfare state, namely, the passive character of benefit programs. The idea of “investment” rather than “security” provides a new set of goals for social policy, i.e., building human capital resources, not simply protecting against lost income. This idea helped many actors accept a new orientation for social policy and led them to reframe their interests. Labor unions and social democrats embraced the idea as a new state project. Conservative parties accepted this as a way to make public policy more effective by building productive capacities in the economy rather than expending resources on unproductive activities. Skilled political entrepreneurs

Conclusion: Ideas and Social Reform | 263

brought together these disparate groups to build coalitions that were able to get the reforms enacted. Still, as the cases demonstrate, there were national differences in the effectiveness of these ideas and their proponents. Juho Saari notes that Finnish reformers embraced the social investment state and were able to implement sweeping reforms based on this new idea. The reforms fell short of their goals in Finland, as Saari shows, but primarily because of an economic downturn that took place at a critical moment. Reinhard Heinisch, in contrast, finds that Austrian policy makers were unable to build the necessary coalitions. Though they used similar ideas, Austrian reformers found that the institutional configuration of Austrian corporatism worked against adoption of the new ideas. Heinisch blames institutional rigidities and Austrian conservativism for the failure of social investment reforms. Robert Cox’s chapter points to a similar difference between the Danish and Dutch cases, on the one hand, and the Belgian case, on the other. Cox argues that the differences between success and failure are attributable to the successful reframing and strength of the new coordination points. These findings are consistent with those of other studies that find the idea of the social investment more successful in Scandinavia than in the Bismarckian welfare states of continental Europe (e.g., Andersson 2006, Vis and van Kersbergen 2007, Palier 2010). The idea of lifelong learning provides another example of a coordination point in the field of education. As Ben Ansell’s chapter and a few others mention, there is a fundamental transformation under way in how the purpose of education is conceptualized. The old idea, that education was something one does in one’s youth in order to prepare for the entire working life, is seen to be inadequate to the demands of a global economy. Today’s economy requires a workforce that can continually adapt and adopt new skills. This requires periodic reentry into the educational system. In a sense, it is a different path for acquiring skills and technology. The old idea assumed that education could get one started and that experience made the worker the technological innovator. Today, technological innovation occurs too quickly for experience alone to serve as the cutting edge of innovation. Lifelong learning allows people to cope with this rapidly changing environment, and it opens a space for more nontraditional educational programs. Such programs have existed for the past forty years, especially in the United States, but they are now being promoted from boutique alternatives to the mainstream for educational innovation. Activation is another such idea in the field of labor market policy. Activation challenges the old passive idea that the purpose of social support was to provide income to people once they leave the labor force. The new ideas emphasize the importance of work and encourage people to return

264 | Robert Henry Cox

to paid employment. In their contributions to this book, Juho Saari and Jorma Sipilä both note how the identification of new social risks leads to an effort to mobilize familiar responses to these novel problems. When we think of isolation as a serious social risk in modern society, active labor market programs do more than put people to work and reduce the number of unemployment claimants. The work also gives people meaningful social interactions that improve well-being. The idea of a new social risk broadens attention beyond income security to the entire well being of a citizen. As coordination points, these new ideas serve to reframe the policy issues, identifying new policy problems or suggesting different causes for issues that are already well understood. Also, they reshape policy preferences, sometimes by causing individuals to reassess their own understandings of the policy problems and sometimes by fostering new coalitions of interests as previously opposed factions come to see the problem in a similar way.

The Newest Politics of the Welfare State Establishing a common coordination point builds coalitions, but it does not eliminate politics. The experiences of the smaller countries remind us that the old politics are particularly enduring. Institutions of corporatist negotiation, which facilitated labor peace during the period of welfare growth, have proven to be vexing obstacles to reform. In some countries, the distributional coalitions that inhabit these institutions have been moved to embrace reform, for example in Denmark, the Netherlands, Spain, and Portugal. However, in countries such as Austria and Belgium, corporatist councils have provided redoubts for those who resist challenges to the benefits and privileges they have enjoyed. The experiences of the smaller democracies show that the fundamental distributional coalitions that have shaped welfare policies for decades still have a powerful effect on current efforts to reform. Despite the changes new ideas and new alliances bring, the old cleavages between labor and management and the fundamental struggle between Left and Right dominate. These old alliances are slower to move and consequently exert a powerful inertial effect on policy making. In countries such as Belgium or Austria, reform has been frustrated not by the lack of new ideas but by the inability of novel thinkers to unseat the forces defending the old politics of the welfare state. While old politics find refuge in corporatist institutions, the new political ideas open space for new groups who move into the policy debates. Thus, debates about new social risks make single mothers an important

Conclusion: Ideas and Social Reform | 265

category of citizen, as Jorma Sipilä discovered. Or, as Paulette Kurzer notices, the discussion of smoking bans has raised the profile of groups that advocate broader approaches to issues of health and environmental safety. The result is a much more fluid environment. While the new coordination points have broken up some traditional political alliances, those groups have not disappeared. Instead, they now either challenge the new groups or they realign with the new groups into completely new constellations of interests. Thus, as Jane Gingrich, as well as Kieke Okma and her coauthors, observe, the efforts to introduce market incentives in health policy place health care providers, insurers, and patients in different relationships with one another. By privileging ideas of market efficiency and patient choice, new ideas in health care have unseated the primacy of physicians as the health care experts and challenged the privileged payment arrangements between state ministries and insurers. Or, as the Austrian example shows, debates over public health can have paradoxical outcomes as Austrians set an innovative course in prohibiting genetically modified organisms while resisting smoking regulation. In the Netherlands, responding to the needs of working women created unusual alliances between women and employers over the issue of flexibility in the workplace. The new ideas reconfigured the political coalitions, but the fundamental dynamics of political exchange have not changed. If anything, expert knowledge and historical experience are treated with more suspicion and less deference.

How Do Ideas Travel? Another important lesson from the smaller democracies of Europe is the way ideas travel, or diffuse, across Europe. At the moment, policy ideas are moving in two directions across Europe. One represents a top-down imposition from the level of the European Union to the member states. The other pattern is a lateral movement across countries, driven by a process of policy borrowing (Radaelli 2000). Because all the countries examined in this volume are members of the European Union (EU), it offers an opportunity to conclude that the cross-lateral diffusion has greater effect than the coordinating efforts of the EU. For years the EU has encouraged member states to reform welfare states in a direction similar to the social investment state. For the EU, reforming welfare states was a concern that resulted from the larger goal of building a single market. Differential levels of welfare support were seen to be impediments to the main pillars of the Single Market, for example, different tax rates affect the costs of doing business, different regulations affect the rights and benefits European citizens are able to enjoy when moving from

266 | Robert Henry Cox

one jurisdiction to another, and different levels of public spending and regulation affect the relative attractiveness of locales. Yet the EU is constitutionally limited to activities that integrate markets. Because social policy falls officially outside the scope of the single market; the jurisdiction of the EU even to promote policy reform in this area is disputed (Scharpf 2002). As a result, the EU has adopted “soft” policy instruments to encourage member states to adopt reforms on their own. For ten years, from 2000 to 2010, this effort at policy coordination was known as the Lisbon Strategy. It involved articulating a number of policy objectives, holding a series of meetings with officials in the member states so they could “benchmark” and share best practices, requiring the member states to submit reports on their progress that were then evaluated by officials in Brussels and heaping praise on those countries that were reforming well and scorn on those making little progress. After ten years, those running the Lisbon Strategy, as well as outside observers, agreed that the program had failed miserably (Idema and Kelemen 2006). Nonplussed by such assessments, the European Commission announced in 2010 another ten-year plan to reform welfare states, called Growth and Jobs. Early speculation is that this program, too, will have little effect. In this volume, Juho Saari, Robert Cox, and Paulette Kurzer each find little influence from the European level. Indeed, in some cases policy reformers were resistant to toeing the line from Brussels. While prodding from the EU seems to have had little effect in spreading ideas about welfare reform, many of the country studies in this volume show that reform ideas did travel laterally across countries. Rather than EU officials, the major drivers of this movement seem to be Social Democratic parties, epistemic communities of civil servants and nonprofit experts, as well as the academic community. The social investment state is an idea that arises from efforts by Social Democratic parties to articulate a new vision for welfare states. Policy reformers, both politicians and bureaucrats, in every country were learning that their counterparts in neighboring countries shared similar problems and had already attempted to address them. Such experiences allowed reformers to learn by someone else’s example, rather than by their own trial and error. Indeed, it was this type of learning that the EU hoped to foster in its reform strategy, though the experiences show that the lessons were learned only when policy makers believed they needed to learn them.

Conclusion: Ideas and Social Reform | 267

Reflections The smaller democracies of Europe are bellwethers, often confronting new policy challenges earlier than their larger neighbors. Their experiences, however, are often overlooked by a scholarly community more focused on the larger countries. This book is intended to serve as a corrective to this imbalance. By examining the experiences of some of Europe’s smaller countries and by comparing successful with not-so-successful reformers, we learn much that drives the process of reform in European countries. Moreover, we see that some of the earliest experiments were undertaken in the smaller democracies. Many of the efforts to reform welfare states in these countries trace to the early 1990s and are now becoming the norm even in the larger European democracies. The smaller democracies of Europe offer a variety of experiences. Some have been successful, some have not. Some have been clever, while others have been rather dull. Regardless of the nature of the lesson, policy leaders in the larger countries would do well to learn from these experiences and draw out the helpful messages. They should reflect on the findings of this book.

Bibliography Andersson, Jenny. 2006. Between Growth and Security: Swedish Social Democracy from a Strong Society to a Third Way. Manchester, UK: Manchester University Press. Béland, Daniel. “Ideas, Institutions, and Policy Change.” 2009. Journal of European Public Policy 16, no. 5: 701–18. Giddens, Anthony. 1999. The Third Way: The Renewal of Social Democracy. London: Polity Press. Idema, T., and D. R. Kelemen. 2006. “New Modes of Governance: The Open Method of Coordination and Other Fashionable Red Herring.” Perspectives on European Politics and Society, 7, no. 1: 108–23. Palier, Bruno. 2010. “Ordering Change: Understanding the ‘Bismarckian’ Reform Trajectory.” In A Long Goodbye to Bismarck? The Politics of Welfare Reform in Continental Europe. Edited by Bruno Palier, 19–44. Amsterdam: Amsterdam University Press. Radaelli, Claudio M. 2000. “Policy Transfer in the European Union: Institutional Isomorphism as a Source of Legitimacy” Governance 13, no 1: 25–43. Scharpf, Fritz W. 2002. The European Social Model. Journal of Common Market Studies, 40, no. 4: 645–70. Vis, Barbara, and Kees van Kersbergen. 2007. “Why and How Do Political Actors Pursue Risky Reforms?” Journal of Theoretical Politics 19, no. 2: 153–72.

CONTRIBUTORS

(

Ben W. Ansell is assistant professor of political science at the University of Minnesota, where he teaches graduate and undergraduate courses in political economy, international relations, and comparative politics. Ansell has also worked as an academic consultant to HM Treasury in the UK and for the Leitch Review of Skills, which is advising the UK government on long-term education policy. His current research interests focus on the politics of education policy, the relationship between inequality and democracy, and the effects of asset price inflation on political preferences. Toni Ashton is associate professor in health economics and director of the Centre for Health Services Research and Policy in the School of Population Health, University of Auckland, New Zealand. She has researched and published widely on the organization and funding of health care systems and health reform. She has been a member of various government working groups on health policy and serves on the editing board of four academic journals. She has consulted extensively for international agencies including the World Health Organization and the OECD, and nationally for government agencies, health-professional bodies, and nongovernmental bodies. Iva Bolgani works as a consultant at the Sezione sanitaria in the Canton of Ticino, with a special focus on improving the quality of hospital services and contracting with public hospitals in the canton. She teaches at universities in Geneva, Lausanne, and Lugano. She has served as an expert to the Swiss national committee on health prevention and communication and is a member of several intercantonal commissions on health policy. She has published several articles in both Swiss and international journals, as well as one book.

270 | Notes on Contributors

David Chinitz, PhD University of Pennsylvania, has held several positions with Israeli government agencies and universities in Israel and the United States. He is the former chair of the European Health Management Association, a member of the management board of the International Society for Priorities in Health Care, and serves as a consultant to the World Health Organization. Chinitz has lectured and published widely on comparative health systems, health policy, and regulation. Gary B. Cohen is professor of history at the University of Minnesota, where he teaches courses on modern European social and political history and East-Central Europe in the nineteenth and twentieth centuries. His research has focused on social development, ethnic-group relations, and education in modern Austria and the Czech lands. His publications include two books, The Politics of Ethnic Survival: Germans in Prague, 1861– 1914 (Princeton University Press, 1981; second edition, revised, Purdue University Press, 2006) and Education and Middle-class Society in Imperial Austria, 1848–1918 (Purdue University Press, 1996); articles in The Journal of Modern History, Central European History, The Austrian History Yearbook, The East European Quarterly, Jewish History, and The Social Science Quarterly; and numerous book chapters. Robert Henry Cox is professor in the School of International and Area Studies at the University of Oklahoma and is coordinator for European studies. Professor Cox teaches classes at the graduate and undergraduate levels in comparative politics, specifically focusing on European politics, comparative public policy, and globalization. He also serves as coeditor for the journal Governance, and is the codirector of the European Union Center at Oklahoma University. Professor Cox’s research focuses on the development and evolution of welfare states in advanced industrialized countries. Reflecting his interdisciplinary interests, his writings on the welfare state have explored the moral foundations of welfare programs, the politics of policy change, and the technical details of systems of taxation and social insurance. Luca Crivelli received his PhD from the University of Zurich and is currently professor of economics at the University of Applied Sciences of Southern Switzerland, as well as an adjunct professor at the University of Lugano. He has been director of the master’s program in health economics and management at the University of Lugano and board member of the Swiss School of Public Health. Crivelli has published several articles on cross-border care (in the Swiss Journal of Economics and Statistics), regulation of pharmaceuticals (Journal of Regulatory Economics), federalism and

Notes on Contributors | 271

health expemditure (Health Economics), efficiency of health care institutions (International Journal of Health Care Finance and Economics) and health insurance reforms (Revue francaise des affaires sociales). Scott R. Eliason is associate professor of sociology at the University of Arizona. He has published widely in the United States and Europe on quantitative methodology and statistics; the sociology of work, occupations, and labor markets; economic sociology; stratification; and life course. Jane Gingrich is assistant professor of political science at the University of Minnesota. She was also a Max Weber Fellow at the European University Institute (2008–2009). She has recently completed a book manuscript “Whose Market Is It Anyways? Making Multiple Markets in the Welfare State,” which looks at the politics of market reform in public service in the Netherlands, Sweden, and the United Kingdom. Her current research continues this analysis of how policy makers alter and tweak incentives by focusing on the increasing use of insights from behavioral economics by politicians and bureaucrats. She is also working on a project examining how different national and subnational political institutions shape public opinion toward government activity. William Hamilton holds two master’s degrees, one in economics from the Andrew Young School of Policy Studies at Georgia State University and one in sociology from the University of Arizona. His current research explores how job insecurity shapes individual preferences for social policies and redistribution. Reinhard Heinisch is Professor of Austrian Politics in Comparative European Perspective and head of the Department of Political Science at the University of Salzburg, Austria. He has published widely in Europe and North America, is an active contributor to the Wiener Zeitung, and is also a commentator on the Austrian evening news. Paulette Kurzer is professor of political science at the University of Arizona, where she teaches courses in comparative politics, European politics, and advanced industrialized states. Kurzer’s research interests include institutions of the European Union and the intersection between domestic and European policy making. Her work has appeared in the Journal for European Public Policy, Journal of Common Market Studies, and other refereed journals, as well as several edited volumes. Her most recent book is Markets and Moral Regulation (Cambridge University Press, 2001). She is

272 | Notes on Contributors

currently working on the perception of risk and the formation of a new public health domain in the European Union. Meng-Kin Lim is associate professor of health policy and management at the Yong Loo Lin School of Medicine, National University of Singapore. A physician by training, Dr. Lim was chief executive officer of the Health Corporation of Singapore and served on eight hospital boards and numerous expert committees. Dr. Lim has published widely in peer-reviewed journals, including New England Journal of Medicine, Health Affairs, Health Policy, Medical Care, Quality and Safety in Health Care, and the British Journal of Public Health Medicine. He serves on several journal boards and represented Singapore on the World Health Organization’s Western Pacific Advisory Committee on Health Research. He has consulted extensively for the World Bank, WHO, the Asian Development Bank in China, and for several other countries in Asia, the Middle East, and Eastern Europe. He is a recipient of the Republic of Singapore’s Public Service Star, the National University of Singapore’s Special Commendation Award for Teaching Excellence and the School of Medicine’s Teaching Excellence Award. Hans Maarse is a professor of health sciences at the University of Maastricht. He has consulted frequently for the World Bank and the World Health Organization. His current field of interest is the impact of the EU on policy making, comparative health system analysis, and market reforms in health care, in particular in The Netherlands. He has published on these topics in the Journal of Health Politics, Policy and Law, Health Policy, Health Care Analysis, the European Journal of Health Economics, EuroHealth, Journal of Medicine and Philosophy, BMC Health Services Research, and Intereconomics, and has contributed several book chapters. He is a member of the supervisory board of several health care providers in The Netherlands. Rachel Meislin is a recent graduate of the University of Pennsylvania with a BA in Health and Societies and has interned as a research associate at the Department of Health Policy and Management, School of Public Health, Hebrew University–Hadassah, Jerusalem. Kieke G. H. Okma received her PhD at Utrecht University, The Netherlands, and is currently an associate professor at the Wagner School of Public Service, NYU, as well as a visiting professor at Columbia University, Cornell University, and Catholic University, Leuven. She has worked with a variety of government agencies and international organizations for over twenty-five years and has lectured and published widely on topics of public policy, health politics, and international comparison. She serves on the

Notes on Contributors | 273

board of several journals including Health Policy, Journal of Health Politics, Policy and Law,and the Journal of Health Services, Research, and Policy. Her recent publications include: Six Countries, Six Reform Models: The Healthcare Reform Experience of Israel, The Netherlands, New Zealand, Singapore, Switzerland, and Taiwan, as coeditor with L. Crivelli (World Scientific Publishers, 2010); Comparative Studies and the Politics of Modern Medicine, as coeditor with Theodore R. Marmor and Richard Freeman (Yale University Press, 2009); Learning and Mis-learning across Borders: “What Can We (Not) Learn from the 2006 Health Care Reform in the Netherlands? Commentary on Rosenau and Lako,” in the Journal of Health Policies, Politics, and Law (2008); and a book review essay on Better But Not Well: Mental Health Policy in the United States Since 1950, by Richard G. Frank and Sherry A. Glied, which appeared in the Journal of Health Politics, Policy, and Law (February 2008). Robin Stryker is professor of sociology and affiliated professor of law at the University of Arizona, where she also is Earl H. Carroll Magellan Circle Fellow and Faculty Research Liaison for the National Institute of Civil Discourse. She currently is a visiting professor at the Ecole des Hautes Etudes en Sciences Sociales, in Paris. She has been a member of the Discrimination Research Group, funded by the Ford Foundation, the American Bar Foundation, and the Center for Advanced Study in the Behavioral Sciences. She writes broadly on theory, methods, the welfare state, organizational and institutional change, and law, politics, and inequality. She has a National Science Foundation grant for her study of “Rights and Their Translation into Practice: Toward a Synthetic Framework.” Juho Saari is professor of sociology at the University of Kuopio, where he focuses on the Finnish welfare system. Professor Saari has also served as the senior advisor at the Finnish Ministry of Social Affairs and Health, where his duties included strategic planning of social and health policy both within Finland and the European Union. Jorma Sipilä is currently professor of social policy and social work at the University of Tampere, Finland, where he also serves as chancellor. Published widely in Europe, Professor Sipilä has written since the 1990s on avant-garde solutions in social work and informal care in Finland. Tim Tenbensel is a senior lecturer in health policy at the School of Population Health, University of Auckland, New Zealand. He serves as editor and member of the advisory board of Policy and Politics Journal. He has written widely on topics of health reform, cost control, decentralization,

274 | Notes on Contributors

interest groups, governance and evidence-based policy decisions in the Journal of Comparative Policy Analysis, Oxford Textbook of Public Health, Social Science and Medicine, Journal of Health Services Research and Policy, New Zealand Government and Politics, Policy Studies, and Public Management Review, and has contributed several book chapters. Eric Tranby is an assistant professor in the Department of Sociology and Criminal Justice at the University of Delaware. His has studied gendered employment inequality, comparative family policies, workplace policies, and symbolic boundaries in American life. His work has been published in American Sociological Review, Social Problems, the Journal for the Scientific Study of Religion, Research in Social Stratification and Mobility, as well as in edited volumes. Tsung-Mei Cheng is a health policy research analyst at the Woodrow Wilson School of Public and International Affairs, Princeton University. She is an expert on comparative health systems with an emphasis on Asian countries. Cheng is the cofounder of the Princeton Conference, an annual national conference on health policy that brings together the US Congress, government, and the research community on issues affecting health care and health policy in the United States. She is an advisor to the China National Health Development Research Center (CNHDRC), the official Chinese government think tank for health policy under China’s Ministry of Health. She is also an advisor to NICE International, an agency of the National Institute for Health and Clinical Excellence (NICE), which advises the United Kingdom’s National Health Service (NHS) on clinical and public-health coverage decisions. In addition, Cheng is a member of the International Advisory Group of Academy Health, the US-based professional association of health-services researchers. Cheng also serves on the International Advisory Board of the Elsevier online Encyclopedia of Health Economics. Sara Watson is assistant professor of political science at Ohio State University. Her research focuses on the comparative political economy of Western Europe, with an emphasis on the politics of the welfare state. She is currently completing a book manuscript based on her dissertation, “The Left Divided: Parties, Unions, and the Development of Welfare Capitalism in Post-Authoritarian Southern Europe.”

INDEX

( A activation, 22, 24-25, 27, 33, 72-73, 84, 263. See also labor markets ˙ del reforms, 120-21 A Alliance for the Future of Austria (BZÖ), 252 Austria, 2-4, 11, 34, 36, 38, 50, 93, 99, 184, 215, 263-65 higher education in, ch 9 passim. See also higher education social policy in, ch. 6 passim. See also health care; social consumption; Social Democrats; social expenditure; social investment tobacco policy in, ch. 10 passim. See also tobacco Austrian Health Reform legislation (2005), 252. See also tobacco B Barro-Lee estimate, 40 Basic Law (Germany), 223 “bed blocking”, 119 “blocked societies”. See social investment state bioengineering. See genetically modified organisms (GMOs) biotechnology, 239, 242, 247-51, 255, 56 Blair, Tony, 82 “blame avoidance”, 19 “breadwinner” concept, 23 British National Health Service (NHS), 181, 184

Bureau of National Health Insurance (BNHI [Singapore]), 203 business competitiveness index, 60 C CACE (compilers average causal effects), 36-37, 39-53 Canada, 36-38, 85, 91-92 cash expenditures, 94 Cederschiöld, Carl, 123 child care, 5, 7, 10, 22, 39, 59, 63, 70 in Chile, 194 and female employment, 22-23, 33, 35, 86-87, 94-98, 100 in Finland, 70 and the social investment state, 84. See also social investment state in Sweden, 120, 123 Chernobyl. See genetically modified organisms (GMOs) Chamber of Doctors (Österreichische Ärztekammer—ÖÄK), 252 Chile, 179, 182, 193-95, 205. See also health care reform Christian Democrats Christian Democratic Parties in the Netherlands, 187 in Germany, (CDU-CSU) 23, 223-25 in Sweden, 112 and corporatism, 23 Clinton, Bill, 22 collective bargaining, 21

276 | Index

in Iberia, 11, ch. 7 passim social bargaining, 155, 160-61 See also corporatism in the Netherlands, 189 in Israel, 196 commodification, 98. See also social investment state “competitive individualism”, 182 Confederación Española de Organizaciones Empresariales (CEOE), 163, 166-67 consensual political system, 3-5, 256 Conservative Party, (Austria), 143, 145-46 Constitutional Reform Act of 1995 (Finland), 63 “coordination points”, 2, 261-65 defined, 262 corporatism, 1, 4, 5, 21, 23, 38, 25, 182, 189, 264 in Austria, 134-37, 139, 141-42, 249, 263 autonomous corporatism, 136 corporatist councils in Austria, 264 in Belgium, 28, 264 in Denmark, 28 in the Netherlands, 28 corporatist negotiation, 264 in Iberia, 157 state-corporatist systems, 157 in Israel, 182 microcorporatist bargaining, 158, 164. See also collective bargaining neocorporatism, 182 in Austria, 132, 136 in the Netherlands, 186 in Israel, 196 and social partnership, 1. See also Social Partnership in Switzerland, 182 Crown Health Enterprises (CHEs), 198 Council of Environmental Ministers, 253 Czech Republic, 243

D decommodification, 21, 98. See also welfare state Dehaene, Jean-Luc, 26 Democratic Progressive Party (DPP [Taiwan]), 202 Denmark, 2, 5, 9, 18, 24-30, 36, 76, 9192, 94, 96, 99, 219, 242, 264 dirigiste parties, 23 “disorganized decentralization”, 154, 156 Dornelas, António, 168 dual system (Switzerland), 232 E “eco-social market” policy, 250 elderly care, 6, 10, 185 in Sweden, 110, 119-25 Esping-Andersen, Gøsta, 18 Politics Against Markets, 81 The Three Worlds of Welfare Capitalism, 81 Estonia, 243-53 European Central Bank, 77, 161 European Commission, 238 and health legislation, 241 and gender discrimination, 24 and genetically modified (GM) corn, 253 and precautionary principle, 254 and welfare reform, 266 EU council of ministers, 241 European Court of Justice (ECJ), 75, 243 European Food Safety Authority, 242, 253-54 Europeanization, 4, 8, 16-18, 21, 27, 29, 74, 131 European Monetary Union, (EMU) 77, 160-61, 170 and Stability and Growth Pact, 74 European Parliament, 239 and Genetically Modified Organisms (GMOs), 242, 256 and Lisbon strategy, 82. See also genetically modified organisms (GMOs)

Index | 277

European Union (EU), 4, 18, 61, 107, 137, 238, 241, 261, 265 and health care systems, 108 and the Single Market, 137, 140 See also Single Market and smoking rates, 238-39, 243. See also Europeanization; tobacco “Euro-sclerosis”, 8 F Family Doctor reforms, 117-18, 122 farming, organic, 239, 249-51, 256 Federation of County Councils (Sweden), 115 Federation of Austrian Social Insurance Carriers, 142. See also welfare state fertility and education, 96, 100 and female employment, 87, 94-97, 100 rates, 34, 96-7, 98, 100-105 Finland, 2, 5, 36, 82, 85, 93, 219, 242, 253, 263 public spending in, ch. 3, passim. See also health care; Social Democrats; social expenditure; social insurance; social investment; welfare state flexicurity, 3, 25, 72, 78 defined, 29 France, 2, 36, 40, 74, 93, 99, 184, 242, 245, 252 Fondo Nacional de Salud (FONASA [Chile]), 193-94 Freedom Party of Austria (FPÖ), 13334, 136, 140-43, 146, 226, 252. See also Haider, Jörg Fukushima nuclear plant. See genetically modified organisms (GMOs) G “general efficacy” principle, 163 general practitioner (GP), 189, 196-98 family physicians, 184

Genetically Modified Organisms, (GMOs) 11-12, 238-42, 247-49, 253-55 bioengineering, 247, 249-50 Deliberate Release Directive, 24 European attitudes toward, 247-48, 251 and Chernobyl, 249 “dread factor,” 248 and Fukushima nuclear plant, 249 “GM-free” stance, 254 See also European Parliament; European Union Germany, 2, 11, 27, 34, 36, 38, 74, 9394, 99, 163, 182, 184, 191 higher education in, ch.9 passim. See also higher education tobacco policy. See tobacco. Giddens, Anthony, 93-94 The Third Way, 81 globalization, 1-4, 6, 59, 74, 161, 261 and Europeanization, 17-18, 21-22, 29 and higher education, 235 and welfare reform, Belgium, 27 and welfare state, 18 “Global Plan”, 26 global warming, 237 grand coalitions, 11 in Austria, 147 and Austrian education, 222, 229 and German education, 222-23 Greens (Austria), 252 H Haider, Jörg, 133-34, 141, 143, 145-46 health care 7, 11, 59, 179, 180-83, 2045, 265 in Austria, 138, 144 consumer-driven, 185, 186, 192 competitive model of, 193 complexification, 205 integrated model of, 181 in Finland, 70, 72, 74 funding for, 181, 183-84, 186 managed care (managed competition), 185, 191-92

278 | Index

in Portugal, 169 privatization, 185 health care reform definition of, 186 after-reform maintenance, 205 in Chile, 193-95 in Israel, 195-97 in the Netherlands, 187-90 in New Zealand, 197-99 dual system, 197 post-reform maintenance, 180, 186, 190 in Singapore, 200-202 in Sweden, ch 5 passim in Switzerland, 190-93 in Taiwan, 202-204 voice and exit strategy, 183, 190, 197, 199 window of opportunity for, 180, 187, 195, 202 health care systems. See health care Health Corporation of Singapore (HCS), 200 “hierarchical collectivism”, 182 higher education, 3, 11, 217, ch.9 passim in Austria, ch. 9 passim Fachhochschule, 228, 233 General University Studies Act, 227-28 Gymnasium system, 227, 233 Hauptschule system, 227 and Länder, 223 in Germany, ch. 9 passim de-Nazification, 215 Gymnasium system, 226 Higher Education Framework Act, 224 and Länder, 234 reform deadlock (Reformstau), 225 reforms, 224 BAFöG, 224 trilemma of, 217-231 public spending on, 219-220 Power Resources literature, 221 in Switzerland, ch. 9 passim See also grand coalitions

horizontal principle, 66 basic principles of, 62-63 human capital, 5-6, 33, 82, 84, 216, 262 human capital development, 33, 51 and social investment, 82, 85. See also social investment Humboldtian mission, 215, 226 Hungary, 243 I “increasing returns” logic, 5 in kind arrangements in kind benefits, 94 cash in kind, 91-92 expenditure in kind, 94 services in kind, 63, 71 “insider-outsider problem”, 23 Instituciones de Salud Previsional (ISAPREs), 193-94 Interconfederal Agreement on the Coverage of Gaps (AICV [Spain]), 167 Interconfederal Agreement on Employment Stability (AIEE [Spain]), 167 Interconfederal Agreement on Collective Bargaining (AINC [Spain]), 167 interest mediation Finnish model of, 69, 71 Israel, 179, 182-83, 186, 195-97, 205. See also health care reform Italy, 2, 34, 36, 38, 74, 93-94, 96, 99, 184, 217, 219, 242, 252 J Johns Hopkins University, 215 K Katzenstein, Peter, 29, 132 Small States in World Markets, 4 Katznelson, Ira, 168 Knightian uncertainty, 66-67 Kohl, Helmut, 223 Krankenkasse, 184 Kuomintang (KMT [Taiwan]), 202

Index | 279

L labor force, 10, 18, 20, 22, 29, 83-85, 263 female participation in, ch. 2 passim in Iberia, 160, 166 in Finland, 60, 64, 81, and social investment, 82 and social investment state, 82 See also labor markets; social investment; social investment state labor markets 4-5, 7, 10-11, 22-24, 85, 98, 153, 262-63 active labor market programs, 20, 264. See also activation in Austria, 131, 137-38, 140, 146, 226 in Belgium, 26-28 in Denmark, 25, 28 and education, 33 female participation in, 94, 98 ch. 2 passim in Finland, 64-65, 70, 73, 75, 84 and flexicurity, 30. See also flexicurity in the Netherlands, 24-25, 28 in Portugal, 6, 155, 157 in Spain, 154-155, 157 160-68, 170-71 in Switzerland, 232 Labour Party (New Zealand), 199 Lacina, Ferdinand, 137 Liberal Party (Sweden), 120 lifelong learning, 20-21, 72-73, 263 Lipponen, Paavo, 82 Local Government Act (Sweden), 120 Luxembourg, 3, 91-92, 99, 184 M Maastricht criteria, 139 managed care. See health care managed competition. See health care market liberalization, 7, 11, 69, 131-32, 134, 136-37, 139 and social investment, 6-10. See also social investment Marx, Karl, 21 Medifund (Singapore), 200 Medisave (Singapore), 200 Medishield (Singapore), 200.

See also health care reform Ministry of Agriculture (Austria), 249 Ministry of Health (Israel), 196 Ministry of Labor (Spain), 159 Ministry of Social Affairs and Health (MSAH [Finland]), 65 Moderate Party (Sweden), 112, 117, 120, 125 “modernization brokers”, 131 Molterer, Wilhelm, 146 N National Competition Authority (NMa [the Netherlands]), 189 National Health Insurance (NHI [Taiwan]), 202-204 National Health Insurance Institute (Israel), 196 National Health Insurance Law (NHI [Israel]), 196-97. See also Netanyahu Commission National Health Insurance Regulator (Israel), 195 National Health Service for Employees (SERMENA [Chile]), 193 National Party (New Zealand), 198 Netanyahu Commission, 195 Netherlands, 3, 5, 9, 18, 24-26, 29-30, 36, 171-72, 179, 182-84, 186-91, 195-205, 242, 253, 264-65. See also health care reform New Zealand, 69, 179, 184, 197-99, 205, 219-20, 226, 229 New Zealand Health Strategy, 199 Primary Health Care Strategy (New Zealand), 199. See also health care reform neoliberal reforms, 1, 2, 17, 22, in Austria, 133, 141-42, 145 in Belgium, 29 in Chile, 193 and German education, 193 in New Zealand, 69 Norway, 3, 36, 38, 85, 92-93, 96, 99, 252 nuclear energy, 249 opposition to, 251 Nuclear Power Prohibition Act, 238

280 | Index

O old-age expenditure, 86-87 “one-and-a half-jobs model”, 24-25 Organization for Economic Cooperation in Europe (OECD), 60, 83, 85, 88, 91, 94-95, 100, 119, 153, 170, 201, 216-17, 219, 222, 226, 233, 243 P “path dependent” effects, 9, 19, 61-62, 99 increasing returns, 19 “lock-in effects”, 19 PAYGO, 19 Pierson, Paul, 19 People’s Action Party (Singapore), 200 People’s Party of Austria (ÖVP), 133, 136, 139, 141, 145-46 and Austrian education, 226-29 agricultural policy, 249 See also Schüssel, Wolfgang political entrepeneurs, 262 political exchange, 10, 265 in Iberia, 77, 154-55, 160-62, 165-66, 168, 170-72 Poland, 93, 96, 99, 243 Portugal, 6, 11, 242, 264 collective bargaining in, ch. 7, passim See also collective bargaining; labor force; labor markets Portuguese Confederation of Industry (CIP), 168 collective bargaining in, ch. 7, passim See also collective bargaining; labor force; labor markets Primary Health Organizations (PHOs [New Zealand]), 199 private-public-partnership (PPP) models, 73 public health insurance, 186, 205 in Austria, 133 in Chile, 195 in Israel, 195 public sector expansion, 35-36

R “rainbow” government (Finland), 68 Reagan, Ronald, 112 Regional Health Authorities (RHAs), 198 Rotterdam Social Service Office, 24 S Sailas lists, 69 St. Göran’s reforms, 122 “sectarianism”, 182 Scharp, Fritz, 18 Schröder, Gerhard, 224 Schüssel, Wolfgang, 136, 141, 143, 146 Sigurdsen, Gertrude, 116 Singapore Medicine, 201 Single Market, 18, 20-21, 131, 134, 137, 241, 253, 265-66 Euro Zone, 140 Sistema Nacional de Salud (SNS [Chile]), 193 Singapore, 179, 182, 200-202, 205. See also health care reform Slovakia, 243 smoking. See tobacco social bargaining. See collective bargaining social citizenship, 20 social consumption, 7, 84, 99 flat-rate cash subsidies, 86 See also social investment Social Democrats Social Democratic parties, 21-22, 35 in Austria (SPÖ), 133, 137, 139142, 145, and the Austrian welfare model, 131 and education, 226-30. See also higher education and genetically modified organisms (GMOs), 249 in Belgium, 29 in Finland, 68 in Germany, 223-225 in Sweden (SAP), 112-13 and health care, 114, 118 and elderly care, 120, 122, 124 in Switzerland, 232

Index | 281

and the social investment state, 266 social expenditure, 59, 82, 84, 85-89, 98-99 in Austria, 133 and Eastern European politics, 99 in Finland, 61, 65, 69, 71, 72 in Scandinavia, 90, 98 and Southern Europe, 99 social insurance, 1, 20, 24, 26, in Austria, 132, 135, 138, 142-43, 147 in Belgium, 28 in Finland, 62-63, 75, 105 in Germany, 184 in Israel, 195 in the Netherlands, 172, 184, 188 in Switzerland, 190 in Taiwan, 202 social investment, 5, 7-11, 22, 82-83, 85, 89, 93-94, 98-101, 263 defined, 1, 85 in Finland, 85-86 and market liberalization, 6-10 and social consumption, 83, 84, 87, 93, 98 social investment index, 87, 91-93, 97 social investment state, 9, 10, 81-85, 93, 100, 262, 265-66 “blocked societies”, 93 Continental model of, 82 Scandinavian model of, 82 social market economy (Sozialmarktwirtschaft), 23 Social Partnership (Austria), 136-37, 139-40, 142, 147 social policy models, 182 Social Security Act (New Zealand), 197 Social Security Coordination Regulation, 75 Sorsa mediation, 69 Spain, 11, 96, 99, 184, 264 collective bargaining in, ch. 7, passim See also collective bargaining; labor force; labor markets Stockholm County (Sweden), 115 Stöger, Alois, 252 “Stop Law”, 118

Sweden, 10, 36, 38, 69, 76, 91-94, 96, 219, 221, 223, 225, 242, 252-53 health care reform in, ch. 5 passim See also health care reform sustainable development index, 60 T Taiwan, 8, 182, 202-205 See also health care reform TEL pensions, 76 Teng-Hui, Lee, 202 Thatcher, Margaret, 112 tobacco advertising regulations, 241-44, 247 Tobacco Advertising Directive (TAD), 241, 243 and biotech crops, 238 and cancer, 241, 246 control, 140, 241, 243, 251-52, 255 public health spending, 244-45 and genetically modified organisms (GMOs), 239, 241 pro-tobacco forces, 245-47, 255-56 and the “Nazi card”, 246 “nico-Nazis”, 246 smoking rates, 239-41, 243 tobacco monopoly (Austria) Benson&Hedges (cigarettes), 246 Austria Tabak, 246 Gallaher, 246 Ministry of Finance, 246 Silk Cut (cigarettes), 246 U United Kingdom, 2, 36, 74, 92. See also British National Health Insurance United States, 85, 89-93, 95-96, 98, 108, 182, 193, 204, 217, 219-220 University of Basel, 230 V “varieties of capitalism” tradition, 160 “vice into virtue” strategy, 162 Vranitzky, Franz, 137 W wage restraint, 166, 170-72

282 | Index

Waitangi, Treaty of, 197 Walloon Socialist Party, 28-29 welfare state, 1-4, 7, 10-12, 17-21, 3436, 41, 47, 66, 74, 78, 182, 184-85, 261-67 in Austria, 131-40, 144, 147-48 conservative welfare regime, 138 basic supplemental security income (bedasfsorientierte Mindestsicherung), 147 childbirth subsidy (Geburtenbeihilfe), 140 infant care money (Kinderbetreuungsgeld), 142 retrenchment packages (Sparpakete), 140-41 social assistance (secondary welfare system), 132, 138-39, 147

in Belgium, 27-29 passive welfare state (Belgium) 27-28 decommodifying welfare state, 82 in Denmark, 18, 24-30 in Finland, 9, 59, 61, 74-75, 77 Golden Age of, 59 passive welfare spending, 84 “Polish plumber”, 75 and Social Democratic Parties, 21 in Sweden, 107, 112-14, 120 welfare state retrenchment, 81, 131 Westerberg, Bengt, 120 Wiener Zeitschrift fur Suchtforschung, 245 Workers’ Statute (Spain), 163 world competitiveness index, 60 World Trade Organization, 253