EIJ THE WORLD BANK 25302 January 2003 New Social Policy Agendas for Europe and Asia Challenges, Experience, and Lessons Edited by Katherine Marshall, Olivier Butzbach Govt announces subsidy cuts, higher taxes for 2003 budget Unemployment figures hide depth of Indonesia's despair Street children hope to be breadwinners of poor families "Setaglianopensiome sanitk la reazione sara durissima" Japan wants retirees to go back to work [11-prepared graduates Aan't fnmd matching jobs 500 infants die every month in W. New Social Policy Agendas for Europe and Asia Challenges, Experience, and Lessons Cover designed by Naylor Design, Inc. New Social Policy Agendas for Europe and Asia Challenges, Experience, and Lessons Edited by Katherine Marshall Olivier Butzbach THE WORLD BANK C 2003 The International Bank for Reconstruction and Development / The World Bank 1818 H Street, NW Washington, DC 20433 Telephone. 202-473-1000 Internet: www.worldbank.org E-mail: feedback@worldbank org All rights reserved. 1234050403 The findings, interpretations, and conclusions expressed herein are those of the author(s) and do not necessarily reflect the views of the Board of Executive Directors of the World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of the World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries Rights and Permissions The material in this work is copyrighted Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law The World Bank encourages dissemination of its work and will normally grant permission promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA; telephone- 978-750-8400; fax: 978-750-4470; www copyright.com All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422; e-mail. pubrights(worldbank org ISBN 0-8213-5201-6 Library of Congress Cataloging-in-Publication Data New social policy agendas for Europe and Asia challenges, experience, and lessons/ edited by Katherine Marshall, Olivier Butzbach p. cm. Includes bibliographical references ISBN 0-8213-5201-6 1 Europe-Social policy 2. East Asia-Social policy. 3 Public welfare-Europe. 4. Public welfare-East Asia. 5. Welfare state-Europe 6 Welfare state-East Asia. I Marshall, Katherine, 1947- II Butzbach, Olivier HN373 5 .N496 2002 361.6'1'094-dc2l 2002034340 Contents Foreword, by Mats Karlsson xi Acknowledgments xv Abbreviations and Glossary xvii Introduction: European Social Policy Lessons: Reflections for East Asia 1 Box I: The ASEM Trust Fund, by Katherine Marshall 13 Box II: European Social Poihcy Lessons: 12 Theses, by Ian Gough 15 Part I Frameworks and Models: European and Asian Approaches to Welfare and Social Policy 23 Introduction 25 1 Welfare Regimes in East Asia and Europe Compared, by Ian Gough 27 Box I 1: Social Policy Priorities in Asia: ASEAN Consultations in 2002, byj Tara Karacan 43 2 The Role of Family Obligations in Developed Economies, by Peter Scherer 45 3 How the East Asian Crisis Changed the Social Situation, by Tamar Manuelyan Atinc 55 Box 1.2: Models and Politics for Asian Social Policies, by Franfois Godemont 75 v vi I Contents 4 The East Asian Crisis: Lessons for the Social Policy Agenda, by Katherine Marshall 81 5 Universal Health Insurance in Japan, by Christian Oberliinder 87 Box 1.3: Ten Principles-for Sound Social Polhcy, by Tamar Manuelyan Atinc 101 Box 1.44: Seminar on European Social Policy and the Asia Crisis, Paris, June 27, 2000 105 Part II Social Policy and Macroeconomic Management: Challenges Following the East Asian Crisis 107 Introduction 109 Box 2.1: Social Policy, Financial Crisis, and Macroeconomic Response, by K. S.]3omo 111 Box 2.2: "Globalization Plus". Fostering Truly Integrated Economies, by Robert Wade 113 Box 2.3. The Great Globalization Debate: Good versus Evil.?, by Cohn Moynihan 117 6 Linking Macroeconomic and Social Policy: The Irish Experience, by F. Desmond McCarthy 121 7 The Solidarity Alternative: Lessons from a Norwegian Experience, by Jonas Gahr Store 135 Box 2.4: A Social Agenda in Practice-France, by3_ean-Michel Severino 143 Box 2.5: The "Pact for Italy ". An Integrated Strategy toward Growth and Employment, by Olivier Butzbach 145 8 Measuring the Social Effects of Macroeconomic Policy in Denmark, by Thorkil Casse 149 Box 2 6: Seminar on Social Policy, Poverty Reduction, and the Life Cycle, Amsterdam, October 5-6, 2000 153 Box 2.7: Seminar on Asian Welfare Policy Responses to the Crash of 1997, Bergen, August 16-18, 2001 157 Contents I vii Part IlIl Policies to Combat Social Exclusion 159 Introduction 161 Box 3.1: A Genealogy of Social Capital and New Polcy Implications, by Gareth Api Richards 165 9 Social Housing Policies in Europe and East Asia, by Soren Villadsen 167 10 Public Policies toward the Urban Poor: Recent European Experience, by Vincent Delbos 177 11 Europe and East Asia Confront Growing Migration, by Dieter Obernd6rfcr and Uwe Berndt 187 Box 3.2. Challenges for East Asian Migration Policy, by Dieter Oberndorfer and Uwe Berndt 199 12 Gender Equality and Social Policy in Europe and East Asia, by Ann-Charlotte Stahlberg 203 13 Gender, Demography, and Welfare State Restructuring in Japan, by Ito Peng 215 Box 3 3: Social Exclusion Unit of the United Kingdom Government, by Tara Karacan 235 Box 3 4: The French Health System: Where It Stands and Where It Is Heading, by Louis-Charles Viossat 237 Box 3.5: Seminar on Social Capital, Social Exclusion, and the East Asian Crisis, Manila, November 5-7, 2001 239 Part IV Facing Aging and Providing Social Security 243 Introduction 245 Box 4.1: A Brief Htstory of Social Security in Europe, by Frank A. G. den Butter and Udo Kock 247 14 Social Security, Economic Growth, and Poverty, by Frank A. G. den Butter and Udo Kock 249 15 Social Security and the Social Safety Net in the Republic of Korea, by Meesook Kim 265 viii I Contents 16 Italy's Pension Reform Process: Where Financial and Political Realities Meet, by Mauro Mare and Giuseppe Pennisi 277 17 Social I[nsurance for Aging Rural Households: A Comparative Perspective, by Luis Frota 289 18 Social Policy and the Informal Sector, by Jacques Charmes 307 Box 4.2: Conference on Social Polhcymaking in Europe and East Asia, Roma/Caserta, February 21-23, 2002 315 Part V Labor and the Welfare State 319 Introduction 321 19 A Work-Friendly Welfare State: Lessons from Europe, by Stein Kuhnle, Aksel Hatland, and Sven Hort 325 Box 5 1: Social Protection and Experience from the Great Depression, by Bernard Gazier and Rimy Herrera 345 20 Youth Unemployment: Challenges and Ideas, by Colin Moynihan with Jacqueline Butler 349 Box 5.2: How the United Kingdom Has Addressed Youth Unemployment, by Colin Moynihan 359 21 Worker Participation in Europe and East Asia, by Jacques Rojot 363 22 Public Employment Services in Europe and Asia, by Ruud Dorenbos 377 23 Active Labor Market Policies in the Republic of Korea and Europe, by Bernard Gazier and Remy Herrera 405 24 Evaluation and Monitoring of Active Labor Market Policies, by Christine Erhel 419 25 Employment Policies: The Case of Spain, by Eduardo Gonzalez Biedma 439 Contents I ix Box 5.3: Seminar on Flexibility vs. Security: Social Policy and the Labor Market in Europe and East Asia, Seoul, December 2000 451 Box 5.4: Seminar on Governance and Social Policy, Bangkok, October 26-27, 2000 453 Part VI Moves toward Global Standards 457 Introduction 459 Box 6.1: SocialAgenda for the European Union, by Willem van der Geest 461 Box 6.2: The Open Method of Coordination: A New Venture for European Social Policy, by Luc Tholoniat 463 26 Global Social Standards, by Bruno Simma and Markus Zoeckler 467 Box 6.3: Social Dumping versus Wage Competition, by Frank Westermann 479 27 Global Standards: The European Experience by Gabriella Battaini-Dragoni 481 Box 6.4: The Social Dimension of Regionalism: An Alternative Approach to Globalization?, by Bob Deacon 491 Box 6.5 Seminar on Social Policies to Cope with Economic Instabtlity: Lessons from Europe and East Asia, Brussels, December 6, 2001 495 Concluding Note: Where Next? 497 Appendix A: Welfare Regimes in East Asia, by Ian Gough 499 Appendix B: Social Policies in Europe, by Ian Gough 513 Select Bibliography 517 List of Contributors 527 Index 537 Foreword The human toll of abrupt macroeconomic setbacks is high. It can be measured in specific facts and statistics-people who suddenly lose their jobs, who cannot buy food, who cannot afford to seek medical help, and whose children drop out of school. We know that there are a host of other effects, swirling in concentric circles, that tear the social fabric Times of trouble sometimes can bolster com- munity ties and leadership, but these heartening examples are not the norm. Suffering and erosion of communities are the rule. Some harm is short-lived, forgotten quickly once good times return, but most human shocks leave painful legacies and can set back years of progress. The international community today is fully forewarned about these social effects of crisis. The issues and calls for action to remedy social problems come quickly to the agendas of national and international institutions, and of civil soci- ety organizations, in times of crisis. Some lessons and attendant remedies are very clear-above all, keep social issues at the fore, work to maintain momentum, and do not allow economic setbacks to deflect focus from the challenge of reduc- ing poverty. There is now a body of lessons that are part of a clear international consensus. fight the tendency to cut social expenditures, work to keep children in school, beware the impact of spiraling food and medicine prices, and keep the spotlight on vulnerable groups, especially children. Examples abound of policies and programs-successful and unsuccessful-to address social fallout of crises. Nonetheless, much more work and reflection are still needed to understand fully what happens to people and communities affected by the specific phenomena of today's sweeping economic ups and downs and, above all, what can be done to cushion their inevitable impact xi xii I Foreword The Asia-Europe Meeting (ASEM) Trust Fund has offered a special oppor- tunity to build on and add to this body of international experience. Its combina- tion of practical, timely support for policy design and implementation has been juxtaposed with reflections on "lessons learned" and on implications for future medium-term strategies. This book is designed to make more widely available the products of the spe- cial ASEM Trust Fund Project that aimed to draw upon and explore European social policy experience and make it available for Asia's decisionmakers and ana- lysts. It is framed above all to inform the reflections and actions of East Asian policymakers as they grapple with issues for social policy in the new millennium, though evidently the experience and the approach have a global application. Europe is, of course, both a continent of great diversity and a continent engaged in unprecedented integration and unification. Although there is a dis- tinctive European flavor to policies and institutions, notably the strong public sector and the explicit social contract, the challenges facing the European public and the range of solutions aimed to address them vary widely. A large part of Europe's recent history can be seen in terms of a social agenda-how govern- ments, the European Union, and communities have responded to social forces and challenges. This evolution has produced a rich array of policies, experience, and institutions. It is both a tale of history, as Europe has contended with so many social issues over the decades, and of current challenge, as the Europe of today addresses employment creation, social exclusion, empowerment of communities, the changing role of the family, immigration, and new trends in work and private life. Europe's experience is neither unique nor always exemplary, but the social policy choices in recent European history demonstrate that equity and efficiency need not be enemies of each other, and that a vibrant economy will support and, critically, be supported by strong commitments to social welfare and protection of the vulnerable. The premises for the European Social Policy Lessons Project are four-first, that the common elements in challenges facing European and Asian countries are useful to explore and highlight (for example, with respect to urbanization, aging, and migration); second, that differences, for example, in employment practices and the dynamics of the family unit, also bear rigorous reflection as they can inform policy choices; third, that the common human challenges ahead for Europe and Asia have more similarities than differences, so networks for policy reflection offer scope for mutual learning; and fourth, that Europe's intensive investment in regional policies, standards, and programs offers a mine of infor- mation to other regions, mcluding East Asia. Foreword I xiii The material in the book reflects a means not an end, a process not a product. The project's goal has been to explore how to build networks designed to sup- port policy reform and implementation, across classic discipline lines and geo- graphic regions, and to explore also what kinds of experience and what manner of presentation of such experience could be most useful to the policymakers who seek ideas, sobering lessons, options, and inspiration The interesting material distilled here is designed in the first instance to be useful and relevant to policy- makers. It is designed even more to encourage continuing networks and continu- ing reflection on cross-regional sharing of experience that is inspired and driven by a concern with the centrality of a broad social agenda. Mats Karlsson Vice President External Affairs and UN Affairs Acknowledgments This book is a product of many partnerships, in the very best sense, and we can acknowledge only a small portion of them here. The entire effort owes a special tribute to those who designed and conceived the Asia-Europe Meeting (ASEM) Trust Fund and its European focus, including, among many others, Geoffrey Lamb, David Peretz, Paul Ackroyd, and Alistair MacDonald. Jean-Michel Severino, then vice president of the WN'orld Bank for East Asia and now director general of the Agence Francaise de Developpement, was an inspiration from start to finish. Many others played important roles in the Trust Fund and project, including Keiko Sato, Paula Donovan, and Jannes Hutagalung. Warm thanks go to Louis-Charles Viossat, who was critical to the design and launch of the proj- ect. In the final stages, Tara Karacan played a pivotal role in organizing the diverse group of actors to move toward production and ensure high-quality results. We acknowledge the major role of our superb editor, Sandra Hackman, and support also from Min Gyo Koo, Manuel Rossini, Kathrin Plangemann, Sonya Woo, Julianne Turner, Rebecca Ling, Mary Imamura, and Monika Ploch at different project stages. Among the dedicated partners who should receive special tribute are Helene de Largentaye and the team of the Conseil d'Analyse Economique in Paris, as well as Jean-Christophe Bas and the organizing teams for the ABCDE (Annual Bank Conference on Development Economics) conferences in Paris and Oslo; Jacques van der Gaag, Wim Driehuis, and the Amsterdam Institute of Inter- national Development; Mitsuko Horiuchi, Moazam Mahmood, and the Asia Pacific Office of the International Labor Organization in Bangkok; Hyekyung Lee and the Korea Social Security Association in Seoul; Sven Hort and Stein Kuhnle in Bergen; Anita Kelles-Viitanen, Isabel Ortiz, and the Asian Development xv xvi I Acknowledgments Bank in Manila; Willem van der Geest and the European Institute for Asian Studies in Brussels; and Giuseppe Pennisi and Italy's National Institute for Public Administration m Rome and Caserta. We also thank several individuals who have taken a special interest in this effort from beginning to end They include Nicholas Barr and Howard Glennerster (London School of Economics), Amar Siamwalla (Thailand Development Research Institute), Keith Bezanson (Institute of Development Studies, University of Sussex), Patrick Guillamont (Centre d'Etudes et de Recherches sur le Developpement International), Alice Sindzingre (Centre National de la Recherche Scientifique), Manfred Bardeleben and Roland Feicht (Friedrich Ebert Stiftung), and Peter Scherer and Peter Whiteford (Organization for Economic Cooperation and Development). Abbreviations and Glossary ADB-Asian Development Bank AF-Arbeidsformidlingen AIID-Amsterdam Institute for International Development AIS-information system of the Swedish Labor Market Administration APEC-Asia Pacific Economic Cooperation organization APIS-Annual Poverty Indicators Survey ASABRI-Indonesian Armed Forces Social Insurance Plan ASEAN-Association of Southeast Asian Nations ASEF-Asia Europe Foundation ASEM-Asia-Europe Meeting ASKES-Asuransi Kesehatan Pegawai Negeri (means-tested free public health center service for civil servants and military) BA-Bundesanstalt fur Arbeit CAP-Common Agricultural Policy CEREQ-French Research Center on Employment and Qualifications CGIL-Confederazione Generale Italiana dei Lavatori CHWE-Council on Health and Welfare for the Elderly CWA-Civil Works Administration DAC-Development Assistance Committee DB-defined-benefit DC-defined-contribution DPJ-Democratic Party of Japan DREAM-Danish Rational Economic Agents Model EC-European Commission xvii xviii I Abbreviations and Glossary EEA-European Economic Area EPF-Employee Provident Fund ES-employment services ESCAP-Economic and Social Commission for Asia and the Pacific ESF-European Social Fund EU-European Union FDI-foreign direct investment FERA-Federal Emergency Relief Administration FIES-Family Income and Expenditure Survey GDI-Gender-Related Development Index GDP-gross domestic product GDR-former German Democratic Republic GNP-gross national product HDI-Human Development Index HPI-Human Poverty Index ICESCR-International Covenant on Economic, Social and Cultural Rights IEFP-Instituto do Empreso e Forma,ao Profissional IFLS-Indonesia Family Life Survey IFS-International Financial Statistics ILO-International Labor Organization IMF-International Monetary Fund JAMSOSTEK-Jaminan Sosial Tenaga Kerja (means-tested free public health center service for private sector) JMA-Japan Medical Association JSP-Japan Socialist Party KLI-Korean Labor Institute LDP-Liberal Democratic Party LES-Local Employment Service LO-Norwegian Confederation of Trade Unions MERCOSUR-Latin America Common Market MISEP-Mutual Information System on Employment Policies MSE-micro- and small enterprises MVO-market value of the output NAFTA-North American Free Trade Agreement NDYP-New Deal for Young People NGO-nongovernmental organization NHI-national health insurance NHS-national health services Abbreviations and Glossary I xix NPS-national pension scheme OAED-Labor Force Employment Organization ODA-official development assistance OECD-Organization for Economic Cooperation and Development ONEM-Office National de l'Emploi OPK-Operasi Pasar Khosus (Indonesian rice subsidy program) PES-public employment services RESEARCH- European Employment Observatory advisory group SADC-Southern Africa Development Community Schengen Treaty-Signed at Schengen, Luxembourg, March 1995, to end internal border checkpoints and controls, together with enhanced visa controls at the external borders through a common approach to visas and asylum requests. There are 15 signatory countries at present. SCI-Sezione Circoscrizionale per l'Impiego SEU-Social Exclusion Unit SMERU-Social Monitoring and Emergency Response Unit SUSENAS-Social and Economic Survey of Indonesian Households SYSDEM-System of Documentation, Evaluation, and Monitoring of Employment Policies TASPEN-Tabungan dan Asuransi Pegawi Negerl (Government Civilian Employees' Savings and Insurance Scheme) TLM-transitional labor market TUC-French public temporary employment scheme UDHR-Universal Declaration of Human Rights UGT-General Worker's Union UN-United Nations UNDP-United Nations Development Programme UNEDIC-French unemployment insurance, run by social partners VTML-Valtion Tyomarkkinalaitos WCIAS-Women's Committee for the Improvement of the Aging Society WPA-Works Progress Administration WTO-World Trade Organization ZUS-Polish state-owned Social Security Office INTRODUCTION European Social Policy Lessons: Reflections for East Asia East Asia is at a crucial stage in its development. Its impressive and rapid eco- nomic and social achievements are unparalleled elsewhere in the world. No other region can boast comparable performance in fostering economic growth; reduc- ing poverty; and improving education, health, and nutrition. East Asian countries have, with few exceptions, shared the fruits of 30 years of economic growth equi- tably within their societies, leading to steadily lower levels of absolute poverty. Partly because of this success, the severity of the economic turmoil that spread after 1997 stunned governments throughout the region. The crisis revealed the devastating vulnerability of many households-especially poor households-to a sudden downturn in macroeconomic fortunes. During the early months of the crisis, the attention inevitably focused on resolving critical problems such as rising unemployment and poverty, mass bankruptcies, and breakdowns in the financial system. However, the need to for- mulate sustainable policy responses soon forced East Asian countries to reflect on longer-term lessons from the crisis, as it shook important assumptions regarding development for both individual countries and the region. This challenge was particularly strong in the field of social policy, as East Asian coun- tries had long essentially relied on economic growth to provide social safety nets, which proved frail and patchy. The crisis prompted a recognition that, with rapid modernization, demo- graphic change, and exposure to global economic and social pressures, the region needed new models for social policy. Governments, trade unions, civil society organizations, and international partners alike embarked on a dynamic process of reflection on the future social agenda and the institutions that would move it forward. 2 1 New Social Policy Agendas for Europe and Asia Challenges, Experience, and Lessons This book is a product of that process of reflection. It is the culmination of an endeavor supported by European and East Asian nations to encourage new think- ing about social policy for East Asia in the shadow of the 1997-98 socioeconomic crisis. The book reflects the guiding objectives of the Asia-Europe Meeting (ASEM) Trust Fund, an informal process of cooperation and dialogue (see Box I and ), and also a specific ASEM-supported project to explore European experience in social pol- icy that might prove useful for postcrisis East Asian countries. ASEM and the ASEM Trust Fund The ASEM process itself has deep roots and an evolving role as a Europe-Asia governmental initiative. Launched in the mid-1990s,1 its central aim has been to strengthen ties between East Asia and Europe as global pressures highlighted the need for both policy exchange and concerted action to address human needs. Heads of state and ministers from both regions have been meeting periodically (with a summit every 2 years) to discuss political, economic, and cultural issues of mutual interest The process has expanded to encompass a civil society forum and ongoing institutions, prominent among them being the Asia Europe Foundation (ASEF) in Singapore. A special context surrounded the second formal meeting of ASEM heads of state in London in April 1998 East Asia's financial and economic crisis was reaching a peak, and the region's governments faced new issues, new demands, and new pressures. Leaders at the London meeting decided to respond by creat- ing the ASEM Crisis Response Trust Fund with contributions of about US$45 million, which would finance a series of projects in three priority areas: social policy, financial restructuring, and corporate restructuring. The ASEM Trust Fund directed the projects to support not only overall reform in the seven most critically affected countries but also analytical and policy work to encourage con- tinued rebuilding. The ASEM Trust Fund represents a remarkable story in development assis- tance; it has supported the launch and completion of more than 80 separate ini- tiatives. Nearly half addressed issues of social policy, generally taking the form of technical assistance projects targeted at specific country challenges and popula- tions. A recent independent evaluation highlights the fact that the ASEM- financed projects played a catalytic role in the postcrisis period for East Asia. European Social Policy Lessons Reflections for East Asia 1 3 The specific project that is the genesis of this book-Beyond the East Asian Socio-Economic Crisis: Lessons Towards the New Social Policy Agenda-was designed, approved, and implemented within the ASEM Trust Fund context The project focused on the countries most affected by the crisis: Indonesia, the Republic of Korea, Malaysia, the Philippines, and Thailand, or the "East Asia five." It was approved in mid-1999 and executed from 2000 to 2002 under the direction of Katherine Marshall. The project sought to cull and disseminate meaningful social policy lessons from East Asia and from Europe, where the wel- fare state is over a century old. A secondary goal was to establish intellectual networks and a closer dialogue between European and East Asian experts and policymakers on social pohcy issues arising from the crisis Vehicles for achieving these goals included a series of semi- nars among European and East Asian policymakers and specialists on specific social policy issues, and some 30 papers to bolster that discussion and explore issues in greater depth. Participants at the seminars discussed most of these papers. A project Web site acted as a resource reaching far beyond the network of experts and policymakers directly involved in the project. The site includes the papers gen- erated through the project, summary notes from the seminars, a calendar of events within the project, and links to related sites and items of interest to the Social Lessons network (see ). Although the project produced tangible outputs, the less tangible outputs were perhaps the most important. Those included personal contacts and friend- ships between European and East Asian participants and a host of informal net- works, with countless examples of follow-up collaboration. The ASEM partnership entrusted the management of the ASEM Trust Fund to the World Bank, to ensure that the activities under the partnership would link tightly to the overall international effort to support East Asia's recovery, in which the World Bank played a leadership role. Some participants expressed doubt as to whether the Bank would fully embrace the spirit of the ASEM Trust Fund, so it was incumbent on the Bank to preserve this spirit. The European Social Policy Lessons Project was a principal vehicle toward that end. The Social Agenda It is worth emphasizing that no government organizes social policy as a single institution or program it is simply much too broad. Social policy encompasses 4 1 New Social Policy Agendas for Europe and Asia Challenges, Experience, and Lessons labor, welfare, health, and education policy; support for older people; gender relations; social exclusion; and the welfare of the poorest communities The point of departure for this book was, nonetheless, the concept of a "social agenda"- essentially the full range of issues that focus on the human dimensions of public policy. A central reason for this focus is the strength of the links between these issues. This social agenda extends beyond common notions of poverty alleviation, social protection, and social development, which are important organizational concepts within the World Bank, for example. The concept of a social agenda is well understood in Europe. As East Asia moves beyond the urgent imperative of raising the standard of living for large masses of people in dire poverty, it con- fronts the familiar challenges of maintaining a broad focus and stimulating dynamic and creative thinking on social policy issues that affect virtually every resident Neither the project nor the book purports to cover this agenda comprehen- sively. Some issues are almost entirely absent-most prominently, education and health policy. Indeed, throughout the course of the project, ideas for exploratory work abounded that available time and resources could not address. Examples include training social policy analysts and leaders, and rethinking drug policies. The East Asian Focus This book is for East Asia but not essentially about that region East Asia enters the discussion at many junctures as the authors apply their analysis to that region, but the book does not purport to analyze exactly where East Asia stands or, more important, exactly wvhat it should do next. Why Europe? A frequent and logical question is, Why European lessons? ASEM leaders and those involved in the ASEM Trust Fund perceived European experience in the social policy realm as deserving special exploration. The welfare state is one of the core characteristics of modern European history and of Europe's current political economies, and European policymakers and institutions have engaged in precisely the kind of building and reforming of welfare systems that East Asian countries are now pursuing European Social Policy Lessons Reflections for East Asia 1 5 Europe is far from homogeneous, of course, which makes for diverse social policy lessons. Indeed, Europe offers a natural and well-studied laboratory for varying social policy responses to broadly similar social problems, and it includes several distinct "welfare regimes."2 Yet despite the vital differences within Europe, the European Union is dis- tinct from other OECD regions, notably the United States, as the role of the state in social welfare is notably greater and its form qualitatively different. The European model of welfare is characterized by high levels of spending, insur- ance-based social programs, marked intergenerational solidarity, and modest-to- high redistribution of resources. In a majority of states, a "breadwinner" model provides considerable employment protection and benefits for the core work force, and substantial investment in human and social infrastructure has achieved moderate to low levels of poverty and inequality.3 ASEM participants built on several hypotheses about this experience.4 First, Europe differs in important respects from other regions. As a very broad approximation, the Americas (North and South) tend to put more weight on individualism (including voluntary charity), whereas Europe emphasizes social solidarity. Second, Europe offers a host of examples. There is no "European model" but many different examples of what might loosely be termed a European approach. Third, European experience makes a critical distinction between structure and scale "Structure" refers to whether activities designed to enhance human well-being are publicly funded or produced, or largely or wholly private "Scale" refers to the level of public spending on the welfare state. Fourth, the European welfare state has not failed. Despite its mixed reputa- tion, it is alive and vibrant and engaged in a dynamic process of adapting to new realities Some countries may have made promises that now seem too generous given demographic trends, but some policymakers have responded by making social guarantees less generous-the reform of the Swedish pension system being an important example.5 Our hope is that the chapters in this book will provide the basis for learning and comparing without conveying linear "lessons" from Europe. There are not any-far from it. Europe's welfare states were created over several decades and have gone through several stages. Countries that face challenges today will need to reflect on both the systems and the dynamics that have shaped them. What is more, despite the deliberate, pointed effort to focus this work on Europe, the project was multidirectional in practice, as the process of reflection led us to far-flung regions of the world. Chapter 5, in fact, shows how an idea and 6 1 New Social Policy Agendas for Europe and Asia Challenges, Experience, and Lessons experience can take root in different societies at different times and in different forms (the example is the evolution of Japanese health systems). A key maxim was that the inquiry would offer a balanced exploration across nations and disciplines into experiences often known only secondhand and in a fragmentary, even distorted fashion. Delving into these rich experiences opened doors for all participants, from whatever sector, country, or continent, to ques- tion, learn, explore, and venture into new ideas. The most remarkable feature of the process was the dynamic interchange it spawned on the issues that confront all societies at the start of the 21st century. The Seminar Program A major feature of the European Social Policy Lessons Project was the series of eight formal seminars and several less formal meetings. Half were set in East Asia and half in Europe, with participants also roughly 50-50 from the two regions. The seminars constituted the heart of the project, allowing participants to pur- sue new insights and to build partnerships whose scope extended well beyond the project agenda itself. Many participants contributed fresh perspectives reach- ing beyond their traditional fields of expertise. The seminars are summarized in short boxes for this volume. Analysts often reveal marked tendencies to separate thinking from action, decisionmaking from implementation, and basic research from policy studies. These divides exist at the international level and within countries. Projects such as this one that aim to amass and refine knowledge run the risk of worsening such segmentation. Thus the project made an explicit effort to encourage and bring together academics and policymakers, unionists and government officials, repre- sentatives of businesses and the media, and people working in the field. Every meeting was designed to blend policymaking and expertise, different points of view, and different experiences. In most cases this produced challenging discus- sions. We identified partners and participants progressively over the life of the proj- ect, allowing us to tap different policy arenas and institutions. Some participants were present at several meetings, others at only one; the group changed at each encounter Partners included the Asian Development Bank, the International Labor Organization (ILO), the Amsterdam Institute for International Develop- ment, the Economic Analysis Council in France, the Korean Social Security Association, the University of Bergen in Norway, the European Institute for European Social Policy Lessons Reflections for East Asia 1 7 Asian Studies in Belgium, and Italy's Public Administration Institute. Each meeting was organized with a different partner. Participants often referred to strong crosscutting themes. One such theme was the need to strike the right balance and coordination between macroeconomic and social policy. Participants from developing and industrial countries alike expressed frustration that countries tend to put macroeconomic policy at the top of the policy hierarchy. Social policy ministers must too often work with choices already determined by their colleagues in charge of finance. Integrating the two arenas is essential to good policy, especially good develop- ment policy, as macroeconomic and fiscal policy strongly affect social welfare, whereas social policy exerts a considerable impact-positive and negative-on public finances and overall economic performance. However, some participants argued that the gap within their countries between finance and economics, and health, education, and social protection is more significant than the supposed gap that separates European and East Asian social policy. Only in times of crisis do the urgency, the breadth, and the links between social and economic issues force policymakers to try to keep all elements in bal- ance. It is not by chance that European welfare states took their modern shapes in such a context, after World War II. The crisis that hit East Asia 5 years ago represented a similar window of opportunity. Many perceive a risk that with a return to more normal growth patterns and less urgency, the traditional divides between macroeconomic policy and all other policy fields could return, and momentum could be lost. Participants also pointed to hopeful exceptions to this rule. Some stressed the traditional importance of social policy and welfare in European government agendas, not least because of the strong political weight of unions and social- democratic and agrarian parties. Some critics have questioned the welfare state as a pillar of the post-World War II social compact, especially in the 1970s and 1980s, when they blamed Keynesian policies-characterized by high levels of public investment, redistribution, and social protection-for stagnation and unemployment. But since the mid-1990s Europe has seen a revival of the wel- fare state, reflecting the central position of social issues on the political agenda, including at the level of the European Union (witness the focus on employment and social protection in the proceedings of the most recent European Councils of Ministers). Participants also saw hopeful signs that social issues would emerge from their traditional political marginalization in East Asia In many countries an active public debate on social policy challenges and options has continued since the 8 1 New Social Policy Agendas for Europe and Asia Challenges, Experience, and Lessons crisis, although some participants doubted that the window of opportunity will remain open. There was clear consensus about the urgent need to improve social protection to avoid similar disasters in future crises A related issue was the need to integrate social polhcy with other polhcy instru- ments. Several seminars (at Brussels and Bangkok) stressed the positive links between social policy and economic growth (the former supports demand and ensures the transferability of skills, for instance). Participants also highlighted links between social policy and poverty (the Amsterdam meeting) and social pol- icy and industrial policy (Bangkok). Participants in the Seoul seminar focused on the need for complementarity withm the social policy realm, such as among employment policies, pension reform, and social security A third crosscutting issue was the regional dimension of social polhcy. European experience highlights a gradual intensification of regional coordination in several policy fields, including immigration policy, labor standards, and employment strategies. Participants in several meetings cited regional trading blocks as the natural place to build regional frameworks for social policy But how to start? Several participants emphasized the rights-based approach as embraced by the ILO and the Council of Europe. This is still controversial, however, and the Rome/Caserta seminar shed light on an alternative and innovative way to coor- dinate regional social policy: through peer review and spreading of best practice exemplified by the "open method of coordination" launched by the European Union.6 The role of social capital as both a goal and an instrument of social policy was the specific focus of the Manila meeting. One of the key objectives of social policy is to empower citizens and social groups rather than simply treating them as "objects of care." This approach underpins many of the recent welfare reforms in Europe, which put more emphasis on individual responsibility and initiative. However, fostering social capital as a force for change presents complex chal- lenges and may exacerbate social tensions if it accentuates divisions within soci- eties and shifts the balance of power among groups The political nature of social policy is also important Participants often stressed social policy as the outcome of political struggle, underscoring the importance of social dialogue and the core need to provide a "level playing field." Social dialogue requires the presence of strong actors such as trade unions, the state, business, and political parties. A vital new issue, addressed in several semi- nars, notably Bangkok and Rome/Caserta, is the role of public information and the media in creating social policies This issue merits much more detailed explo- ration. European Social Policy Lessons Reflections for East Asia l 9 The importance of gender in policymaking (hearteningly) recurred often dur- ing the seminars. Gender relations have emerged as a vital engine of social change-witness the impact of growing female participation in the labor force in most countries in Europe and East Asia Gender is also one of the most prob- lematic areas of social policy, given the heavier burden that women bear during crises and given the "social risks" of maternity, unemployment, and disease. Virtually every discussion, from Paris to Bangkok and Manila, raised the point that policymaking forums too seldom include women's voices, with the result that social policies do not fully take into account gender relations. For example, because policymakers have often designed welfare schemes, especially in Europe, during times when the sole-breadwinner household prevails, changes in gender relations can mean that this traditional design runs the risk of perpetuating exist- ing inequalities instead of correcting them. Social Policy Lessons? Against the backdrop of a project that focused on process more than on specific recipes-that eschewed formulas for successful social policy and celebrated the diversity and dynamism of experience-some core ideas nonetheless emerged time and time again They form the structure and argument of the book A first lesson is the need to find effective ways to link social and economic practices. Successful country examples from Europe offer insight into the politi- cal drive and specific mechanisms (such as social pacts) that encourage effective integration of social and macroeconomic policies. A second insight is the need to incorporate exceptional dynamism into social policy. A corollary is that decisions that affect the welfare of millions prove exceptionally difficult to change. Social policy decisions and designs often have very long lead times and application periods. The experience of many countries shows that designs well suited to certain socioeconomic and political configura- tions may well no longer work as social demands, individual behavior, and demo- graphics change. In Europe, for instance, unemployment insurance designed for a period of high employment and short length of joblessness spells cannot cope with high unemployment levels and joblessness that lasts more than a year. Another classic example is the fate of pension systems conceived in the baby- boom era, which face fundamental problems as societies struggle to cope with aging populations. These outcomes call for building "permanent learning" mechanisms into social security systems at the design stage Such flexibility can 10 I New Social Policy Agendas for Europe and Asia Challenges, Experience, and Lessons take the shape, for instance, of mechanisms to evaluate the employability of indi- viduals and provide further training, which some European countries are begin- ning to include in their mechanisms to address unemployment. A final insight turns on the complexity and interrelationships within social policy. Examples abound of unexpected consequences of social policy. Many case studies that emerged from the project highlighted the links between pensions and health issues, and between social policies and gender relations Box II reflects a set of hypotheses, articulated by Ian Gough, about social pol- icy issues and lessons as they emerge from the juxtaposed analysis of European and East Asian experience. The Content and Structure of the Book This book includes many but not all papers discussed at the project seminars. The papers are abridged to make them more accessible; full versions remain on the project Web site, together with full proceedings of seminars and some papers that could not be included in this volume. The book's coverage is exploratory and illustrative, and it does not aspire to investigate fully the rich array of European experience or its possible implica- tions for East Asia. The book is designed to whet the appetite but not to satisfy it. Each section and chapter aims to highlight references and experiences for fur- ther exploration The book's structure follows the basic narrative of the project's trajectory. Part I provides a framework for the analysis in the rest of the book, offering back- ground on how Europe defines welfare regimes, juxtaposed with an illustration of how reality may differ from image in the classic issue of the family as a core element of social policy. Two chapters provide background on the East Asia cri- sis to ground further discussion The final chapter describes the evolution of Japan's health insurance system, with its complex origin, borrowing, adapting, and ultimately succeeding. Part II tackles links between macroeconomic and social policies through case studies of Ireland and Norway and a Danish planning tool. The book does not delve into history and debate on macroeconomic policy, often deeply influenced by the East Asia crisis, but project participants do summarize arguments about the implications of that crisis for future policy, especially in the social arena Part III turns to the complex issue of social exclusion, a topic woven through- out the book from start to finish. Poverty is clearly one form of social exclusion, but the challenges are much deeper and wider, and present themselves differ- European Social Policy Lessons Reflections for East Asia I 11 ently in different societies. The chapters cover a range (though far from compre- hensive) of current issues, such as gender relations, migration, and policies toward poor citizens in contemporary Europe. Part IV focuses on a central challenge facing Europe today (as well as many East Asian countries). the phenomenon of aging societies and welfare systems. Strong patterns and challenges emerge in countries as diverse as Italy and Scandinavia; also at the heart of the discussion is the vital role of political econ- omy in addressing them. Part V highlights the wealth of experience on what many consider the "end and the beginning" of social policy-labor policies. The chapters include case studies from different countries on topics ranging from worker participation to public employment services. The section also addresses the important role of the European Union from different angles. Part VI follows up this orientation with chapters on the complex topic of set- ting international standards for social policy. The most concrete example is the Council of Europe, which stands out in monitoring and moving toward enforc- ing social standards. This section also treats the broader issue of human rights, now very much at center stage in global debates on labor standards, the role of children, and the right to minimum health care and education. The discussion is set against the background of the growing international consensus about the urgency of fighting poverty reflected in the Millennium Development Goals that were the focus, for example, of the March 2002 United Nations Summit on Financing Development held in Monterrey, Mexico This book represents both a record and stock-taking. Its central purpose is to encourage continued exploration and even stronger networks. Notes I See for the ASEM homepage of the opening event in Bangkok, Thailand 2 See Chapter I for further details on the different regimes 3 Ian Gough, "Social Aspects of the European Model and Its Social Consequences," in The Social Quality, of Europe, ed Wolfgang Beck (The Hague Kluwer, 1997) 4 This framework owes a particular debt to Nicholas Barr, professor at the London School of Economics, who in discussions and notes helped participants explore these hypotheses 5 A more sophisticated set of arguments can be found in A. B Atkinson, The Economic Consequences of Rolling Back the Welfare State (Cambridge, Mass MIT Press, 1999) 6 See Box 6 2 BOXI The ASEM Trust Fund by Katherine Marshall The ASEM Trust Fund for the Asian Financial Crisis was created at the initiative of leaders of the Asia-Europe Meeting (ASEM) at the ASEM summit in London in April 1998. The fund was launched in June 1998 with the signing of the first letter of agreement between the United Kingdom and the World Bank. Ten partners contributed a total US$45 million. The fund's purpose was to facilitate recovery in countries affected by the Asian economic crisis and to support reform programs by providing technical assistance and advice to address immediate policy issues in the financial, corporate, and social sectors. East Asian countries participating in the ASEM process could draw on the fund if they were eligible to borrow from the World Bank or the International Development Association and they were in crisis or vulner- able to financial or economic shocks or contagion; thus China, Indonesia, the Republic of Korea, Malaysia, the Philippines, Thailand, and Vietnam were eligible. The fund was not viewed as an injection of capital but rather as a vehicle to provide expertise, especially from Europe. It formed part of the broader ASEM effort to enhance relations between Europe and East Asia. As the fund administrator, the World Bank established a special gov- erning structure to ensure that all activities and programs met high stan- dards commensurate with agreed country and sector strategies, follow- ing the World Bank's operating procedures. An ASEM committee within the World Bank oversaw its administration. The ASEM Trust Fund agreements provided that ASEM members meet with the World Bank every 6 months to review country strategies and progress reports, and that the ASEM donors approve specific projcct proposals over US$1 million. The letters of agreement between the World Bank and donors pro- vided that: continued on next page 14 I New Social Policy Agendas for Europe and Asia Challenges, Experience, and Lessons BOX 1, continued * Activities supported by the ASEM Trust Fund were to address two key priorities, with roughly 50 percent of funding for each: - Technical assistance and training to governments for policy and institutional reform in the financial and corporate sectors - Technical assistance and training to governments to design interventions to mitigate adverse social effects of the crisis and reorient poverty reduction efforts to meet the new social agenda * ASEM Trust Fund was to fund consultants and experts engaged by the World Bank, and grants to host governments, nongovern- mental organizations, educational entities, and other entities to finance technical assistance, training, and related costs, for activi- ties supported by the fund * At least 50 percent of ASEM Trust Fund allocations were to go toward activities in lower-income and lower-middle-income countries * The Trust Fund closed on December 31, 2002. In March 2002, an independent evaluation of ASEM was conducted by European Commission consultants; their findings and recommenda- tions were discussed in a workshop in Bangkok in March 2002 An ASEM II Trust Fund was launched at the ASEM leaders' Third Summit in October 2000 and endorsed at the Third ASEM Finance Ministers' Meeting in Kobe in January 2001. Activities supported by the ASEM II Trust Fund will address two key priorities. (a) providing technical assistance and training to governments for sustainable reform of their financial and corporate sector policies and institutions, and (b) providing technical assistance and training to governments to design and implement interventions to reorient sustainable poverty reduction efforts to meet evolving country requirements. Total contributions amount to about US$30 million; the expected closing date is December 31, 2005 Further information about ASEM Trust Fund may be found at . See also the European Union Web site on ASEM at . BOX 11 European Social Policy Lessons: 12 Theses by Ian Gough Europe's modern welfare state, at least the "social insurance state," is well over a century old, if the introduction of Bismarck's sickness funds in 1883 is treated as its foundation. What lessons might this experience offer for the developing world? Social welfare policies must be seen in context, taking careful account of existing patterns of social provision, institutional responsibilities, and the interests that these express and perpetuate. Policy learning, transfer, and change can occur, and universal principles should apply to policy goals. However, universal policy designs should not be sought and any social policy idea must account for political economies and inherited institutions. In that spirit, I offer 12 lessons, with both negative warnings and positive role models. The 12 Theses 1. From "strong" to "weak." Social insurance was the dominant form of social protection in all European countries from the start (even in the United Kingdom after 1925). It began by covering state workers and manual workers in large industrial firms, gradually rippling out to include employees in medium and small enterprises; agricultural, white- collar, and service workers; the self-employed; and later, in some coun- tries, the unemployed and homemakers. Thus the social insurance state proceeded "from the strong to the weak"2 or, more accurately, from groups central to an industrial economy to those nearer the periphery. Social insurance offers a transitional strategy to universal coverage that starts with those employed in the formal economy. It is the opposite of today's dominant "target the poor" approach. 2. Industrialization. Social policies in Europe developed in societies that were industrializing rapidly and came to fruition during the mass rural exodus following World War II (except in the United Kingdom, continued on next page 16 New Social Policy Agendas for Europe and Asia Challenges, Experience, and Lessons BOX 11, continued where the shift from agriculture to industry took place a century earlier). These countries initially followed an "extensive" rather than an "inten- sive" growth path; they grew through expansion of capital and the labor force rather than through increasing productivity.3 Labor accepted a cap- italist economy in exchange for capital's tolerance of collective represen- tation and bargaining, social services, and social protection. This welfare system was ill suited to deindustrialization and postindustrial capitalism from the 1970s onward. 3. Civil society and labor movements. Proletarian struggles, trade unions, and socialist parties formed a backdrop to the emerging welfare state throughout Europe. In many northern European countries, these developments were linked to a range of other class mobilizations, notably by agricultural workers and later the "service class", other social move- ments, such as the temperance movement in Scandinavia; and self-help institutions, such as Friendly Societies in the United Kingdom The imprint of these movements on social policies differed according to dom- inant class coalitions, the orientation of business and the propertied classes, and the form of the national state. Thus there have been "etatist" (France) and "corporatist" (Germany and Italy) routes to social policy. Fears of social unrest and breakdown provided a constant backdrop. 4. Crowding out versus crowding in. The growing state role crowded out some prior forms of social provision and protection, such as friendly soci- eties and private hospitals, but "crowding in" has also occurred This is evident in the state role as a supplier of resources for the third, or non- profit, sector in every modern European country Civil society social movements have served to strengthen the voice and claims of the poor. 5. The productive welfare state. Originally introduced in Sweden in the 1930s, this term recognizes the contribution to modernization and pros- perity of high-quality and equitable education, health care, family plan- ning, and other family strengthening policies. This notion has undergone a recent resurgence as human capital assumes even greater importance in the postindustrial economy. Thus social policies have a "public goods" role; they were never solely about redistribution. Innovation in social policy often reflects "reforms from above" by progressive elites who rec- ognize the collectivc benefits of managed capitalism. European Social Policy Lessons 12 Theses I 17 BOX II, continued 6. Open economies and social protection The inverse statistical relation- ship between extent of trade and levels of spending on social protection in Organization for Economic Cooperation and Development (OECD) countries amounts almost to a law A stark argument holds that social protection is the only alternative to trade protection if societies are to avoid social disintegration, implying that a globalized world "requires" more social protection.5 For such systems to develop in practice, though, preconditions need to be in place: pressure from below and reforms from above 7. The family and the houtsehold economy The family and household still play a critical role in providing care and managing security, even in Nordic countries, whose governments have developed the most deliber- ate strategies to cope with modern changes in household structure. In Japan and southern Europe, the extended family and cohabitation across generations are declining much more slowly. In these countries-and in countries such as the Republic of Korea and Taiwan, China-"grand- mother welfare" may have bought respite for one generation from the contradictions between women working and providing childcare. Even in such countries, the capacity of the family to provide protection against insecurity and invest in human capital is limited unless income distribu- tion is equitable or labor and social mobility ensure that most families have at least one member in the protected formal sector (the garantismo). 8. Labor markets and welfare states Postwar Keynesian welfare states were founded on extensive employment opportunities and complemen- tarity between labor markets and welfare systems Employment systems supplied incomes and taxes, whereas welfare systems moderated the effects of the trade cycle and employed men and growing numbers of women in public services. Many now see contradictions between the two. Greater mobility and structural power of capital forces greater flexibility and insecurity on the labor market, whereas high taxes and labor market regulation impede employment. This is most evident in the conservative regimes of continental Europe. Some countries with social democratic welfare regimes, such as Denmark and the Netherlands, appear to be continued on next page 18 I New Social Policy Agendas for Europe and Asia Challenges, Experience, and Lessons BOX 11, continued overcoming these challenges through a combination of social bargaining, mild labor market regulation, and generous universal support services, a "win-win" scenario dubbed "flexicurity." (The United Kingdom and Ireland have had recent success in overcoming mass unemployment but poverty and insecurity remain high.) The lesson is to favor education, health, and family services above cash transfers. 9. Pensions. Classic social insurance, pay-as-you-go pensions, face a crisis of unsustainability. Reasons include falling contributor-pensioner ratios stemming from demographic change, high nonwage labor costs, early retirement, and the shift from an extensive to an intensive growth path, coupled with a rising profit share that limits wage growth The hardening of political support for the "breadwinner" model of social insurance, which impedes pension reform, is another critical factor, though Europe also offers examples of successful incremental pension reforms. 10. From safety nets to activation policies Europe's extensive means- tested social assistance has been attacked for creating disincentives and moral hazards. Countries across Europe now generally complement such assistance with a variety of policies designed to activate the labor force These policies, which differ from U.S.-style "workfare," offer several policy lessons: successful activation efforts are expensive, they require substantial administrative capacity, and they may redistribute employ- ment opportunities more than they create new jobs 6 11 Health care. European countries have all developed universal, publicly guaranteed rights to health care, largely because health care consumers organized before private health care providers did.7 Countries broadly divide into those with national health insurance (NHI) and those with national health services (NHS). Beginning in the 1970s several southern European countries switched from NHI to NHS, partly in recognition of the macro-inefficiency of NHI. Public insurance schemes, when coupled with fee-for-service payment to providers, generate moral hazard and cost inflation. Compared with NHI and private medical insurance, tax-financed national health serv- ices are superior in delivering standardized medical treatments within fixed budgets. European Social Policy Lessons: 12 Theses 1 19 BOX 11, continued 12. Universal citizenship-based services. Universal citizenship-based services and benefits offer several advantages over both social insurance and social assistance. They are administratively simple: a child benefit and an elder pension require only a birth certificate or other proof of identity and age. They are "minimally presumptuous,"8 making the fewest assumptions about people's income, pattern of activities, living arrangements, and livelihood strategies. And they are thus well adapted to postindustrial breakdowns in families and labor markets. Such systems appear to be one basis for the Nordic countries' economic revival, and the United Kingdom's decision to raise universal child benefits substan- tially follows such reasoning. These arguments apply still more in devel- oping countries with large informal sectors and complex livelihood strategies. A general and obvious overall conclusion is that all this costs money. The weight of the public sector must and should grow as economies modern- ize, so fiscal reform moves to center stage. This seems to swim against strong tides in East Asia, but universal education, health care, and citi- zenship transfers could in time rebuild middle-class support for new taxes. Implications for East Asia There are new threats to the revenue base of all states, especially in East Asia,9 including growing competition for taxes with the mobility of capital and skilled labor; for fees imposed on off-revenue services such as airports; for middle-class spending; and for more opportunities to evade taxes through e-commerce. Downward pressure on import duties from the Association of Southeast Asian Nations (ASEAN) Free Trade Area and weakening of pro-tax, pro-spending political coalitions through middle-class exit also undercut government revenue. Mean- while, other factors are boosting demand for public spending, notably continied on next page 20 New Social Policy Agendas for Europe and Asia Challenges, Experience, and Lessons BOX 11, continued Asia's high postcrisis debt burden, and rising insecurity and aging. Helpful European experience includes the importance of value added, consumption, income, and capital gains taxes; an emphasis on social security contributions to fund transfer expenditures, and the benefits of harmonizing regional taxes. These experiences show that welfare states can be competitive if they invest in people and communities and avoid large-scale, unproductive transfers.1O Productive social policies form the foundation for high- quality, intensive economic development East Asia's welfare regimes embody some of these attributes, notably substantial public investment in basic education and health; they can and should build on this legacy. An important avenue could be to extend universal citizenship bene- fits to secondary education and curative health. East Asian countries would be wise to avoid committing totally to social insurance, though it may prove useful in specific circumstances. Given constrained resources, priority should go to health, education, family services, and housing, though cash transfers will need to expand with urbanization and the growth of nuclear families. In-kind services are likely to contribute in the medium term to economic competitiveness and social justice 1 See also Anthony B Atkinson and John Hills, "Social Security in Developed Countries Are There Lessons for Developing Countries?" in Social Security in Developing Countries, ed Ehtisham Ahmad et al (Oxford, England Clarendon Press, 1991) 2 Bob Deacon, Michelle Hulse, and Paul Stubbs, Global Social Policy International Organisations and the Future of Welfare (London Sage, 1997) 3 Paul Krugman, 'The Myth of Asia's Miracle," Foreign Affairs 73 i 1994) 62-78 4 This also appears to be the case in Asia Even zakatlthe Islamic wealth tax earmarked for the poor, the mdi- gent, and others in need) raises significant sums of money only if it has strong government backing See Mohamed Ariff, "Resource Mobilization through the Islamic Voluntary Sector in Southeast Asia," in The Islamic Voluntary Sector in Southeast Asia, ed Mohamed Ariff Singapore Institute of Southeast Asian Studies, 1991) 5 Elmer Rieger and Stephan Leibfried, 'Welfare State Limits to Globalization," Politics and Society26, no 3 11998) 363-90 6 Fresh experience in southern European countries in developing activation policies in a context of extensive informal labor markets and continued familial employment deserves careful study 7 In the United States, the medical profession, private insurers, and private hospitals organized first a century ago and have managed to block proposals for national health insurance ever since European Social Policy Lessons: 12 Theses 21 BOX II, continued 8. Robert Goodin, "Towards a Minimally Presumptuous Welfare Policy," in Arguing for Basic Income, ed. Philippe van Parijs (London: Verso, 1992). 9. Mishra Ramesh with Mukul Asher, Welfare Capitalism in Southeast Asia: Social Security, Health, and Education Policies (New York, St. Martin's Press. 2000). 10. Ian Gough, Global Capital, Human Needs, and Social Policies: Selected Essays 1994-99 (Basingstoke, England: Macmillan, 2000). PART I Frameworks and Models: European and Asian Approaches to Welfare and Social Policy Introduction Part I offers a comparative overview of European and East Asian social policy and provides background on the social consequences of the 1997-98 financial crisis in East Asia, as wvell as lessons learned from the crisis. Chapter I offers a theoretical and practical framework for the book's discus- sion of social policy experiences in Europe and East Asia. Ian Gough categorizes "welfare regimes" in Europe into four broad types that place varying emphasis on social assistance versus social insurance and on private versus public providers of social services Social systems in East Asia, in turn, reveal common aspects such as low levels of government expenditure, but as in Europe they more often mask wide differences, some linked to varying levels of economic development (For more detailed information on social policy patterns in Europe and East Asia, see Appendixes A and B ) Chapter 2 questions the commonly held view that Asian and European social systems differ greatly because they place different weight on family obligations. Peter Scherer argues that, contrary to conventional wisdom, family obligations are still very important in Europe and are usually legally enforceable. European welfare history shows that formal social insurance did not develop as a substitute for family obligations but rather as a reaction to the burden such obligations impose. The chapter holds that better understanding of the true basis for social systems in Europe and Asia will shed light on alternative approaches, and on how countries that are suspicious of legal obligations can best cope with weakening family and community ties as they unfold. Chapter 3 shifts to the aftermath of the 1997-98 East Asian crisis and its role in revealing serious weaknesses in the social safety nets in countries that had been part of the East Asian "miracle." Tamar Manuelyan Atinc highlights major 25 26 1 Frameworks and Models European and Asian Approaches to Welfare and Social Policy household responses to the crisis as well as two central findings. the wide differ- ences in the effects by country, region, and social group (including severe effects on women), and the remarkable coping strategies that many families exhibited in the face of unprecedented economic challenges The chapter holds that growing urbanization and an aging population will make further strengthening of social safety nets in the region essential. In Chapter 4 Katherine Marshall builds on the previous analyses to encapsu- late lessons that East Asian countries and their international partners have learned from the 1997-98 crisis The lessons highlight the strong interaction between macroeconomic and social policy and suggest a continuing need to learn from this experience to prepare for future shocks. In Chapter 5 Christian Oberlander offers a case study of a conscious effort to adapt a European welfare model to an Asian situation: Japan's National Health Insurance. The Japanese experience is notable because the country quickly achieved universal health coverage with more reach than even European insur- ance schemes. At the design stage, policymakers skillfully relied on research of international as well as domestic experience in providing health insurance. The Japanese experience is an example to its neighbors and one worth referring to in Japan's own development plans. The crisis period generated a remarkable series of reflections and analysis that continues to this day. One institution taking an active part in the reflections is the Association of Southeast Asian Nations (ASEAN) (see Box 1 1). Throughout the crisis years and since, ASEAN has worked to move regional social policy issues to the fore, devoting considerable effort to framing lessons from the crisis and to understanding the role of social safety nets in ensuring a healthy economy. This is one illustration of the extensive international reflection on crisis lessons that has also marked much work by the multilateral development institutions, the United Nations (Economic and Social Commission for Asia and the Pacific), and the Asia Pacific Economic Cooperation Forum. Box 1.2 discusses some of the political aspects of the challenges posed by the legacy of decades of economic growth and then the financial crisis on the tradi- tional social contract in Asia. Although each country must clearly develop solu- tions to social and economic challenges that reflect its culture and history, Box 1.3 offers key principles that can inform effective programs and improve the overall efficacy of government spending in East Asia. Box 1.4 offers a summary of the Paris seminar on European Social Policy and the Asian Crisis. This seminar set out to frame the analytic work of the ASEM European Social Lessons Project, first, by providing background on the East Asian crisis and, second, by exploring possible theoretical frameworks for future discussion of social policy issues. CHAPTER 1 Welfare Regimes in East Asia and Europe Compared by Ian Gough ABSTRACT. This chapter uses the concept of "welfare regimes" to com- pare the social policy systems of East Asian countries-including Indonesia, the Republic of Korea, Malaysia, Thailand, and the Philippines-with those of European countries. European welfare regimes include a large, active public sector and high state expenditures with small market and declining family provision of social services. East Asian welfare regimes, in contrast, rely on substantial family provisioning and a very small public sector, with the private sector providing a growing share of social services. The two regions' wvelfare regimes also differ in outcomes European regimes provide much more security and equality than do those in East Asia. The chapter focuses on policy areas where East Asian countries might benefit from exploring the experience of European welfare regimes. 'An integrated approach of social policies and poverty reduction in the region has been lacking so far. Today Asia deeply needs a framework which brings syn- ergy between the different social policy instruments " (jean-Michel Severino, Vice President, East Asia and Pacific Region, World Bank; Philippines, November 9, 1999) The Asian financial crisis of 1997-98 called into question a core principle of East Asian welfare regimes that economic growth obviates the need for an elaborate system of social protection l This chapter aims to bridge the gap highlighted in Jean-Michel Severino's statement by comparing social policy in Europe2 to that in East Asia. Overall, European social welfare regimes providc much more secu- rity and equality than do East Asian regimes. 27 28 j Frameworks and Models European and Asian Approaches to Welfare and Social Policy The chapter first recaps the concept of welfare regimes3 and then analyzes them in four countries of Southeast Asia (Indonesia, Malaysia, the Philippines, and Thailand) and one in Northeast Asia (the Republic of Korea).4 The second section analyzes welfare regimes in Europe, and the third section compares East Asian welfare regimes with those in Europe, where policies and outcomes often represent opposite ends of the welfare spectrum. The chapter concludes by examining policy areas where East Asian countries might benefit from the expe- rience of European welfare regimes. Welfare Regimes in East Asia A welfare regime extends well beyond social safety net programs such as cash income supports, unemployment insurance, and old-age pensions, to encompass health and education. The concept also embraces the interdependent way states, markets, and households produce and allocate social welfare. Analysts have derived this definition of the welfare state from European experience, but several factors distinguish East Asia from the European experience.5 First, the role of the state in ensuring social welfare in East Asia is tiny, even by the standards of the developing world. Second, East Asian societies retain significant agricultural pop- ulations, a fact that implies a lesser role for formal social welfare institutions in shaping economic stratification and interest groups. Finally, the effect of external factors-the global economy and supranational institutions-is larger than in the world of the Organization for Economic Cooperation and Development (OECD) Thus the welfare mix in East Asia entails relatively low public responsibility for welfare expenditure, provision, and regulation; instead it is characterized by reliance on the family and community to provide services and redistribute wealth and by growing involvement of both private markets and community-based organizations in providing social services. In East Asian countries the public sec- tor gives priority to social investment in basic health care and primary education, paying little attention to social safety nets that aim to maintain minimum levels of income. Until the 1997-98 financial crisis, East Asian countries had been cur- tailing their dependence on foreign aid but increasing their openness to com- mercial penetration from abroad in providing social services. In all countries in the region, government personnel receive the most generous state support. In terms of welfare outcomes, the region achieves high scores in public health, education, and poverty reduction. However, persistent gaps and inequalities remain, especially in less well-monitored areas such as morbidity, school dropout Welfare Regimes in East Asia and Europe Compared 1 29 rates, working conditions, and patterns of exclusion from society of specific social groups (social exclusion). Variations do occur within this common pattern. Korea, by far the richest economy in the region, enjoys higher standards of education and other social out- comes and is embarking on a rapid expansion of social insurance in areas such as unemployment benefits. The Philippines exhibits much lower growth; a long- established, segmented, and partial social insurance tradition; and high levels of unemployment, poverty, and inequality, yet good access to education. The result has been labor emigration and high remittances from workers abroad that aug- ment the role of the family. Malaysia maintains a different policy profile, with its provident fund,6 alongside a U.K -influenced national health system, and rela- tively low levels of private finance for health care. Welfare regimes in Indonesia and Thailand are less institutionally developed and differentiated (For more information on all these regimes, see Appendix A.) Welfare Regimes in Europe Europe encompasses several distinct welfare regimes, there is considerable con- sensus on this typology, which is described, for example, by Gosta Esping- Andersen.7 They include different welfare mixes and outcomes as well as vary- ing development paths and degrees of economic stratification. Many scholars now delineate four regime types in Europe (see Table 1 1): 8 * Liberal- Ireland and the United Kingdom * Social democratic the Nordic countries (excluding Norway) * Continental Austria, the Benelux countries, France, and Germany * Southern: Italy, Greece, Portugal, and Spain Table 1.1 Dominant Elements in the Social Regimes of the European Union and East Asia State Market Community Family Education EU,a EAa EA Health EU,' EA EA EU EAa Pensions EU3 EU, EA EA,a EU Safety net EU' EA,a EU EU = European Union. EA = East Asia a Dominant insitiutional responsibility 30 I Frameworks and Models European and Asian Approaches to Welfare and Social Policy The classic continental model is characterized by high levels of spending on income transfers (notably pensions), highly regulated labor markets, and rela- tively good outcomes in terms of poverty and inequality, but poor labor market performance. In the southern European variant, inequality and poverty are more pronounced, but the role of the family is greater. Both these systems generate high old-age dependency through two mecha- nisms. First, work force participation rates are low, especially among women and middle-aged men, owing to weak demand for labor. This low demand, some argue, stems from the high contributions that the labor force and employers must make to fund social insurance.9 Second, fertility rates are very low, possibly owing to late household formation and the difficulties that mainly women face in com- bining work and family roles. The result is an unsustainable ratio of contributors to welfare recipients-I- in Italy at present Yet short of an external crisis, political support for the continental and southern European regimes is self- sustaining; as long as every family has one member in the protected labor mar- ket, it can benefit from his or her entitlement to a social safety net The results are a lack of employment opportunities for women and young people and con- cerns about social exclusion. However, this situation does not occur in the other two "social Europes," which offer a market and state solution to the contradictions of the continental social model. In the liberal regime of the United Kingdom, employment growth has been high, owing partly to labor market deregulation and low benefit levels. But like other liberal regimes, this one generates poverty and low wages, a prob- lem that the U.K. Labour government is now beginning to tackle The social democratic regimes in the Nordic countries (excluding Norway) have managed to avoid the continental trap through a different mechanism sub- stantial public investment in childcare and other forms of family care. This investment permits women and men to combine employment and child raising and, incidentally, provides a growing number of service sector jobs. In these countries, employment and fertility are higher, so dependency ratios are lower (with a ratio of 2.4 workers to I dependent in Sweden). The essence of the Nordic social model is relatively high spending on state services coupled with surprisingly unregulated labor markets, at least in Denmark. Social programs are typically universal rather than based on social insurance or targeted, though this is changing. Universal benefits do not discourage or distort labor market activity and do not make presumptions about people's behavior and preferences, and thus universal benefits are better adapted to a world of rapidly changing work, family, and life-cycle patterns.1O (For a more detailed account of education, health, pensions, and safety nets in Europe, see Appendix B )1i Welfare Regimes in East Asia and Europe Compared I 31 Comparing Welfare Regimes Overall, Western Europe retains the classic welfare state form: a large, active public sector; high state expenditures (particularly on income transfers), declin- ing family social provision (if a crucial assumption, we ignore unpaid domestic labor); and small but growing private sector involvement. Southeast Asia is almost the opposite: substantial family provision, a small public sector, and a growing market share of social welfare. Differences between the two regions are less pronounced in terms of program outcomes, especially in education, although European superiority in health provision and life expectancy is still marked (see Table 1 2). In terms of equity and security, European welfare regimes are far superior to their East Asian counterparts, notwithstanding strong families in the latter. Europe does better on measures of poverty, decommodification (that is, the extent to which social goods and services are protected from market transac- tions), inequality, the real income of the poorest, female employment, and other measures Europe also compares well with the "liberal" U.S. welfare regime in terms of poverty and inequality. Recent research has shown that the poor in the United States are more deeply poor, more frequently poor, and remain poor longer. 12 The last finding is especially important. once poor, people in the United States have less chance of escaping poverty. One commentator notes, "Whatever it is we want from welfare regimes-whether it is income stability, income equal- ity, low poverty, or high economic prosperity-the social democratic welfare regime seems the best on offer. . . . Conversely, the liberal welfare regime is unambiguously the worst on offer "13 Apart from income growth, the same ver- dict would hold true for comparisons with East Asian regimes. Annual growth of the gross domestic product (GDP) in the 15 countries of the European Union from 1994 to 1998 averaged 3.2 percent, the same as in the United States. However, the employment effect of growth in Europe is notably lower than in the United States, achieved through productivity growth stemming from falling average hours of work and falling employment levels. The desirabil- ity of this is difficult to evaluate. On the one hand, intensive growth of this sort can maximize leisure and choice between consumption and leisure On the other hand, it creates social exclusion-a divide between those inside and outside the labor force in countries that lack income supports (such as those of southern Europe). However, families in southern Europe can redistribute the benefits from insiders to outsiders. Overall, the European social model must continue to explore ways of combin- ing wage moderation and labor market flexibility with social security and public 32 1 Frameworks and Models European and Asian Approaches to Welfare and Social Policy Table 1.2 Social Expenditure and Outcomes: Europe and East Asia Compared United Kingdom East Nordic Continental Southem and EU 15' Asia Europe Europe Europe Ireland Income per head (US$ thousand) 22 3 31 Income per head, purchasing power parity (US$ thousand) 21 3 6 2 Public social expenditure (% of GDP) 341 7 0 Public social expenditure I% of total government expenditure) 85 39 Education (% of GDP) 5 0 3 4 5 7 5 0 4 7 4 6 Health(% of GDP) 68 12 65 77 55 58 Social security 1% of GDP) 22 3 2 2 Education Secondary enrollment l% of eligible population) 90 66 Tertiary enrollment I% of eligible population) 44 24 Average years of study 14 6 104 Health Under 5 mortality rate I% of children under 5) 66 32 Life expectancy (years) 77 4 69 2 Doctors per 1,000 population (%) 28 04 Hospital beds per 1.000 population (%) 72 20 Private health expenditure (% of total health expenditure) 23 59 Elderly living with children (% of elderly population) 17 80 7 14 38 16 GOP = gross domestic product a Averages weighted by population or total GDP Welfare Regimes in East Asia and Europe Compared 1 33 services A recent study suggests that Denmark and the Netherlands (as well as Ireland and Portugal) offer practical models for adapting to new external and internal pressures (see Table 1.3 for information on those pressures). 14 European Social Policy Lessons for East Asia A basic premise of the welfare regime paradigm is that policy proposals must take into account existing patterns of social provision, the distribution of institutional responsibility, and the interests that these express and perpetuate. That does not mean that policy learning, transfer, and change cannot occur, nor does it neces- sarily reject applying universal principles to policy goals 15 But it does caution against recommending universal policy designs and instead favors context- specific proposals that consider political economies and inherited institutions. These "lessons from Europe" are framed in that light.'6 Health Notwithstanding their generally good record,17 East Asian health systems are unable to address several emerging problems. These include a falling share of GDP devoted to public expenditures; low pay, training, and incentives for Table 1.3 Pressures Affecting Welfare Systems Sample consequences for advanced Source Pressure countries External Trade competition Deindustrialization. loss of unskilled (globalization) lobs Capital mobility and integrated Tax competition, "social dumping," reduced production bargaining power for states and labor Internationalized financial Decline of states' autonomy in setting markets macroeconomic policy Internal Low service sector productivity Slow growth in service sector productivity. (postindustrialization) "trilemma" of employment, equity, and budget stability Aging Growing pension and health expenditures Household transformation Smaller household sizes, more single-parent households, more women working Maturing of social entitlements Automatic growth of social expenditures 34 j Frameworks and Models European and Asian Approaches to Welfare and Social Policy doctors and nurses; a multi-tiered division of responsibility among public agen- cies, low government capacity for regulating, coordinating, and implementing health care, and weak regulation of the growing private sector involvement. A 1999 World Bank report presents a convincing analysis of these problems and offers options for reform.'8 These include raising health expenditures, empowering consumers through health promotion campaigns, strengthening the regulatory and administrative capacities of health ministries, improving providers' pay and incentives, changing price structures to favor primary care, extending social insurance (beginning with catastrophic health insurance), requiring higher co-payments, and mobilizing the resources of the private sector. Some of those options may be contradictory. European experience suggests that public insurance schemes, when coupled with fee-for-service payment of providers, the usual approach in Southeast Asia, generate cost inflation. Tax- financed national health services (NHS) are better at delivering standardized medical treatments within fixed budgets When NHS-type systems are in place, "there is little insurance gain from switching to social health insurance." 19 For example, the commendable Malaysian health system continues to be threatened with corporatizationi, privatization, and integration with the medical accounts of the Employee Provident Fund (EPF). Further inefficiencies stem from growing private provision alongside an underfunded public sector (for example, maldistribution of doctors results in high workloads and low incentive to work in the public sector).20 Moreover, health insurance typically excludes more costly treatments, which providers offer privately for higher remuneration. This gives providers a political incentive to keep such treatments out of the health insurance fee schedule More generally, the gradual move of the middle classes into the private health care sector undermines political support for the public system.21 Countries can reduce the major drawback of NHS systems- inefficiency in public provision-through a clearer purchaser-provider split and managed competition, as the United Kingdom, Sweden, and other European countries have done. East Asian health systems should aim to move from (mainly) private financ- ing and mixed provision toward (mainly) public financing and mixed provision This unusual prescription may be complicated by recent reports that foreign multinationals are purchasing indigenous health care providers. For example, AIA United Healthcare bought the oldest managed care business in the Philippines in 1999.22 Forthcoming liberalization under the Association of Southeast Asian Nations (ASEAN) Free Trade Area, which will allow unre- stricted foreign investment in hospitals, clinics, and laboratories by 2003, will Welfare Regimes in East Asia and Europe Compared 1 35 only reinforce this trend. Foreign investment may well bring improved practices, but it could inhibit national health care planning and coordination. The compar- ative history of public health care suggests that to avoid the domination of providers over consumers, the latter need powerful independent financial bod- ies, whether state run or owned or nonprofit, to protect their interests. The dan- gers of admitting powerful international health care providers into developing markets is that they, rather than health care consumers, may dominate the agenda. Pensions To draw lessons we must consider the emerging environment for pension sys- tems in East Asia and Europe, including demographic trends (see Table 1.4). In East Asia over the last 30 years, falling fertility rates and rising life expectancies have cut the share of children and raised the share of working-age adults in the population. The share of aged people rose slightly. The result has Table 1.4 Demographic Trends in East Asia Korea Malaysia Thailand Philippines Indonesia Average /%J (%) /%J /%) (%) /%J 65+ dependency ratio,a 2000 93 68 84 61 72 7 56 65+ dependency ratio,a2030 258 145 196 122 142 1726 Change in 65+ dependency ratio,2000-30 165 77 112 61 70 97 Change in total dependency ratio,b 2000-30 151 -16 0 3 1 -19 8 -8 2 -516 Change in weighted dependency ratio,' 1995-2030 16 7 -2 5 8 3 -4 0 07 3 84 a The 65+ dependency ratio = the number of persons over age 65, expressed as a percentage of persons aged 15-64 years b The total dependency ratio = the number of persons over age 65. plus children under 1 5 years. expressed as a percentage of persons aged 15-64 years c The weighted dependency ratio = the number of persons over age 65, plus one-third of children under 15 years. expressed as a percentage of persons aged 1 5-64 years Source David I Stanton and Peter Whiteford, Pension Systems and Policy In the APEC Economies (Manila Asian Development Bank. 1998) 36 l Frameworks and Models European and Asian Approaches to Welfare and Social Policy been a sharp fall in the overall dependency ratio. The prospects for the next three decades are very different, as the larger working-age cohorts enter old age and the smaller child cohorts enter working age. The share of those 65 years and over is projected to double in East Asia and to grow by two and one-half times in Korea alone. Thus the demographic profile of East Asia wVill mature in the com- ing period. The pattern of employment growth is important in determining whether these potential dependency ratios translate into actual dependency ratios. I am unaware of any projections of future work force participation rates, but the region's economies will probably shift gradually from high employment and wage growth to slower, more labor-intensive growth. This will disadvantage pay- as-you-go schemes for financing old-age pensions. The role of the family both in effecting transfers of wealth and in providing shared housing and other services will probably dimnish, but at a slow rate typ- ical today of Japan and southern Europe. In any case, the high levels of income inequality in East Asia (except in Korea) undermine the ability of household sav- ings and transfers of wealth to act as a source of old-age security.23 Thus, in comparison with Europe, East Asia faces a rapid demographic tran- sition, plus a gradual shift from extensive to intensive growth, in a context of extensive household transfers plus weak public pension schemes. The possibili- ties for learning from Europe appear weak. Nevertheless, the retirement age in East Asia is low in relation to life expectancy, and raising it offers a potential source of improvement. Europe offers practical examples of how to raise the retirement age within existing schemes (as discussed later). A strong case can be made for establishing universal first-tier pension systems in East Asia. Such systems can take one of two main forms: an unconditional grant or a means-tested, income-tested, or pension-tested allowance. 24 In coun- tries such as Thailand, where formal pension provision is weak, a universal citi- zenship pension has much to recommend it Such a system would offer admin- istrative simplicity, equity (for women and workers in the informal sector), adaptability to less-patterned work lives owing to global economic change, national solidarity, and political support. Scandinavian countries introduced cit- izenship pensions when a large proportion of the population was employed in the informal agricultural sector.25 For example, Finland introduced a flat-rate pension for all in 1956 when its GDP per capita (at purchasing power parity) was US$4,600, a level that is below that in Malaysia and Thailand today and only slightly above that in the Philippines. On the other hand, in countries such as the Philippines and Malaysia, where pension entitlements are accruing, a pension- Welfare Regimes in East Asia and Europe Compared 1 37 tested supplement available to all over the retirement age, as can be found in Korea, makes sense. When we turn to potential second-tier pcnsions, clear differences emerge between the social insurance schemes in the Philippines, Korea, and Thailand and the provident funds in Malaysia and, to a more limited extent, Indonesia. The Malaysian EPF was the world's first mandatory provident fund. It remains one of the most significant and has many unique features. It is therefore worth noting its institutional strengths as well as its weaknesses: * Adequacy of benefits. The fund posted an annual real rate of return of more than 4 7 percent from 1987 to 1996, enabling millions of workers to share in the returns to capital. Administrative costs are only 2 percent,26 and most members are happy with the proposed benefits However, the scheme is costly, with a combined contribution rate of 23 percent ( 11 percent from employees and 12 percent from employers). The ability of members to withdraw 30 percent of their entitlement for house purchases or improve- ment and another 30 percent at age 50 before retirement undermines the fund's role as an old-age pension. * Equity. The system offers weak protection against poverty in old age, given the unequal distribution of work incomes and the lack of any provision for redistributing benefits. Furthermore, extensive tax exemptions favor higher-income groups, and the system excludes casual workers and those in the informal sector. * Security. Capitalized defined-contribution schemes are at risk from falling share prices, as demonstrated during the East Asian financial crisis The result is that payouts can fluctuate substantially among cohorts of retirees and that guaranteeing payments against inflation (let alone keeping pacc with rising living standards) can be difficult Recipients can also spend their lump-sum payments unwisely, although housing purchased with EPF withdrawals may offer an important alternative to social security27 Raising the pension age and introducing a citizenship pension would over- come some of these drawbacks But more specific reforms, such as removing tax exemptions, introducing annuities as the normal form of payment, and improv- ing the rate-of-rcturn guarantee, are also necessary. The fundamental case for social insurance-based, defined-benefit, pay-as- you-go pensions is that they replace the traditional intergenerational solidarity of the household and extended family with a more formal and wider-based 38 1 Frameworks and Models European and Asian Approaches to Welfare and Social Policy solidarity in tune with modernized societies. They also permnt and usually undertake intragenerational redistribution. The result is a lower rate of poverty among the aged. Crntics claim that an aging population creates a fundamental problem for these programs, because the programs must demand higher contributions from the working population, lower pensions, or both, producing intergenerational con- flict. However, this analysis is almost always premised on a two-generational model. With a more realistic three-generational model, the nature and degree of intergenerational conflict change.28 A falling birthrate leads at first to a rise in the working population and a fall in the total dependency ratio The key measure is the total dependency ratio over a cohort's lifetime. In Southeast Asia falling num- bers of children will continue to offset the aging process, except in Korea and Thailand (see Table 1.4). This suggests that countries can finance pensions and health care by transferring resources from education.29 Nevertheless, reforms in addition to raising the retirement age will be needed to ensure that the implicit pension debt does not grow out of hand. Here European experience has much to offer. Pension reform in Europe has entailed hngher contributions and new sources of financing, shifts toward defined- contrnbution benefits, discouragement of early retirement, and the incremental introduction of a capitalized element into the total pension package Economists have debated the negative effects of higher contributions on the labor market sup- ply. However, the alternative-means-tested pensions-may dissuade work and savings. The fears of an adverse effect on labor supply may well be exaggerated.30 The final lesson to be learned from Europe is political. Proposed pension reform in France in 1995 brought millions out on strike and into the streets, sug- gesting that any attempt to challenge rights in a mature pay-as-you-go pension regime is doomed to failure. However, France did successfully retrench excessive pension entitlements in 1993, as did Italy in 1995 and 1997 3' The difference was the active involvement of trade unions and other parties in these reform exer- cises Pension reform requires open democratic government to succeed. The attempted French pension reform in 1995 was the first to tackle excessive public sector pensions-a warning about the potentnal difficulties that East Asian coun- tries will likely face in equalizing public and private pension entitlements Safety Nets Safety nets can operate at three levels: employment, income, and consumption. At the first level, government offers destitute people work programs, usually at wages so low that they automatically exclude those who are not poor At the Welfare Regimes in East Asia and Europe Compared 1 39 second level, government gives needs-based cash transfers or unemployment insurance to supplement income At the third stage, government provides direct services to satisfy the needs of poor families, such as food programs, "stay in school" programs (which include specific measures to prevent dropouts from school), emergency health cards, and housing assistance. A World Bank revicwe32 highlights how, in the wake of the East Asian crisis, many countries in East Asia employed all three strategies, as do European welfare states.33 In assessing lessons, we must consider how these efforts interact Nvith other components of the welfare regime. Asian families are a far more important source of assistance (in the form of work, cash, and kind) than European families-with the partial exception of those in southern European countries, which also have weak formal safety nets In Greece, Italy, Portugal, and Spain, family members employ around three-fifths of young people. Those countries rely on political organizations and faith-based organizations (for example, Catholic charities) to provide for the elderly and people with disabilities Except for Portugal, they exhibit low rates of female employment. This cluster of factors explains the absence of national safety nets in those countries.34 For this reason, recent exper- iments in Italy and elsewhere to adjust assistance programs to situations involv- ing strong families, weak compliance with regulations, and the participation of nonstate institutions such as political parties are of obvious relevance to East Asia. Yet, despite their common elements, southern Europe and Southeast Asia differ in their degree of labor market regulation and flexibility. The recent East Asian crisis resulted in lower wages rather than lower employment, an outcome that undermines the effectiveness of work programs and enhances the effective- ness of cash and in-kind benefits.35 Future Priorities for Welfare Regimes in Southeast Asia Productive social policies form the foundation for a high-quality, intensive path of economic development.36 The welfare regimes of East Asia embody some of these attributes, notably substantial public investment in basic education and health. East Asian countries need to build on this legacy, extending citizenship benefits to include universal secondary education and curative health, as well as giving priority to services such as health, education, and housing rather than to cash transfers Experience shows that welfare states can remain competitive by investing in people and communities rather than by relying on unproductive, large-scale cash benefits 37 Urbanization and family nucleation will steadily undermine the capacity of households to protect their members against 40 F Frameworks and Models European and Asian Approaches to Welfare and Social Policy cconomic risks and insecurities, hence it is likely that East Asian countries will need to augment their social protection systems. However, social services are more likely to contribute to economic competitiveness while satisfying people's needs and promoting social justice Notes I NVorld Bank, Towards ai Fast Asian Social Protection Strategy (draft, Human Development Unit, East Asia and Pacific Region, September 1999) 2 Bv "Europe," I mean the 15 member states of the European Union This definition excludes, among other countries, Norway, Switzerland, and all the countries of Central and Eastern Europe 3 There remains a problem of terminology The term "welfare" in the Asian context frequently denotes state handouts or charity It is often contrasted with "development," as in human resource development or social development, concerned with investing in people and productiv- ity-enhancing social instutions Thus the term "welfare regime" is likely to be seriously misun- derstood Perhaps "social policy regime" or "human development regime" would be a preferable term In this chapter, I use the term "welfare regime," with the understanding that it refers to a wide definition of social programs encompassing broad social and developmental goals 4 In places in this chapter, the countries are ordered in terms of their present income per head Korea, Malaysia, Thailand, the Philippines, and Indonesia 5 For a more detailed argument see Ian Gough, "NVelfare Regimes On Adapting the Framework to Developing Countries," SPDC PapLr No 1, University of Bath, 1999 6 Provident funds arc private savings funds for pension purposes 7 See especially Gosta Esping-Andersen, The Three Worlds of Welfare Capitalism (Princeton, NJ Princeton University Press, 1990) 8 See, for example, Maui izio Ferrera with A Hemerilck and M Rhodes, The Future of Social Europe Recasting Work and Wellare in the New Economy (Oeiras, Portugal Celta Editora, 2000) 9 Gusta Esping-Andersen rejects this argument on the grounds that the alternative-high direct and indirect taxes-will fall on employees, who will in return demand higher wages and thus push up wage costs through another route See Gosta Esping-Andersen, "Do Spending and Finance Structures Mattcr," in The Social Quaity of Europe, cds IVolfgang Beck, Laurent van der Maesen, and Alan WNralker (The Hague, Netherlands Kluwer, 1997), p 122 In my view, the likelihood of this scenario wvill depend on the balance of advantage in the labor market if labor is weaker owing to high unemployment, it may be unable to recoup its losses In this scenario, the form of taxation may have an effect independent of its level 10 Robert Goodin, "Towards a Minimally Presumptuous XVelfare Policy," in Arguing for Basic Insomie, ed Philippe van Parijs (London Verso, 1992) 11 For further comparative and prospective works on European welfare states, see Gosta Esping- Andersen, ed , Welfare States in Transition National Adaptations in Global Economies (London Sage, 1996), Beck, van der Maesen, and Walker, op cit, Martin Rhodes and Yves Meny, eds, The Future of European Velfare A sVen Social Contract 2 (New York Macmillan, 1998), Gosta Esping- Andersen, ed , Tie Social Foundations of Post-Industrial Economies (Oxford, England Oxford University Press, 1999), Giuliano Bonoli, Vic George, and Peter Taylor-Gooby, eds, European Ielfare Futures Towa?ds a Theory of Retrenchiment (London Polity Press, 2000), Stein Kuhnle, Welfare Regimes in East Asia and Europe Compared I 41 ed , Survival of the Euiropeatn Welfare State (London Routledge, 2000), and Ferrera, Hemertick, and Rhodes, eds , op cit 12 Robert E Goodin et al, The Real WIorlds of Ielfare Capitalismn (Cambridge, England Cambridge University Press, 1999), and Chris Picrson and Francis Castles, eds , The WVelare State Reader (Cambridge, England Polity Press, 2000) 13 Goodin, p 184 14 See Ferrera et al, op cit 15 See Len Doyal and Ian Gough, A Theor)' of Iluman ANeed (Basingstoke, England Macmillan, 1991) 16 Owing to lack of space and knowledge, I omit discussion of education policy here 17 According to Mishra Ramesh, "Most sick people have some access to health care " Mishra Ramesh with Mukul Asher, Welfare Capitalism in Southeast Asia (New York St Martin's, 2000), p 113 18 World Bank, East Asia Health, Nutritiorn and Population Strategy (Human Development Sector Unit, East Asia and Pacific Region, October 1999) 19 Paul Gertier, "On the Road to Health Insurance '['he Asian Experience," World Development 26, no 4 (1998) 725 20 Ramesh, op cit 21 Gertler, op cit 22 J Haresnape, ed, Health Care Global Oultlook 2000 (London Economic Intelligence Uniit, 1999), p 148 23 Ramesh, op cit, pp 73-74 24 Pension-testing raises existing pension income to a minimum level, income-testing takes into account all other income, and means-testing takes into account income and capital assets 25 See Evelyne 1-uber and John Stephens, "The Political Economy of Pension Reform Latin America in Comparative Perspective," UNRISD Occasional Paper no 7, May 2000 26 The lessons from Europe here are the mis-selling and excessive administrative costs cxperienccd in the decentralized, individually managed personal pensions accounts introduced in the United Kingdom, wherc various fees and costs consume 40-45 percent of the saluc of individual accotnts See Peter Orszag and Joseph Stiglitz, "Rethinking Pension Reform Ten Myths about Social Securit' Systems" (paper presented at thc WVorld Bank Conference, Washington, D C, 14-15 September 1999), p 31 27 See Frank Castles, "Thc Really Bi3g Tradeoff Flome O inership and the WelfIarc State in he New \Xtorld and the Old," Acta Politica 33, no 1 (1998) 28 David Collard, "Generational Transfers and the Genei ational Bargain," jovurnal oJ Intiernational Development 12, no 4 (2000) 453-62 29 The WVorld Bank argues that the costs of aging will exceed the savings on education by a substan- tial amount One attempt to take this into account is to calculate a weighted dependency ratio, as shown in 'Fable 1 4, which N%cights the costs of a child at one-third that of an eldeily person This suggests small changes, except in Korea and TIhailand, which still show a growing burdcn of dependency But given high education spending and low pension spending in East Asia, this over- states the problem See \Vorld Bank, Averting the Old Age Crisis (Oxford, England, New York Oxford University Press, 1994), pp 34-36 30 See Orszag and Stight7, op cit 31 See Giuliano Bonoli, "Pension Politics in France Patterns of Co-operation and Conflict in Two Recent Reforms," West Eulropean Politcs 20, no 4 (1997) 160-81 42 1 Frameworks and Models European and Asian Approaches to Welfare and Social Policy 32 World Bank, Towardls an East Asian Social Protection Strategy, op cit, Table 3 33 Unlike in the United States, where Medicaid and food stamps are important, in-kind assistance in Europe is limited mostly to housing According to Atinc and Walton, estimates of the cost of restoring the consumption of poor households in Indonesia following the crisis to 1996 levels are I percent of 1998/99 budgetary revenues for full targeted benefits, 3 5-5 percent for public works programs, and 6 5 percent for untargeted cash transfers These are all relatively small amounts. See Tamar Manuelyan Atinc and Michael Walton, "The Social Consequences of the East Asian Crisis," Background paper prepared for Responding to the Global Financial Ctisis (Washington, D C World Bank, 1998), p 22 34 Ian Gough, "Social Assistance in Southern Europe," South European Society and Politics I, no 1 (1996) 1-23, and Ian Gough, Global Capital, Hu,inan Needs, and Social Polices. Selected Essays, 1994-99 (Macmillan, 2000) 35 World Bank, Towardi an East Asian Social Protection Strategy, op cit, p 25 36 Bowles also arrives at this result, using an economic model that assumes completely unimpeded mobility of capital among countries He concludes, "Globalization does not rule out all egalitar- ian interventions There remain a large class of governmental and other collective interventions leading to substantial improvements in the wages, employment prospects, and economic security of the less well off Included are redistribution of assets which are productivity-enhancing, namely those that provide efficient solutions to incentive problems arising in principal agent rela- tionships such as wage employment, farm and residential tenancy, and the provision of environ- mental and social public goods in local commons situations " Sam Bowles, "Globalization and Economic Justice," Benjamin H Hibbard Lecture, University of Wisconsin, March 2000, p 5 37 Ian Gough, "Social Welfare and Competitiveness," New Political Economy 1, no 2 (1996) 210-32, and Ian Gough (2000), op. cit BOX 1.1 Social Policy Priorities in Asia: ASEAN Consultations in 2002 by Tara Karacan Indonesia, Malaysia, the Philippines, Singapore, and Thailand estab- lished the Association of Southeast Asian Nations (ASEAN) in 1967 in Bangkok to represent "the collective will of the nations of Southeast Asia to bind themselves together in friendship and cooperation and, through joint efforts and sacrifices, secure for their peoples and for posterity the blessings of peace, freedom and prosperity'Q" Brunei Darussalam, Cambodia, the Lao People's Democratic Republic, Myanmar, and Vietnam later became members. Today the 10-nation bloc represents 500 million people and a combined GDP of US$737 billion. Its secretary- general, Rodolfo C. Severino Jr., has observed that "closer integration among the economies of Southeast Asia in ASEAN has made them more competitive than they would otherwise have been."2 ASEAN has moved cautiously on the social policy front, focusing more on learning lessons than on defining common standards or approaches. Each of the 10 member countries has different concerns and priorities, generally reflecting its income and level of social development. For example, Singapore, the richest, is focusing on issues comparable to those confronting Western Europe-such as subsidies and welfare, uni- versal health insurance schemes, upgrading its citizens' skills, and fos- tering lifelong learning. On the other end of the scale, Cambodia is con- tending with large numbers of orphans, elderly people, people with disabilities, and demobilized military personnel from its civil war; low rates of participation in education; growing levels of HIV/AIDS; and human trafficking. Nonetheless, ASEAN countries share many challenges, and an analy- sis of best practices among their social programs can allow them to develop regional programs. Thus, since the East Asian crisis began in 1997, ASEAN has organized efforts to reflect on the future policy agenda. It was in this spirit that the Secretariat convened in January 2002 a conference in Jakarta that focused specifically on social policy continued on next page 44 I Frameworks and Models European and Asian Approaches to Welfare and Social Policy BOX 1.1, continued challenges Participants in "Social Development in the National Development Agenda" included senior officials responsible for key social sectors, representatives of nongovernmental organizations (NGOs) and international organizations, and academics and specialists. Country papers for the conference reflected common concerns such as containing the HIV/AIDS epidemic, reducing poverty, creating employment and income-generating activities, bringing basic health and sanitation services to the poor and rural areas, and assisting people with disabilities, children, and the elderly. Other common concerns included fostering participation by governments, NGOs, private businesses, com- munity-based organizations, mass organizations, and international organizations in delivering social services, promoting more involvement by women; and building local technical and financial capabilities. Participants' central conclusion was that the challenge is less to inte- grate social sectors into countries' national agendas-practically all have done so-but rather to determine how governments can reach social development goals in a context of globalization and market liberalization, economic cycles and financial crises, and demographic transitions. A related need is to ensure that economic planning encompasses social development, especially by allocating adequate resources. Overall the region is rich in innovative poverty alleviation programs, including those mobilizing and devolving responsibilities to local govern- ments and community organizations. Regional action should aim at link- ing these practices and practitioners in a continuous learning process. As President Gloria Macapagal Arroyo of the Philippines has observed, "One of the ways to make ASEAN continue to be relevant is to make agreements binding But another way . .. is to make sure that when there is a new cataclysmic development in the world, ASEAN knows how to respond, taking into account our cultural diversities, the different social organizations, and the different historical antecedents." What will count most in establishing a regional action plan will be actual practices and a clear learning curve rather than any well-concocted blueprint 1 2 Julia Clerk, Internationa/ Herald Trbune online, , January21. 2002 CHAPTER 2 The Role of Family Obligations in Developed Economies by Peter Scherer ABSTRACT Analysts often portray a formal European system of social insurance that gradually crowds out family-based assistance, while main- taining that in East Asia family welfare systems dominate However, family obligations in Europe are still a reality. In fact, formal social insurance is much less a substitute for family obligations than a reaction to the finan- cial impact of those obligations. That is why, for instance, public-funded pension schemes are still very popular in Europe and their reform meets such resistance Family obligations, in sum, are much less culture specific than is often assumed. They are common to all welfare systems. The issue today is how Asian countries can move toward law-based welfare schemes as families gradually disengage from assistance in the context of economic development. Discussions of social security in Asian countries have been strongly influenced by "positive orientalism"-the tendency to associate rapid economic growth and social progress in Asia with the absence of the dependency culture that has sup- posedly developed in the West l This chapter examines the role of family obliga- tions in Western social security systems. It argues that the perception that such systems have displaced family obligations is inaccurate. Furthermore, this per- ception has tended to divert debate over changing Asian social security systems away from key questions. The views in this chapter are those of the author and do not necessarily represent those of the OECD or its member countries 45 46 1 Frameworks and Models European and Asian Approaches to Welfare and Social Policy Obligations for Social Support under European Civil Law In 1996 Singapore passed the Maintenance of Parents Act, which introduced a legal obligation for children to support their elderly parents and established the Tribunal for the Maintenance of Parents to mediate and, if necessary, adjudicate such claims. A Singapore minister summarized the need for such a measure by contrasting it with the "negative experiences" of other countries: The key question behind the proposed Parents Maintenance Bill is who should support vulnerable old people who cannot maintain themselves adequately. Over the last four decades, the people in Europe, North America, and Australia have chosen the welfare state with high taxes and high social security transfers.2 The WVorld Bank draws the same contrast between the "family solidarity-based" systems of Asia and the social protection systems found in member countries of the Organization for Economic Cooperation and Development (OECD) For example, a 1994 World Bank report on pension system reform makes a clear dis- tinction between informal systems based on intrafamily obligations and formal systems based on wealth-transfer entitlements 3 A recent WAlorld Bank strategy paper also implies that these approaches are distinct: The dramatic negative effects of global financial crisis revealed the impor- tance of having well-designed formal SP [social protection] systems in place, which were lacking due to governments' resistance to the adoption of OECD-type SP programs and reliance on a different tradition of family-based support.4 Such analyses imply that formal social protection systems "crowd out" informal ones. However, despite this conventional wisdom, civil law in most continental European countries does require family members to support one another when in need. This requirement is not confined to underage children living in the parental home: parents must support adult children and grandchildren who are still engaged in full-time education or unable to support themselves, and chil- dren and grandchildren must support elderly parents and grandparents. Under French law, these provisions appear today in almost the same form as the Code Napole'on, the 1804 codification of civil lamv Although many elements of the code (such as the requirement that a wife obey her husband and the stipula- tion that a husband has authority over the couple's common property) have been The Role of Family Obligations in Developed Economies 1 47 abandoned, the family obligations remain. These requirements entail far more than a bare minimum of support French civil tribunals have stated that family support must extend beyond basic needs to allow the beneficiaries to "live in a decent manner and even to enjoy a certain level of comfort, a function evidently of the resources of the payer."5 Civil tribunals can enforce these obligations, and over the past two centuries an extensive jurisprudence has interpreted howv they apply in particular circumstances. Social insurance as it has developed in Europe and Japan is thus partly a response to family obligations rather than a substitute for them.6 Such systems also provide a powerful incentive for popular support of public pension and health care systems, and for opposition to their reform. This is because in the absence of such social insurance provisions, citizens would be legally liable for providing for extended family members who were in need One consequence has been a growing tolerance of subsidies from the general budget to social insurance systems For example, in Germany this subsidy has grown from 4.1 percent of pension payouts in 1960 to more than 25 percent in 1998.7 The impact of these provisions varies greatly from country to country, and often within each country. In most cases the requirement to contribute is means tested- a poor family does not have to contribute to its relatives' upkeep. However, the requirement to provide support is not necessarily confined to cases of dire poverty: within each family, members are required to support each other in line with their established standard of living. The concept in this respect is similar to alimony between divorced partners, but extends further than the nuclear family These provisions characterize the European countries that have now devel- oped comprehensive social insurance systems. But under the common law that originated in England and that has been adopted as the basis of civil law in English-speaking immigrant countries (including the United States), support of one family member by another is not enforceable in law, with the exception of underage children.8 The laws governing public assistance, the "poor laws," did stipulate that family could be required to pay for the upkeep of someone who would otherwise have been reliant on social assistance. But in contrast to European civil law, this provision was purely ancillary to the assistance law: its purpose was to minimize calls on public expenditure 9 No right to support existed for those who had not applied for public assistance.10 These provisions have now disappeared from public policy in these countries- legal liability for the upkeep of adult family members no longer exists in the English-speaking world (except in Singapore). In those countries social insurance institutions are found, but the level of coverage is in general more modest than in continental Europe or in Japan. 48 1 Frameworks and Models European and Asian Approaches to Welfare and Social Policy It would, however, be an exaggeration to suggest that family responsibility laws are the sole reason for the growth of European social insurance systems. In particular, the European countries with the highest rates of social transfers, the Nordic countries, have long abolished such laws.ii In those countries, commu- nity rather than family responsibilities are emphasized. Family Responsibility and Assistance Arrangements In Italy and Spain residents can request municipal assistance only if no family member can afford to provide enough support. In Austria, Germany, Luxem- bourg, and Switzerland-and generally in France-a person in need has a right to have local authorities meet those needs. However, parents or children of the recipients can be required to compensate authorities for the portion they can afford. In France the legal duty to provide support extends to daughters and to sons-in-law (though not to grandchildren-in-law), and in Italy, Portugal, and Spain it extends to siblings and half-siblings.i2 In all these countries social restraints discourage potential applicants from exploiting the system. In any country where governments can, in principle, recover social assistance costs from relatives, the shame of having to nominate relatives means that the number of applicants and recipients is always far lower than the number who are eligible. In Switzerland, for example, only one-fifth of eligible applicants apply in rural areas, and only about one-half in urban areas,13 even though benefit rates are high compared with average incomes.i4 In most countries local discretion also determines whether family members eligible to contribute must do so. For example, in Germany, local governments under social democratic administration tend not to enforce the rules. Another indirect but important factor also keeps the take-up rate low. Local municipalities usually administer social assistance, and in most countries the expenditures are charged to local budgets. (Exceptions include Australia and New Zealand, and to some extent Ireland and the United Kingdom.) This often means that local councillors make final decisions on assistance, even if they usu- ally act on the advice of social workers who have reviewed applicants' needs In Switzerland, this usually occurs at the level of the commune. Although Zurich consists of one commune of 450,000 inhabitants, the typical rural commune has 200 inhabitants, so councillors know applicants personally and understand the impact of granting assistance on local taxes. This type of pressure operates even in Finland, Norway, and Sweden, which do not require relatives to contribute. A social worker must still thoroughly examine each applicant's personal circum- stances and often report the results to the local committee, which decides, for The Role of Family Obligations in Developed Economies 1 49 example, whether a homeowner must sell his or her property if rental accommo- dations are available Relying on municipalities makes sense because people may need assistance for reasons that go beyond low earnings. In most countries municipalities assess an applicant's other assets and determine whether the income shortfall results from problems within the household. National administrations are not well suited to undertake these detailed investigations. Overall, these provisions do not greatly affect affluent families, as few mem- bers are likely to have incomes low enough to apply for assistance. However, the more general requirements of the civil law still apply, and these can result in claims for support by families that would not be eligible for social assistance. Certainly, for middle-class families with modest incomes, the impact of support- ing indigent relatives can be considerable This obligation also imposes the risk that a nuclear family cannot control its own affairs, because no matter how well a couple manages its finances it may still be liable for destitute parents Defining Minimum Living Standards In Europe, Norway and the Mediterranean countries-notably Italy and Spain-allow regional governments to set their own standards for social assis- tance. However, most other OECD countries establish national standards, although even these countries, which have fully monetarized economies and almost universal tax and administrative systems, can find it difficult to agree on such standards. Until the early 1990s Canada and the United States addressed the issue by offering matching grants to provinces or states, with the result that provinces and states with high rates of assistance received higher federal subsi- dies. However, in the mid-1990s these countries began providing block grants to provinces and states that no longer depend on actual assistance rates, although past rates of assistance do influence the grant levels Provinces and states have been left free to set their own standards. In the Czech Republic and the Netherlands local authorities claim that some 90 percent of their expenditures are financed by national authorities. This means that localities have no incentive to control expenditures (and might, in the course of seeking local influence, exaggerate their constituents' needs). To avoid this form of political "moral hazard," municipalities more commonly finance assistance out of their own tax revenues but receive additional grants according to their socioeconomic characteristics, such as the local unemploy- ment rate. 50 I Frameworks and Models European and Asian Approaches to Welfare and Social Policy Safety Nets, Family Obligations, and Poverty Alleviation Those examples show that drawing a line between informal systems based on family solidarity and formal systems based on social insurance is difficult. Formal systems based on family solidarity are common and form an essential part of today's safety nets. The introduction of social insurance systems that originate from such European sources into societies with a large informal sector can have perverse effects. By definition, only those in the formal sector-that is, in general, those with higher incomes and better social connections-will be covered. Experience has shown that it is very difficult to run such systems without incurring a deficit As incomes are generally higher in the formal sector than in the informal sector, the subsidy itself will be regressive in its overall impactiS1-and may still not address the most urgent needs.'6 But this does not mean that families alone can provide a safety net, formally or informally. The informal networks that tie families together are, of course, the basis for both economic development and mutual protection-economists know well that economic development does not result from the actions of disconnected individuals. Business networks and arrangements, within both families and com- munities, are essential for a successful economy; understanding their contribu- tions propels current interest in "social capital" as a component of economic growth. Social networks can block change as well as promote it. Families work together to build up businesses and pool investment funds, but they can also work together to exclude outsiders and protect established ways. Ethnic groups draw on common understandings to launch new ventures, but they sometimes close ranks to exclude outsiders and prevent competition. The sense of obligation and gratitude toward parents by children, an under- lying feature of the family relations in many societies, can have perverted conse- quences, particularly when fast economic development is driving aspirations to rapid increases in wealth. In Thailand, for example, concepts of obligation and gratitude toward parents can lead young people to leave school and their home villages in order to provide remittances to their parents-sometimes by entering harmful occupations, including prostitution 17 Such defensive uses of family and community ties are particularly likely to occur during times of crisis Competition that was tolerated, if resented, during periods of economic growth can be feared and attacked when times are hard. When there is no safety net to limit social distress resulting from failure, such tensions can become violent and destructive. These tensions were a major cause The Role of Family Obligations in Developed Economies I 51 of the rise of fascist regimes in Europe, and the generalization of social insurance after World War II was largely a response to this tragedy. For countries with very low per capita income, policies that promote eco- nomic growth are clearly essential to reduce durably the rate of absolute poverty. Any safety net program that impedes growth is, therefore, likely to be potentially counterproductive, because it risks, at best, protecting part of the population from sharing endemic poverty. Safety nets in countries with a large formal sector can easily fall into this defensive role. Economic growth that leaves significant portions of the population feeling insecure, with only community and family resources to rely on in times of crisis, is vulnerable to social unrest. Such unrest will pose a continuing threat even if the disposable income of the majority of the population continues to grow The growth process is particularly vulnerable to the withdrawal of investor capital that can quickly follow the first sign of unrest-a process that, in inducing a decline in capital values, can become self-perpetuating Supplementing Family Support in Asia The continued importance of family obligations in most developed Western economies shows that such obligations occur in all types of societies-there is nothing particularly "Asian" or "non- Western" about them. Even in countries that confine formal family obligations to underage children, families provide mutual support to their members throughout the life cycle For example, rela- tives provide most care of the frail elderly in all countries. 8 Countries with a strong Confucian influence on public affairs will be uncom- fortable with the emphasis on formal legal processes found in Romano- Germanic law and English common law. For example, the Republic of Korea is now revising its social assistance legislation to make aid an entitlement in cases of established need While Korea's Confucian tradition establishes a strong moral expectation that family members will care for each other, the law has no role in enforcing this. However, the newly defined entitlements are conditional on an assessment of the capacity of family members to assist Authorities can refuse to provide public assistance if an able-bodied relative should provide assistance, even if that relative does not have the resources to do so.19 Countries with a colonial past, such as Indonesia, tend to mistrust formal lawv as a foreign imposition. Customary and religious laws have standing in formal courts, but community leaders rather than courts generally enforce the law. As internal migration increases, other mechanisms will have to supplement these 52 1 Frameworks and Models European and Asian Approaches to Welfare and Social Policy traditional means of ensuring social support, particularly in urban areas, but it is not clear what form these should best take. Families remain a significant focus of social obligations and support in almost all societies. The question is how systems that are suspicious of legal obligations can best cope with the attenuation of family and community ties that accompany economic development. Must these societies invent formal obligations, as Singapore has done? Can the expansion of social insurance, as in Japan, Korea, and Taiwan, China, fill the gaps? The answers await further research and experience. Notes I Roger Goodman and Gordon White, "WVelfare Orientalism and the Search for an East Asian Welfare Model," in The East Asian Melfare Model IYelfare Orientalism and the State, eds Roger Goodman, Gordon White, and lIuck-Ju Kwon (London. Routledge, 1998), 3-24 2 Boon Heng Lim, "FamilyValues" (speech at the opening of the Singapore National Trade Union Congress Seminar on FamilyValues, Singapore), available at An account of the origins and functioning of the Singapore tribu- nal can be found in Pcter Scherer, "Formalizing the Informal Family Obligations in Modern Asia," in Children and Social Security Itternational Studies in Social Security, ed Jonathan Bradshaw (Aldershot, England Ashgate, 2002) 3 Vorld Bank, "Averting the Old Age Crisis Policies to Protect the Old and Promote Growth" (policy research paper, Oxford, England Oxford Unimersity Press, 1994) 4 R Holtzmann and S. Jorgensen, Social Risk Management A Arew Conceptual Frameiworkfor Social Protection, and Beyond, World Bank Social Protection Discussion Paper No 0006, 2000 5 Leopold Peyrefitte, "Considerations stir la Regle 'Aliments N' Arreragent Pas," Revue Trinestrielle de Droit Civile 67, no 3 (1968) 286-308 6 My full paper prepared for the ASEM Project, available on the ASEM WVeb site, lays out this argument in more detail See 7 A Borsch-Supan, "A Modcl Under Siegc A Case Study of the German Retirement Insurance System," Economncjournal 10, no 461 (2000) F24-F44 8 These countries began to formalize and enforce child-support obligations of absent parents only in the 1980s Public officials used rhetoric about family obligations to justify this change, but it was largely inspired by the desire to reduce the budgetary burden of public support for single parcnts 9 Alvin L Schorr, Filial Responsibility in the Modern American Family (Washington, D C US Department of H-ealth, Education, and NVelfare, Social Security Administration, Division of Program Research, 1960). 10 David TIhompson, "I Am Not My Father's Keeper Families and the Elderly in Nineteenth Century England," Law and History Review 2, no 2 (1984) 265-86. 11 For example, these obligations were deleted from Finnish law in 1970, when Finland adopted the Nordic welfare ideology See MarJatta Mann, "Generational Relations and the Law," in The Myth of Generatiotial CoGnflict, ed Sara Arder and Claudine Attias-Donfut (London Routledge, 2000), p 108 The Role of Family Obligations in Developed Economies 1 53 12 The principle applies in all continental members of the European Union except the Netherlands (where, though still formally present in civil laws, the provision has disappeared from the rules governing the provision of social assistance) and Denmark See J Millar and A WVarman, Famtily Obligations in Europe (London Fimily Policy Studies Centre, 1996) The principle also survives in some former members of the Soviet bloc, such as Hungary See OECD, Social and Labour Market Policies in Huungary (Paris OLCD, 1995), p 168 13 OECD, 7he Battle against Esciusion, Volume 3 SocialAisivtance in Canzada and Switzerland (Pans OECD, 1999) p 158 14 Ibid , p 87 15 The overall regressive impact of the subsidy depends, of course, on the net incidence of the taxes through which revenue is raised. If consunmption taxes and import levies are a significant revenue source, they will clearly have an impact, at least in part, on the informal sector 16 For a discussion of how a system of income security for employees in the formal sector could be designed to avoid this problem, see A C Edv ards and C Manning, "The Economics of Employment Protection, and Unemployment Insurance Schemes Policy Options for Indonesia, Malaysia, the Philippines, and Thailand," in East Asian Labour Markets and the Economic Cnsis Impacts, Responses and Lessons, cd Gordon Betcherman and Rizwanul Islam (Washington, D C, and Geneva NVorld Bank and International Labor Organization, 2000), pp 345-78. 17 1 am grateful to Chongcharen Sornkaew (Country Director for T hailand, Global Alliance for WVorkers and Communities) for this observation in the course of the ASEM Social Lessons Bangkok seminar 18 OECD, Caringfor Frail Elderly People Policies in Evolution (Paris OECD, 1996) For a discus- sion of the relation of social protection to family living arrangements in Asian countnes, see Peter Scherer, op cit 19 OECD, Labour Market and Social Safety Net Policies in Korea (Paris OECD, 2000) CHAPTER 3 How the East Asian Crisis Changed the Social Situation by Tamar Manuelyan Atinc ABSTRACT: This chapter reviews the effect of the 1997-98 East Asian crisis on households in the five most affected countries-Indonesia, the Republic of Korea, Malaysia, the Philippines, and Thailand-and their responses to the crisis. It focuses on what actually happened to poverty lev- els and inequality during the crisis years. It addresses the four major ways that the crisis affected households: through falling demand for labor asso- ciated with the economic contraction, relative price changes caused by massive devaluation of the region's currencies, a potential public spending squeeze in response to falling revenues, and possible erosion of the social fabric. It details what is known about how households in fact responded and how governments pursued measures in response. The main messages are that households and governments largely reacted to the crisis in sensi- ble ways and that the impact on household welfare, though substantial, was less than originally feared. However, the crisis exposed important limita- tions in the ability of private and public safety nets to cope with a shock of this magnitude. The chapter ends by briefly outlining major social and demographic changes (aging, urbanization) that lie ahead. In the wake of the Asian financial crisis of 1997-98, households and governments in East Asia had to adjust to the first serious economic contraction in a genera- tion, and with it wild swings in the relative prices of food, manufactured prod- ucts, and services (see Table 3.1). This chapter reviews the impact of the crisis on households in the five East Asian countries most affected by it-Indonesia, the Republic of Korea, Malaysia, the Philippines, and Thailand-and the responses, of both households and governments, to the crisis. 55 LA Table 3.1 Main Indicators of Economic Activity and Household Impact (%) Per capita Per capita private Inflation (Consumer Poverty Unemployment Government GDP growth consumption growth Price Index) incidence' rates' spending' 1990-96 1998 1990-96 1998 1990-96 1998 1996 1998 1996 1998 Educaton Health Indonesia 57 -144 68 -47 88 576 113 167 49 55 723 878 Korea 63 -66 65 -102 60 75 96 192 20 68 942 968 Malaysia 70 -93 54 -126 42 53 82 na 25 32 863 903 Philippines 04 -26 10 13 98 97 375 na 86 101 1038 922 Thailand 70 -108 64 -151 50 81 114 130 18 45 987 893 n a =not available a The numbers for poverty incidence are derived using national poverty lines and are based on income for Malaysia, the Philippines, and Thailand, and on consumption expenditure for Indonesia and Korea Poverty incidence numbers for Malaysia and the Philippines are for 1997, figures for Korea are for urban areas only b Unemployment rates are annual averages c In general, the figures show the ratio of govemment spending in 1998 to government spending in 1997 These figures are adjusted using the gross domestic product (GDP) deflator from Intemational Financial Statistics (IFSI for Korea, the Philippines, and Thailand The figures for Malaysia reflect the GDP deflator from the Malaysian Department of Statistics For Indonesia, the figure for education spending is the ratio of spending for 1998 to spending for 1996, and the health spending was obtained from World Bank staff estimates How the East Asian Crisis Changed the Social Situation 1 57 Effects of the Crisis on Poverty, Inequality, and Households Poverty Poverty increased substantially in all countries hit by the crisis.' Korea experi- enced the sharpest rise, as the proportion of poor people in Korea's cities rose from 7 5 percent just before the crisis in the first quarter of 1997 to a peak of 22 9 percent in the third quarter of 1998 Poverty rates then declined rapidly to 15.7 percent in the last quarter of 1998. Indonesia was hardest hit by the confluence of financial, political, and El Nifno shocks, and saw the ranks of the poor swell dramatically The percentage of households falling below the poverty line rose from 11 3 percent in 1996 to 16.7 percent in 1998; an additional 10 million to 12 million people were thus pushed into poverty 2 In Thailand the poverty head count grew from 11 4 percent in 1996 to almost 13 percent in 1998, as an additional I 1 million people fell below the poverty line. In Malaysia simulations suggest a rise in poverty incidence from 8.2 percent in 1997 to 11.2 percent in 1998 In the Philippines more than 90 percent of families reported being adversely affected by higher prices of food and other commodi- ties, whereas 17 percent reported a job loss within the country, and a further 5 percent reported a job loss from retrenchment of migrant workers 3 Throughout the r egion, investment fell much more sharply than consumption-a typical pat- tern in systemic economic shocks. Inequality Concern about rising inequality in many East Asian economies wvas growing before the financial crisis Inequality had been rising particularly sharply in China, Hong Kong (China), Malaysia, Thailand, and the Philippines, as indi- cated by Gini indexes (a measure of inequality) ranging from the mid- to high 40s, close to averages in Sub-Saharan Africa and Latin America, the two most notoriously unequal regions of the world. Strikingly, aggregate indexes for inequality in Indonesia, Korea, and Thailand, appear to have moved little during the crisis years However, the crisis affected various social groups in quite differ- ent ways and degrees In Thailand the percentage of people living in poverty in urban areas changed little (1.6 percent in 1996 versus 1.5 percent in 1998) despite the clearly urban nature of the initial shock, whereas rural poverty rose from 15 percent to 17.2 58 1 Frameworks and Models European and Asian Approaches to Welfare and Social Policy percent. The drop in real incomes was most severe for those with education below elementary levels. In Indonesia urban areas and Java were hit substantially harder than other parts of the country, and the percentage of people living in poverty in urban areas rose more sharply than in rural areas. However, because a majority of Indonesia's population lives in rural areas, more rural people became poor as a result of the crisis. While the nominal Gini coefficient changed very little over the crisis period, there were important differences in urban and rural areas. The nominal Gini does not account for price changes in measuring inequality, which is impor- tant in Indonesia because the poor have faced higher inflation than the rich (because of the increase in the relative price of food, which accounts for a larger share of the consumption basket of the poor). Calculating the Gini coefficient on the basis of appropriately deflated household incomes yields a slight drop in urban inequality (0.299 to 0.289) but an increase in rural inequality (0.265 to 0.289). There was practically no change in the Gini coefficient for urban Korea cal- culated on the basis of household consumption levels, whereas the Gini coeffi- cient that was based on income registered a significant rise from 0.271 to 0.301 between 1997 and 1998. This is caused, in part, by the differential savings behav- ior of households during the crisis. The data show very sharp increases in the ranks of the ultrapoor (from 1.2 million in 1996 to 3 million in 1998; see Table 3.2), consistent with the higher increases observed in the distributionally more sensitive poverty indicators. Between the first quarter of 1997 and the third quar- ter of 1998, the head-count index tripled, the poverty gap quadrupled (1.3 per- cent to 5.6 percent), and the severity of poverty quintupled (0.4 percent to 2 0 percent). Poverty incidence rose more for households headed by individuals with lower levels of education, although households headed by high school graduates and above accounted for a full two-thirds of the poor, reflecting Korea's high overall levels of educational attainment. Despite the material effects of the crisis on the welfare of poor households throughout the region, increases in poverty and inequality were generally smaller than originally anticipated. Five main factors help explain this unexpected out- come. First, contraction of economic output was smaller and less protracted than expected. Second, labor mobility between the formal and informal and urban and rural sectors may have cushioned the impact of the crisis by distributing the burden more broadly. Third, the relative price changes induced by currency devaluations probably favored the rural poor engaged in producing marketable surpluses, especially for export. Fourth, poor households reduced savings and reallocated items within their budgets to protect their consumption of critical How the East Asian Crisis Changed the Social Situation 1 59 Table 3.2 Changes in Korea's Poor Show the Vulnerability of the Poorest People 1996 1997 1998 Ultrapoor Number (millions) 1 2 11 3 0 Growth (%) -277 -111 1787 Marginal poor Number (millions) 1 9 1 7 3 3 Growth%) -20 6 -8 4 90 0 Poor Number (millions) 3 1 2 8 6 2 Growth (%) -23 5 -9 5 123 8 Near-poor Number (millions) 2 9 2 7 41 Growth (%) -13 4 -6 5 50 6 Note The ultrapoor are households with per capita consumption at less than 80 percent of the poverty line The marginal poor and the near-poor have per capita consumption at 80-100 percent and 100-120 percent of the poverty line, respectively Source Nonak Kakwani and Nicholas Prescott, "Impact of Economic Crisis on Poverty and Inequality in Korea," mimeo, World Bank, Washington, D C , 1999 items, such as staple foods. Finally, public transfers in the form of unemploy- ment insurance and other safety nets may also have played a role. Households In the book East Asia" The Road to Recovery, I identified four main mechanisms through which the crisos affected household welfare. falling demand for labor associated with the economic contraction, relative price changes caused by the massive devaluation of the region's currencies, potential public spending squeezes in response to falling revenues, and possible erosion of the social fab- ric.4 In the following sections I discuss the first two areas, for which reasonable data are available. Labor Markets A drop in demand for labor was the most important channel through which the crisis affected most households. High-income households also suffered substan- tial losses on property incomes (from dividends, capital gains, and rents). Shocks in labor demand hurt households through lower real wages, higher unemploy- ment or underemployment (those working less than 20 hours), and reduced self- employment earnings Whether the major impact stemmed from a change in the number of people working or a change in their income depends on the structure 60 I Frameworks and Models European and Asian Approaches to Welfare and Social Policy of the labor market, in which regulations make it more difficult or costly for employers to lay off workers, and collective bargaining arrangements, which affect the extent of downward wage flexibility. The prevalence of migrants, domestic or foreign, in urban areas also influences the outcome. Households, in turn, respond to shocks in various ways Labor force participation (including child labor) and the number of hours worked may rise or decline to compensate for lower wages Sectoral employment patterns can also change in response to the impact of currency devaluation (for example, away from construction and manu- facturing, initially toward agriculture), and labor may move from the formal sec- tor, which sustains the initial impact, toward self-employment and informal activities. Those initial effects and responses can produce different outcomes with respect to gender, age, income level, and differentials between skilled and unskilled wages What does the evidence show for the impact of the crisis on labor markets in the five most affected countries? The largest shocks to the domestic labor force occurred in Indonesia, Korea, and Thailand The Philippines had participated less in the recent economic boom (and hence suffered less from the crisis), and in Malaysia foreign labor bore the brunt of the adjustment burden. In all coun- tries, younger, less-skilled casual workers, as well as workers in the informal sec- tor, were particularly vulnerable Much of the adjustment to the fall in labor demand occurred through a decline in real wages. However, substantial variation occurred across the five countries, depending on the importance of the formal sector and the rigidity of the labor market. Unemployment and underemploy- ment increased significantly in Korea and Thailand. Work force participation rates fell in Korea and Malaysia, stayed constant in Thailand, and rose in Indonesia, where women, especially those age 25 and older, scrambled to supple- ment household resources in the wake of huge income drops 5 Work became much more informal, as labor shifted from the formal sector and employee sta- tus into self-employment, unpaid family work, and agriculture. In Indonesia and Thailand the relatively larger (smallholder) agriculture sector appears to have acted as a shock absorber, leading to smaller increases in open unemployment in those countries (especially Indonesia) than in Korea. Korea experienced the sharpest rise in open unemployment, from 2.5 percent just before the crisis to a peak of 8.7 percent in February 1999, before it fell to 4.6 percent in October 1999. Many people gave up searching for another job when they became unemployed and thus became part of the inactive population, between the second quarter of 1997 and the fourth quarter of 1998, the econom- ically inactive population grew by 9 percent, or 1.2 million people Most of the newly unemployed were low-paid workers-temporary and daily workers, self- How the East Asian Crisis Changed the Social Situation I 61 employed workers, and unpaid family workers-and therefore could not benefit from unemployment insurance. Women were significantly affected, representing nearly three-quarters of those who became economically inactive. Female work- ers were laid off before their male counterparts, and many opted for voluntary retirement in anticipation of layoffs; some were subsequently rehired on short- term contracts with reduced benefits. From October 1997 to October 1998 the number of regular female employees dropped by nearly 20 percent, compared with a 7.4 percent fall for males, and the number of female workers with short- term contracts rose while the same figure fell for males. Real wages in Korea also saw a substantial decline of 12.5 percent from mid-1997 until the end of 1998 before recovering during 1999. A large part of this decline may represent changes in the composition of employment: if workers shifted from higher-wage, formal- sector jobs to lower-wage, informal jobs, the average wage level would decline. Indonesia presents a marked contrast to Korea-not unexpectedly, because Korea is the wealthiest, most urbanized, and most industrialized country in the region, whereas Indonesia is still poor, rural, and largely agricultural. In Indonesia, the declinc in rcal wages proved far more important than the rise in unemployment. Open unemployment rose by very little (4.7 percent in August 1997 to 5.5 percent in August 1998) despite massive contraction in economic out- put. Underemployment also rose: 3.7 million more people were working fewer than 35 hours per week in 1998 than in 1997 6 Labor force participation rates, meanwhile, rose overall, remaining constant for men but increasing substantially for women (49 9 percent to 51.1 percent). Only 55 percent of women age 25 and older participated in the labor force in 1996, compared with 72 percent in 1998. Indonesia is the only country among the "crisis five" in which labor force partic- ipation rose, reflecting relatively low levels of initial income (and therefore heightened vulnerability of poorer households), the severity of the economic cri- sis, and the inability of formal and informal safety nets to cope with the shock. The main effect on family welfare occurred through a decline in real wages. Formal sector wages fell by 34 percent in real terms between 1997 and 1998, and agricultural wages also fell massively-by 40 percent in real terms. The period also saw a major shift from formal to informal sector employment and from mod- ern to agricultural sector employment (a net 5 million workers increase in agri- culture). In Thailand open cmployment increased more than in Indonesia but less than in Korea; unemployment rose from 2.3 percent in February 1997 to 4.8 percent in February 1998 and to 5.4 percent in February 1999. Unlike in Korea, the pickup in Thai economic activity does not appear to have boosted labor demand, so unemployment remained stubbornly high. Underemployment also 62 1 Frameworks and Models European and Asian Approaches to Welfare and Social Policy increased sharply during the crisis, and some cuts in working hours observed between 1997 and 1998 may have translated into open unemployment by 1999. But falling wages had the largest effect on household incomes. Between February 1997 and February 1998, wages fell by 6 percent in real terms; the decline was higher in urban areas (-8.3 percent) than in rural areas (-4.7 per- cent). Lower wages appear to explain two-thirds of the decline in real per capita incomes, with lower employment accounting for the remaining third.7 Well- integrated rural and urban labor markets in Thailand explain the widely distrib- uted drop in wages and the rise in rural poverty-many unemployed migrants sought work in rural areas. Changes in patterns of internal migration during the crisis were significant. The effect of the economic crisis on domestic labor was less adverse in Malaysia and the Philippines. Malaysia's large number of foreign workers (20 percent of the labor force) provided an extensive cushion for Malaysian workers because foreign migrants bore the brunt of the adjustment burden. With a fall in participation rates, particularly for women, open unemployment was limited; the unemployment rate rose modestly, from 2.7 percent in 1997 to 3.2 percent in 1998. Unemployment rates in the Philippines tend to be high (8-9 percent) com- pared with those in Korea, Malaysia, and Thailand, and, like Indonesia, the Philippines is a net exporter of labor to the region and beyond. Unemployment rates rose throughout 1998 but fell thereafter, whereas real wages remained essentially constant throughout the period. Price Changes Massive depreciation of the region's currencies meant large changes in relative prices (see Figure 3.1). This raised incentives for the production of tradable goods and should therefore have benefited the agriculture sector in particular. These changes help the rural poor if they are net producers of food and if increases in consumer prices or in world prices in terms of domestic currency are passed down to producers. Regrettably, important unknowns remain here; detailed information on household expenditure patterns is available, but there is less information on the production side. Food prices rose relative to other com- modities in all five crisis countries, but these changes were minor except in Indonesia, where food prices rose by 40 percent between mid-1997 and mid- 1998. The sharp increase in relative prices of food was particularly hard on the urban poor and on rural workers who are net consumers, but most rural house- holds benefited from the favorable shift in the terms of trade. In Indonesia the direction of change in farmers' terms of trade varied by region, from +10 percent to-15 percent. Regions with substantial production of How the East Asian Crisis Changed the Social Situation I 63 Figure 3.1 Relative Food Prices 150 - Indonesia 140 - --- Malaysia Thailand - ------ Philippines i ' 130 - --Korea o~~~~~~~~~~~~~~~~~~~~~~~~ I O~~~~~~~~~~~~~~~~~~~~~~~~ 11120 - 110 _ 100 --- ----- ---------- 901 02 Q3 Q4 0.1 02 Q3 Q4 QI Q2 Q3 0.4 Q1I 2 Q3 Q4 Q01 0.2 '95 '95 '95 '95 '96 '96 '96 '96 '97 '97 '97 '97 '98 '98 '98 '98 '99 '99 key export commodities, such as sea products, plywood, and rubber (Kalimatan, Sulawesi), experienced favorable changes in terms of trade, while rice-producing regions (Java) did not reap net benefits from the rise in rice prices. Concern arose early in the crisis that prices for pharmaceuticals, essential drugs, and contraceptives would rise to unaffordable levels, sharply curtailing use, with adverse effect on morbidity, fertility, and HIV/AIDS rates, in particular. Higher medical costs did hit the poor and the elderly particularly hard in several countries, including Malaysia. In Korea, among all categories of household spending, prices of medical expenditures rose at the fastest rate, averaging 1.8 per- cent per quarter compared with an overall quarterly inflation rate of 1.4 percent. Household Responses to the Crisis Apart from experiencing changes in labor supply, households in the five crisis- affected countries reacted to the shock in a multitude of ways. Households 64 Frameworks and Models European and Asian Approaches to Welfare and Social Policy reallocated expenditures to protect critical items in their budget. In Indonesia, per capita food expenditures declined in absolute terms for urban and rural households between 1997 and 1998, but the share of expenditures allocated to food rose.8 Despite the overall decline in food expenditures, households raised per capita spending on staples by 5 percent in urban areas and by 11.5 percent in rural areas (see Table 3.3). Data for Indonesia and Korea suggest that households also reduced their expenditures on nonessential expenditures that could be delayed (clothing, recreation, household goods), so the share of these expendi- tures within household budgets fell. The extent to which households protected their expenditures on education and health varied considerably across countries. In Indonesia, such expendi- tures fell in both absolute and relative terms. In Korea, in contrast, cuts in spending on education and health were smaller than overall reductions in household expenditures. Meanwhile, in Indonesia per capita expenditures on adult goods-alcohol and tobacco-declined only slightly (7 percent) in urban areas, while the share of these goods among total household expenditures Table 3.3 Consumption Patterns for Households in Indonesia during the Crisis (thousand rupiahs per month, in 1997 prices) Urban Rural 1997 1998 % change 1997 1998 % change Household expenditure 319 211 -33 9 194 168 -134 Fooda 1881 134 9 -28 3 147 8 135 8 -81 Staples 41 4 43 5 4 9 59 3 66 2 11 5 Meat 405 219 -458 242 164 -323 Dairy 117 79 -324 52 44 -144 Oil 62 61 -1 0 52 42 -205 Vegetables 284 180 -368 223 217 -23 Nonfooda Alcohol/tobacco 13 0 121 -6 9 8 6 6 8 -21 0 Health 5 5 31 -43 0 2 3 1 2 -48 5 Education 15 7 9 5 -39 2 4 6 3 0 -341 Household goods 261 14 3 -44 9 7 0 5 3 -23 5 Transport 100 68 -328 35 25 -274 Clothing 9 4 5 2 -44 2 4 3 2 5 -41 0 Housing 344 193 -439 119 81 -320 Recreation 8 2 4 3 -47 4 3 6 2 9 -19 6 a Per capiia expenditures on different data Source Indonesia Family Life Surveys (IFLS), IFLS2 and IFLS2+ data E Frankenberg, D Thomas, and K Beegle, "The Real Costs of Indonesia's Economic Crisis Preliminary Findings from the Indonesia Family Life Surveys,' Rand Labor and Population Program, Working Paper Series 99-04, Santa Monica, Calif, 1999 How the East Asian Crisis Changed the Social Situation 1 65 actually rose In rural areas, although the overall share of alcohol and tobacco declined slightly, the percentage decline was among the lowest among nonfood items. It is somewhat surprising that households did not reduce savings by more to protect their overall consumption levels Indeed, the data for Korea suggest that, in the aggregate, households increased savings in 1998 relative to previous years, although saving behavior varied significantly depending on household income. Households unable to smooth consumption (by spending savings, borrowing, selling assets, and receiving public or private transfers) may take shorter-term coping measures, some of which have severe long-run implications. Reduced nutritional intake, a delay in seeking preventive health care, and shortened schooling may spell irreversible losses in human capital and related income- earning potential. Although family survival may depend on a child's income, such responses are costly children are denied an opportunity to escape poverty and can face physical and mental harm. It is precisely the role of public policy to ensure that families do not have to resort to destructive coping mechanisms while taking care not to displace effective family and community responses to tempo- rary income shocks. In Thailand the use of public education and health services expanded during the crisis, including the use of basic services. By contrast preliminary data from Indonesia showed a large drop in health care utilization rates whereas 53 percent of those reporting an illness sought modern medical care in 1997, only 41 per- cent sought care in 1998. Among people who sought care, fewer went to public facilities (27 percent in 1997 and 20 percent in 1998), perhaps reflecting a per- ceived drop in quality following budget cuts. But there is no evidence of an increase in the number of children with low weight for their height (wasting) or low height for their age (stunting). In Indonesia aggregate school enrollment rates did not change much during the crisis: primary enrollment rose slightly from 1997-98 to 1998-99, from 94 percent to 95 percent, whereas secondary enrollment dropped slightly, from 54.5 percent to 53 2 percent However, whereas in 1997 there was no correlation between per capita expenditure and enrollment rates, in 1998 lower per capita expenditure was associated with a lower probability of enrollment, indicating that poorer families had more diffi- culty keeping their children in school.9 There are also indications that some stu- dents elected to prolong tertiary studies until labor market opportunities improved. In Thailand evidence points to a limited or negligible impact of the crisis on school dropouts. A 1998 nationwide survey in the Philippines suggested that 7 percent of families included in the study pulled children out of school dur- ing the crisis period. 66 5 Frameworks and Models European and Asian Approaches to Welfare and Social Policy Government Responses to the Crisis The onset of the financial crisis was sudden, and its social consequences caught policymakers off guard. Designing policies and interventions quickly in the con- text of severe information gaps, rapid change, declining revenues, weak institu- tional capacity, and a limited menu of pro-poor programs proved a considerable challenge. Citizens, meanwhile, did not place clear demands on governments for social programs, possibly because of an ethos of self-reliance or an absence of channels for citizens' voices. Even so, governments generally pursued sensible policies under the circumstances. Many East Asian countries made substantial efforts to collect more information on social conditions during the crisis, using rapid qualitative assessments to diagnose particular points of vulnerability and quantitative surveys to monitor more systematically the effect on households and benefits of government programs. Korea went the furthest in expanding safety nets. This reflected greater demands for public action-possibly encouraged by democratic institutions-as well as a greater capacity of the government to respond and a more extensive net- work of precrisis pro-poor programs. All countries tried to restore health and education spending once fiscal targets were loosened, but they succeeded to vary- ing degrees. Efforts were least successful in Indonesia, where total public sector health spending fell by 8 percent in 1997-98 and a further 12 percent in 1998-99, whereas total public expenditures on education collapsed by 41 percent between 1996-97 and 1997-98, but rebounded somewhat in 1998-99 to 72 percent of pre- crisis levels. Those dramatic reductions in real spending reflect significant infla- tion in Indonesia during the crisis period. The effect of real spending cuts on service delivery is difficult to determine, as the bulk would have translated into real wage declines for teachers and health workers. In Korea and Thailand social sector spending also declined in real terms between 1997 and 1998 but stayed constant as a share of gross domestic product (GDP) There is evidence that within the smaller envelope for health, countries protected spending on primary care. Differences also appeared in the instruments that countries used to mitigate the effect of the crisis (see Table 3.4) In light of the considerable corruption prob- lems associated with public service delivery, Indonesia stayed away from reliance on direct cash transfers. Indonesia pursued specific interventions to keep children in school, as did Thailand (with the highest ratio of youth in the labor force). Food security was a primary concern in Indonesia and the Philippines, where the impact of the drought was most severe. Social funds-special institutions and How the East Asian Crisis Changed the Social Situation 1 67 programs designed to support community and nongovernmental organization (NGO)-sponsored social programs-were introduced in countries with an active NGO community or where government institutions were failing, such as Indonesia. Finally, training programs, not particularly effective as a crisis response, were perhaps most suitable for Thailand, where a skills gap had emerged before the crisis. Prior to the crisis, publicly provided safety nets tended to be limited in scale and coverage, reflecting reliance on rapid growth and full employment, strong family-level and community-level supports, and social stigmas associated with public transfers. In Thailand major safety net programs represented only 2 per- cent of the government budget in 1996; in Malaysia spending on social assis- tance amounted to 1.7 percent of the current budget in 1997 After the crisis began, social safety nets expanded in all countries. This was most evident in Korea-which made concerted policy decisions-where the share of GDP for social protection rose from 0.6 percent in 1997 to 1 26 percent in 1998, and to almost 2 percent in 1999. In Indonesia outlays for redefined "core" safety net programs rose from about 1 percent of GDP in 1998-99 to 1 25 percent in 1999-2000. A dominant strategy across virtually all countries was the use of public works to generate employment. Indonesia substantially expanded its padat karya pro- gram in response to the crisis, creating an estimated 58 million days of labor that annually benefited more than 400,000 people. In Korea, since most of the jobless did not benefit from unemployment insurance despite expansion of the scheme, the government introduced a new public works program in May 1998. By January 1999 the program was providing 437,000 jobs and making a dent in the unemployment rate. However, the numbcr of applicants was even higher at 650,000. Community-based investment funds, notably newly introduced pro- grams in Indonesia and Thailand, were intended to generate both employment- driven and demand-driven infrastructure. Among the five countries, only Korea had an expanded unemployment insurance scheme. The country expanded this relatively young program of mandated insurance in 1998 from firms with more than 30 employees to cover firms with at least 5 employees, as well as to cover temporary and daily work- ers. This meant a substantial increase in coverage, from 5.7 million people in January 1998 to 8.7 million people by the end of the year. The government also shortened the contribution period required to activate eligibility and extended the duration of benefit payments. As a result of these changes, the number of unemployment insurance beneficiaries rose but still stood at a low Table 3.4. Upgraded Social Safety Net Programs in East Asia during the Crisis Unemployment assistance! Health and Workfare insurance! Active labor Food secunty Cash transfers Social funds educaton programs severance pay market policies Indonesia New program of Community-based New back-to-school program Expanded and targeted distribution programs (provided scholarships for the redesigned padat of cheap rice (OPK) poorest students and school karya program grants for schools in the poorest communities) Subsidies to maintain prices of essential drugs Korea New temporary, New public Expanded Expanded active noncontributory, workfare unemployment labor market means-tested scheme insurance programs (vocational livelihood protection program training, wage program and social subsidies, job pension for the elderly placement) Malaysia Expanded training for unemployed workers Philippines New computerized job assistance network and some expansion of training Thailand Expanded social pension New community- Expanded programs to provide New public Higher Expanded training for the elderly and cash based programs low-income health card for the workfare severance for unemployed transfers to needy poor and voluntary health scheme payments workers families insurance card for the near-poor New Employee Self-employment Expanded scholarships and Welfare Fund loans educational loan programs to partially Provision for families to pay fees finance unpaid in installments, for fee waivers, severance and for free uniforms to students claims for workers from bankrupt firms Expanded school lunch program 70 1 Frameworks and Models European and Asian Approaches to Welfare and Social Policy 10 percent of the unemployed labor force in March 1999, up from 1.9 per- cent initially. Korea also implemented a large wage-subsidy program, which paid one-third to two-thirds of a worker's wage if the firm could demonstrate a need for eco- nomic adjustment.'0 In 1998 the program covered 800,000 workers, and in 1999 it was expanded and the duration of the subsidy increased from 6 to 8 months.11 A survey of enterprises suggests that 22.3 percent of jobs would have been lost without the subsidy, confirming international experience that wage subsidies usually subsidize jobs that would have existed in any event. Korea, Malaysia, and Thailand modified their severance pay schemes. Thailand extended the maxi- mum amount of severance pay to 10 months and set up a fund to pay legislated severance amounts to workers whose firms went bankrupt. Malaysia also extended its severance pay scheme, allowing workers who quit voluntarily to receive severance pay. Other strategies for income maintenance included cash or in-kind transfers. In Korea, the government expanded the means-tested Livelihood Protection Program-which provided cash and in-kind aid to those unable to work-in terms of both the number of people covered and the budget (nearly a 40 percent increase in 1998). In drought-stricken Indonesia, the government increased the public provision of rice while introducing a targeted rice subsidy program (Operasi Pasar Khosus-OPK) in July 1998 to replace a general rice price sub- sidy. Education scholarships and block transfers through the Indonesian Stay in School campaign provided incentives to families to keep children enrolled in school. Thailand expanded the low-income health insurance card program. How effective was the public response to the crisis? Full evaluations are not yet available for most programs in most countries.'2 However, although we do not yet know the full effect of these public programs, we do have important indica- tors (rates of enrollment, health care utilization, and child labor) that can change in the short term. We also know that severance pay, unemployment insurance, and wage subsidies largely benefit formal sector workers; there is evidence that the Indonesian padat karya program did not adequately reach women, among other specific findings. In sum, the five governments implemented largely sensible policies in response to the economic crisis, expanding some safety net programs and attempting to protect expenditures in key sectors such as education and health (albeit with only partial success). But governments throughout the region were clearly hampered by their own lack of preparedness and menu of social programs to upgrade, by poor information, and by poor coordination. How the East Asian Crisis Changed the Social Situation I 71 Looking Ahead This chapter highlights three main conclusions. First, the recent economic crisis in East Asia exerted a large, negative effect on household welfare. Poverty became more acute throughout the region, and other social indicators, such as school attendance, took a turn for the worse in some countries. However, the effect of the crisis on household welfare was less than had originally been feared. Second, households and governments reacted to the crisis largely in sensible ways. Households protected their consumption of critical items, such as staple foods, while the labor market reflected the impact of falling wages and the reduced availability of formal sector jobs. Governments beefed up safety net programs and worked hard to gather data that could be used to inform policies. Nonetheless, the crisis exposed important limitations in the ability of private and public safety nets to cope with a shock of this magnitude. This is a source of con- cern as the region's growing integration into the international economy may make incomes more volatile even as it heightens the need for institutions that can help households manage risk. This is especially true because the faces and structures of East Asian societies are changing rapidly- they are becoming grayer, more urban, more formal, and more open economically and politically. These trends have important implica- tions for social policies. Demographic changes will bring the issue of old-age security to the forefront and require changes in patterns of public spending. Economic changes, driven by shifts in employment, urbanization, and continued globalization, will be associated with greater labor mobility and more household insecurity. Expanded political participation will put social issues on the political agenda. These trends will exert pressure on informal, family-based mechanisms for household protection and expand demands for more formal, government- mandated schemes. Demographic Trends East Asia's population is aging rapidly, thanks to falling fertility and rising life expectancy at all ages; the region has one of the world's most pronounced aging patterns. The old-age dependency ratio is still relatively low, ranging mostly from 5 percent to 8 percent, but it is projected to double in 30 years, reaching the cur- rent old-age dependency ratio of Organization for Economic Cooperation and Development (OECD) economies. 13 The ratio is expected to triple to 27 percent by 2040-near projected levels in Europe and Central Asia and well above those 72 1 Frameworks and Models European and Asian Approaches to Welfare and Social Policy of other developing regions. This aging is extremely rapid by historical compar- isons: the elderly population doubled in France in 140 years and in Sweden in 86 years. Changing demographics imply important consequences for public spending patterns and the design of social protection schemes. The changes suggest poten- tially greater vulnerability among the elderly population and the need to focus on old-age security programs, including pensions and health care. The youth dependency ratio is projected to fall from about the current 41 percent to 31 per- cent by 2040 14 This, too, has important consequences for government spending, particularly in education. Changes in fertility and life expectancy are closely related to income levels; Korea is furthest along in the demographic transition, with an old-age depend- ency ratio projected to reach 40 percent by 2040, whereas the corresponding fig- ures for Lao People's Democratic Republic and Cambodia are 7.8 percent and 11.3 percent, respectively There are exceptions: China's population is aging more rapidly than its income level would suggest, and the reverse seems true for Malaysia. Most East Asian countries will see fewer working-age people support- ing a growing proportion of elderly people, with particularly significant implica- tions for pension schemes. Economic Trends Urbanization and Services In 1960 all East Asian countries were still largely agrarian, with less than one- third of their populations living in urban areas. Over the last 40 years the per- centage of the urban population has grown, most dramatically in Korea, where the urban population rose from less than 30 percent in 1960 to over 80 percent in 1995. The urban share of the population is expected to top 50 percent in all but a few countries by 2030 and to reach more than 70 percent in Malaysia and the Philippines. Urbanization tends to go hand in hand with economic transfor- mation. In 1996 the agriculture sector still accounted for a significant share of GDP and an even higher share of employment in many countries, reaching 70 percent and more in China and other transition economies. With continued industrialization and market liberalization, the share of agricultural employment is expected to drop quickly. Despite the recent crisis, more workers are projected to form part of the formal labor market in both market economies and those in transition. Those trends are likely to weaken traditional social support systems and coping mechanisms. In predominantly agrarian societies with large families, How the East Asian Crisis Changed the Social Situation 1 73 informal arrangements, such as intergenerational income support for the elderly, have worked wvell Cultural traditions in Asian societies, which emphasize strong family values, have helped preserve this system. However, the experience of Japan and Korea suggests that the forces of modernity are strong More than two-thirds of parents age 60 and older still live with an adult child in Malaysia, the Philippines, Thailand, and Vietnam, but only 49 percent (1994) live with an adult child in Korea, down from 78 percent in 1984. In Japan, the proportion of people age 60 and above whose main source of income is the family was 5.7 per- cent in 1990, down from 15.6 percent in 1981 but still substantially higher than in Western countries (0.1 percent in the United Kingdom and 0.7 percent in the United States). Asian societies may exhibit great family solidarity, but the extent and coverage of private transfers is unlikely to be sufficient. Rapid urbanization, combined with falling fertility rates and overall increases in old-age dependency ratios, wvill place growing pressure on informal family- and community-based income support systems. An increase in formal sector jobs generates a greater need for more formal mechanisms of social protection Korea's recent experience illustrates the point agriculture and the informal sector were unable to absorb the unemployed (unlike in Indonesia), spurring the largest increases in levels of open unemploy- ment. Rising incomes, greater institutional capacity, and a more accessible tax base are all associated with growing formalization, making it more possible for countries to meet these demands. The transition toward greater democracy and public participation is influenc- ing the social agenda in much of East Asia NGOs and civil society more broadly are demanding greater accountability in the use of public resources, most strik- ingly in Indonesia but also in Thailand. As political systems in the region mature, concerns articulated by civil societ) groups will put pressure on mainstream polit- ical institutions New demands on governments will tend to push up government spending on social safety nets, as market economies respond to a more vocal middle class and transition economies ratchet down their support for state employees in favor of more affordable and equitable systems. Labor relations are an area in %Nhich such pressure seems particularly likely Labor unions are seeking to redefine industrial relations, both to protect workers during hard times and to ensure that they share the benefits during good times. Striking a balance between the legitimate concerns of workers and the need to maintain flexible labor markets will be important. Mechanisms that mitigate the downside risks-unemployment insurance, for example-while retaining wage flexibility may offer a way forward. An important risk is that political and social debate may revolve largely around the needs and aspirations of the new middle class, to the detriment of the 74 1 Frameworks and Models European and Asian Approaches to Welfare and Social Policy needs of poor households and regions. A greater voice for the poor and for the champions of the poor is essential. The international community can play a part by keeping this issue squarely on the table. Notes I Comparable pre- and postcrisis data are available for Indonesia, Korea, and Thailand The 1999 round of the Malaysian Household Income Survey has not yet become publicly available The Philippines conducted a new survey (Annual Poverty Indicators Survey-APIS) in 1998, but, unfortunately, consumption data are not comparable between the regular survey (Family Income and Expenditure Survey-FIES) conducted in 1997 and the APIS because the latter used a trun- cated consumption questionnaire Income data are also not comparable because the APIS used a 6-month reference penod, whereas the FIES used an annual reference period 2 This estimate is based on the 1998 Social and Economic Survey of Indonesian Households (SUSENAS) subsample of 10,000 households; other sources suggest that poverty may have been higher at the peak of the crisis 3 These results stem from the initial findings of the APIS survey 4 World Bank, East Asia. The Road to Recovery (Washington, DC. World Bank, 1998) 5 Susan Horton and Dipak Mazumdar, "Vulnerable Groups and the Labor Market The Aftermath of the Asian Financial Crisis" (paper presented at a meeting of the World Bank, International Labor Organization, Japan Ministry of Labor, and Japan Institute of Labor on "Economic Crisis, Employment, and Labor Markets in East and Southeast Asia," Tokyo, October 13-15, 1999 6 Ibid 7 Nanak Kakwani and J Pothong, "Impact of Economic Crisis on the Standard of Living in Thailand" (Bangkok National Economic and Social Development Board, Development Evaluation Division, 1999). 8 Data are from the Indonesia Family Life Survey (IFLS). 9 Data are from IFLS2+ 10 The firm could demonstrate need by shutting down 2 or more days a month, cutting hours by more than 10 percent, granting leave of a month or more to employees, transferring workers to affiliates, or switching to a new line of business while retaining at least 60 percent of existing employees. See Horton and Mazumdar, op cit 11 See Horton and Mazumdar, op cit 12 In 1999 Thailand conducted a special round of the regular household survey (Socioeconomic Survey) containing a crisis module, which included information on the assistance that each household member received from specific social safety net schemes 13 The old-age dependency ratio is defined as the ratio of population age 65 and above to the popu- lation age 15-64 14 The youth dependency ratio is defined as the ratio of population age 0-14 to the population age 15-64 BOX 12 Models and Politics for Asian Social Policies by Frangois Godemont Throughout the 1980s and early 1990s the European welfare state repelled Asians more than it attracted them.' It was seen as addressing issues specific to Europe (high social costs and unemployment, graying demographic forecasts, and conflictual labor relations), whereas the Western liberal market model seemed better suited to East Asia's needs. Traditional employment systems based on customary relationships and stratification were seen as ensuring equitable economic growth and social stability, making a redistributive welfare system such as Europe's irrele- vant and unnecessary Two developments jolted this Asian social model out of its compla- cency It was shaken first by a decade of accelerating growth that pushed several Asian countries into thc ranks of developed or nearly developed economies, and second by the Asian financial crisis of 1997-98 Challenge from Above: Impact of Economic Growth on Welfare Policies As Asia's most advanced economies (after Japan)-Hong Kong (China), the Republic of Korea, Singapore, and Taiwan (China)-grew richer and began exhibiting characteristics inherent to the European welfare sys- tem, they created new social protection policies such as medical care, social or housing assistance, and minimum retirement levels. Japan itself was the precursor; Japan's social spending rose in the late 1980s to levels greater than those in the United States-social security disbursement was 10.9 percent of national income in 1987 (against 8.8 percent in the United States). Other examples of far-reaching social policy reforms include the national health system in Korea; the unified medical care sys- tem in Taiwan, China, which has mandatory pricing of doctor fees; the This summary is based on a presentation at the June 2000 Paris Asia-Europe Meeting (ASEM) seminar conttnued on nea t page 76 I Frameworks and Models European and Asian Approaches to Welfare and Social Policy BOX 1.2, continued Provident Fund in Singapore, which in effect functions as a potential insurance scheme; the complex welfare systems in Hong Kong, China, which in effect guarantee minimum incomes to citizens, and generous public housing schemes in Hong Kong (China) and Singapore. A striking political feature of these reforms is how they came about and who supported them. They have been promoted either directly by political forces that defend the status quo, or indirectly in response to new challenges from electoral competition Increased social spending was often backed by established political forces. Because long-standing political majorities were traditionally associated with chentary relation- ships and promotion of Asian family values, their conversion to social spending was not viewed as a revolutionary development but as a tradi- tional form of public action, countering market forces and forcing a col- lective approach to social problems. Challenge from Below: Social Welfare after the Asian Financial Crisis The financial crisis revealed the huge underlying inequality created by gains in capital stock. As asset values collapsed, so did the fortunes of the well-connected entrepreneurial class and the new urban class Gross domestic product (GDP) decline and resulting business failures sud- denly brought lower wages and high unemployment to both formal and informal sectors. The financial crisis had two contradictory results. reversing the trend to underlying property inequality that had been growing since the early 1980s, and increasing absolute poverty. Urban middle classes were proportionally harder hit by a crisis that was finan- cial in nature than lower-income groups. The trend toward poverty erad- ication by way of economic growth was reversed and East Asia faced increased unemployment and dislocation of many public services. The crisis presented further challenges to the traditional social con- tract, and it accelerated the transition from "old" economies, based on labor costs and delocalization of mature and labor-intensive industries, to the "new" economy and technology-based enterprises, especially in Models and Politics for Asian Social Policies 77 BOX 1.2, continued the electronic and information sector The shape of recovery in Asian economies dictated a shift to industries and activities based on knowl- edge and new skills New directions of growth will surely accentuate edu- cation gaps, with an added twist Asian economies have relied on Asian graduates and returnees from top Western universities to fill important technical and managerial positions. This educational flow slowed because of lower incomes and higher relative costs. In Malaysia, students going abroad dropped sharply, and in poorer countries, particularly Indonesia, enrollment at every level of education declined. These fundamental changes will leave an imprint on future politics and necessitate changes in public policies. Public demand for accounta- bility, repudiation of past corrupt practices, and a general move toward a more open democracy and social system are one feature. In Korea, unions formerly torn between recognized entities and extralegal organi- zations have acquired a new collective sense of legitimacy, and with it increased social bargaining power From the political point of view, it makes sense for Asian governments to pursue proactive social policies. These policies will almost always run the risk of being tainted with locally corrupt practices, but social subsi- dies are not the only or the most vulnerable areas of public policies for corruption; privatization policies in the Philippines, for example, have been described as a key area of public fraud.3 An intermediate course can be to supplement market economics with social welfare actions managed, if not entirely financed, by nongovernmental organizations (NGOs) rather than by state bureaucracies. Charity and private welfare are a tra- dition in many Asian cultures, from Buddhism to the philanthropic actions of merchant classes and local organizations. NGO action com- plements state authority but does not necessarily compete with it. It may bypass corrupt practices in local bureaucracies. The concept of NGO action fits the image of a minimal safety net that would cost little, imply minimum overall administrative reform, and interfere little with what is seen as the only desirable course: a speedy transition to free market eco- nomics and open competition, including for social practices. continued on next page 78 I Frameworks and Models European and Asian Approaches to Welfare and Social Policy BOX 12 continued Models It is important to beware the temptation to characterize East Asia's social policies simplistically in terms of models. Japan, for example, suggests a zigzag course, from the rise of a public and corporatist welfare system along with economic growth to that model's present demise after a decade of economic and financial crisis. The more advanced emerging economies of Asia each created elements of a public welfare system. At the other end of the scale, China is an emerging economy that created very early on ele- ments of a complex social security system, chiefly to deal with unemploy- ment and retirement issues. The legacy of the socialist era "iron bowl" is compelling, but China's mixture of experimentation, pragmatism, and legal enactment is unique among developing economies. These reforms acted as an essential stabilizer during the Asian crisis. East Asian governments have tinkered with social formulas that answer partial and temporary needs, but this has had much less impact than self-help on family and subsistence sectors. Korea is an exception, as the only country with a balance of active social forces, mirroring some European examples; it is also the only one that has implemented signifi- cant social reforms, a mix of liberal (or new "third way") approaches enhancing the back-to-work priority of more classic welfare policies. This is definitely not a safety net approach but a welfare system. Indonesia and Thailand have pursued much more limited social policies, in keeping with both their income levels and the safety net concept. The poorer of the two (Indonesia) has the most comprehensive and nonideo- logical approach, mixing NGO, community, and local administration approaches in a state led (and internationally financed) approach. Civil servants, if not the state itself, are agents of these programs that hinge on community watchdogs. Thailand's pro bono approach is a gratifying story of community action and NGOs, though on a limited scale. Reform and transition from a political or customary model to a mar- ket economy seem unavoidable. This political dynamic will increasingly dictate public social policies that would have seemed unimaginable before the Asian crisis. Democratic debate and voter preferences will cer- tainly push in this direction. Models and Politics for Asian Social Policies 1 79 BOX 1.2, continued 1. An exception is the Philippines, where after the fall of the Marcos regime in 1986, the Aquino and Ramos administrations called often for a more humane social model involving European features. 2. Except for those who kept hidden assets overseas invested in Western financial and real estate markets. A major problem today (for example, in Indonesia), is to persuade owners of this hidden capital to retuan it. 3. See Walden Bello, "Payoff Scandal Hits ADB-Backed Power Privatization in the Philippines," available at the Web site of the Asia-Europe Dialogue on Altemative Political Strategies, , CHAPTER 4 The East Asian Crisis: Lessons for the Social Policy Agenda by Katherine Marshall ABSTRACT: This chapter highlights key lessons from the East Asian eco- nomic crisis about the implications of economic swings and crises for the social agenda. The first and most obvious lesson is that countries and their partners need to better prepare for socioeconomic crises by developing more extensive safety net programs, as well as the information essential to designing, managing, and assessing them. The means by which countries transform social policy-especially how far they enhance transparency and accountability and involve civil society-emergcs as critical. Although the 1997-98 socioeconomic crisis in East Asia affected different coun- tries differently, its rapid progression and the clear role of economic and social contagion jolted preconceptions of national and international policymakers about economic management, development, and social safety nets. The resulting rethinking of many assumptions has been perhaps most evident in the realm of social policy This policy rethinking focused on three major elements The first entailed a quest for effective measures to address the immediate and visible consequences of the crisis Countries launched programs to create jobs, created mechanisms to protect social expenditures during budget cuts, and sought to cushion the impact of increases in food and medical prices and the tendency of parents during hard times to take children out of school. Second, as it became apparent that "nor- mal" times would not return for some time, if ever, leaders pursued a broader- based effort to redesign the social safety net to respond to both current and future crises Finally, individual countries and the region as a whole began to reconsider the prevailing assumptions-that economic growth and job creation 81 82 1 Frameworks and Models European and Asian Approaches to Welfare and Social Policy should constitute the core of social policy, and that the dangers of the "welfare state" (creating dependency on government and undermining family values) out- weigh its social and economic benefits. Civil society, including nongovernmental organizations and trade unions, drove this reflection and rethinking in the initial period. The process emerged most quickly and evidently in the Republic of Korea, but debates in Indonesia, the Philippines, and Thailand quickly became vocal and widespread, and other countries followed suit. Regional organizations, including the Association of Southeast Asian Nations (ASEAN) and the United Nations Economic and Social Commission for Asia and the Pacific (ESCAP), took an early role. At the interna- tional level, the Asia Pacific Economic Cooperation (APEC) organization focused sharply on the social implications of the crisis and launched task forces to explore lessons for social policy. The topic was the focus of various meetings at different levels, including meetings of heads of state and government ministers. The region's major development partners, meanwhile-notably the World Bank and the Asian Development Bank-inspired by the unfortunate experience of the social costs of crises in other regions of the world, focused on social dimen- sions almost as soon as it was apparent that the crisis would not end quickly. They convened two international meetings-in Bangkok in January 1999' and in Manila in November 19992-to build support for immediate assistance and reflect on future policy issues and changes. A ministerial meeting organized by the Australian government in Melbourne in March 1999 gave further impetus to rethinking and action within the broader international community; numerous other gatherings at national and international forums focused on specific topics such as pension reform, labor-intensive public works, and migration issues. At the 1999 Bangkok meeting it was striking how far the national leaders expressed the widespread fear that the crisis was bringing severe and lasting social costs. They called then for urgent and ambitious new social programs requiring large-scale international assistance. A year later the social agenda was much less crisis driven and more hopeful and optimistic. However, policy responses across the board reflect, in retrospect, how far the crisis did change the social policy agenda in fundamental ways. As hope returned and allowed a longer-term vision of the future, the need to prepare for hard times and unexpected shocks, and to deepen efforts to combat poverty, emerged as new imperatives. The lack of effective systems for gathering information on the results of social policies was and remains a critical concern. In January 1999 gaps in knowledge from all quarters and all actors-policymakers and researchers alike-stared forth. But by then, not far into the crisis, policymakers already had launched many remarkable information-gathering efforts and had begun to note their The East Asian Crisis Lessons for the Social Policy Agenda 1 83 results. This information focus had a marked and growing impact on policy as time progressed. The basic messages at the later conference in Manila (thus by November 1999) were threefold. First, the impact of the crisis was much more complex- more varied by place, time, and social group-than initial assumptions and reports had suggested Second, the resilience of societies in the face of crisis offered an endless series of inspirational stories and underscored the realization that new social policies needed to take careful note of traditional coping mecha- nisms, tailoring programs to the cultural views and wishes of beneficiaries. Third, the enormous suffering of many groups brought home vividly the human face of globalization, the costs of economic swings, and the need to prepare much better for future uncertainties and troubles. The Manila conference also focused on the many examples of successful part- nerships among official agencies, members of the civil society, and private-sector institutions. It also, though, highlighted duplicated efforts and missed opportu- nities. Participants expressed a determination to build on the best experiences, improve communication and synergy, and ensure a broad understanding of the development partnership. A consensus about key lessons for social policy began to emerge from these gatherings. This chapter summarizes those lessons.3 They represent a distillation of thinking at the World Bank and other institutions, and reflect conscious efforts to mine "good practices" regarding the use of public policy to manage economic and social crises Seven Lessons from the Crisis Lesson 1: The crisis was a wake-up call that countries could no longer take uninter- rupted economic growth for granted. The price of deeper integration with the world economy is likely to be fluctuating levels of economic growth. This had signifi- cant implications for economies that had relied on growth and full employment to provide their social safety nets. Lesson 2: Policymakers need to concentrate more of their analysis and response on the key factors that are critical to household welfare in times of crisis. In 1997-98, for example, the focus was too much on formal unemployment and too little on changes in wage levels and shifts to the informal sector. Changes in demand for labor exert the largest short-run impact on households. In East Asia, unemploy- ment increased significantly during the crisis, but falling wages and the movement 84 1 Frameworks and Models European and Asian Approaches to Welfare and Social Policy of workers into low-paying, low-productivity jobs in the informal sector were even more significant. Few policy measures addressed these issues with effect. Currency devaluation and price changes had harsh consequences for the poorest segments of society, because they have urgent daily needs and tend to hold a greater proportion of their wealth in cash Price increases for food hurt house- holds that are net consumers of food In East Asia, these include the urban poor and poor farmers, whose food production and food stocks dropped sharply because of the El Niimo drought. And across-the-board budget belt-tightening tends to harm those who rely on public services, especially the poor. Overall, the social consequences of macroeconomic shocks can be large and unpredictable, undermining years of investment in education and other forms of human capital, notably health. Some households' rational short-term responses (like pulling children out of school to supplement household income or delaying medical treatment) have lasting adverse effects that are sometimes irreversible. Efforts by governments, the civil society, and their partners to keep children in school worked remarkably well in several countries. The public focus on "stay- in-school" programs linked to material financial support for parents, as well as practical solutions to specific problems that emerged, underscored for parents and communities alike the importance of education. Lesson 3: Safety nets work much better if theJJ are established before a crisis. Setting up effective safety nets, let alone social insurance mechanisms, during a crisis is particularly difficult financial resources are severely curtailed and subject to numerous pressures, there are inevitable implementation lags for programs designed essentially from scratch, and challenges in managing programs and navigating political pressures generated by crisis circumstances are great During the crisis of 1997-98, social assistance, despite ambitious goals and announce- ments, fell far short of demand, as existing safety nets were patchy and served a small minority of people. Poor communities were often unaware of programs designed to help them, and leakage of benefits to people for whom these pro- grams were not designed was sometimes substantial. When no crisis is in sight, effective safety nets help ensure that various eco- nomic shocks or other unexpected setbacks (such as droughts or other natural calamities) do not halt development, and that households continue to invest in health and education, thereby boosting productivity and spurring growth. Effective safety nets aimed at poor families and communities are thus a long- term investment. In economies in which the formal sector is dominant, unem- ployment insurance or savings plans are likely to be most appropriate. The East Asian Crisis Lessons for the Social Policy Agenda 1 85 During a major national crisis, safety net programs can provide a solid foun- dation for rapid expansion to meet crisis needs Experience suggests that work- fare programs are likely to be a key element that can help directly mitigate house- holds' loss of income. Several countries in East Asia, along with Latin American countries, have developed effective programs Targeted transfers of cash or food to specific subgroups of the poor, such as pregnant women, the elderly, and those who should not work, such as school-age children, can also play an important role. Lesson 4. Meclhanisms to monitor the impact of a crisis and evaluate policy responses are essential, not desirable Timely and complete information on the effects of a crisis on various groups and areas helps policymakers design appropriate responses and allows for corrections if the desired improvements do not materi- alize. In retrospect, there are ample examples of misdirected policy initiatives in East Asia that can be traced to the door of poor information In contrast, several examples of effective information systems were developed during the 1997-98 crisis, such as Indonesia's Social Monitoring and Emergency Response Unit (SMERU) Public information on the intended and actual results of social poli- cies can also help reduce leakage and corruption Lesson 5 Enhanced social dialogue and participation bY civil society are keys to effec- tive crisis management Social conflicts divert resources from economically pro- ductive activities. Dialogue among workers, employers, the government, and the civil society can limit such conflicts and generate cooperative solutions to serious problems Empowering marginalized groups and actors not only enhances social cohesion but also improves decisionmaking about social services and economic development East Asian governments face a particularly urgent need to explore better and more diverse mechanisms that allow the intended beneficiaries of social policies to participate in decisionmaking. Lesson 6 The 1997-98 crisis highlighlted a host of interconnected governance issues. Government accountability requires effective institutions, clear rules, and respect for individual rights. The two peak crisis years in East Asia focused sharp attention on weak regulatory systems and financial mechanisms, as well as on lack of coordination among social service agencies Efforts to strengthen and reform the public sector will remain crucial if East Asian countries are to respond more effectively to future shocks Policymakers also need to consider how attempts to decentralize and privatize social service functions affect accountability An active 86 1 Frameworks and Models European and Asian Approaches to Welfare and Social Policy press and civil society can prove crucial in ensuring accountability under these conditions Corruption thrives in the absence of transparency and accountability. Countries cannot afford, and donors cannot accept, social and developmental losses because of corruption. Anticorruption efforts are now a central focus of East Asian governments as well as of international donors and multilateral agencies. Lesson 7: No group needs more focus than the very poor, yet they are easily left aside as middle classes and organized groups dominate the agenda. The weakness of insti- tutions and programs directed to the poorest communities-whether govern- ment or nongovernmental organization-was apparent during the crisis. In many instances these communities had nowhere to turn, and despite goodwill, it took many months to identify their needs and gear up to respond to them. Conclusion Overall, the crisis revealed how far the social agenda encompasses but also extends well beyond strategies for reducing poverty. Policies directed to the very poor could not be isolated from the broader social issues for the population. The need to link social and economic policies stood out as an imperative. Policy- makers still can learn much from the experience of how these two domains inter- sected and interacted during the crisis years. The risk, of course, is that once the crisis has passed, all parties will return to the precrisis norm. It is worth recalling the maxim that those who forget history are condemned to repeat it; thus, we should focus on the urgent need to continue to mine the crisis experience to prepare for the future Notes I The background document prepared for the meeting by the World Bank is titled "Social Impact of the East Asia Crisis A Work in Progress," January, 1999 2 The work of the Manila Conference is reflected in a joint Asian Development Bank and World Bank publication, The New Social Policy Agenda in Asia Proceedings of the Manla Social Forum (Manila Asian Development Bank, 2000). 3 The author presented this list at the November 1999 Social Policy meeting in Manila CHAPTER 5 Universal Health Insurance in Japan by Christian Oberldnder ABSTRACT: This chapter gives a historical overview of the construction of a national health insurance system in Japan that is designed to illustrate the adaptation of models from one region (Europe) to another (Asia) Universal coverage through a multi-tiered insurance system and a uniform fee sched- ule have ensured the success of the Japanese health system. In the 1930s pol- icymakers made extensive use of international research as well as local pilot projects to study possible schemes for national health insurance. The mixed nature of those schemes-half voluntary, half mandatory-built on a decen- tralized network of existing institutions such as cooperatives. When World War II broke out, coverage was almost universal. In the postwar period implementation and financial consolidation of national health insurance resumed through the granting of state subsidies and the creation of a spe- cific tax. A nationwide drive to expand coverage again relied on a decentral- ized approach to win support from local politicians and stakeholders. Japan recently achieved the world's lowest infant mortality rate and highest life expectancy. In 1996 infant mortality in Japan was 3.8 per 1,000 live births, whereas life expectancy at birth was 84 years for women and 77 years for men.1 Long-time observers of the Japanese health care system Naoki Ikegami and John Creighton Campbell have concluded that "Japan's health care system . . helps to keep its population healthy at an exceptionally low cost." They "rate the Japanese system as excellent in cost control and access and very good in equal- ity,"2 Universal health insurance has been key in bringing about those results The Japanese health care experience, though not free of flaws-witness issues relating to long-term care, drug policy, and control of medical technology-is 87 88 i Frameworks and Models European and Asian Approaches to Welfare and Social Policy relevant to other nations for several reasons Confronted with the Great Depression after 1929, Japanese policymakers recognized early the close connec- tion between improving health and reducing poverty. They boldly aimed to ensure broad access to health care among workers in the large informal sector because they realized such availability was crucial to enable that population to weather economic crises. The search for relief for a distressed rural population led to expansion of health insurance with the ultimate aim of universal coverage Japan took these key steps when it was still at an early stage of economic devel- opment, with average per capita income close to that of Latin America, and was still borrowing from the World Bank.3 The nation's early drive to expand health insurance allowed Japan to achieve universal coverage in 1961, well before many other industrialized countries. This chapter examines the Japanese drive for universal health insurance, highlighting innovations that met the particular needs of Japanese society as well as potential lessons for other nations The next section introduces key features of Japanese health care financing that appear essential to its success. Then the chap- ter reviews the historical process that led to universal coverage from the initial planning stages by the Home Ministry during the Great Depression to the com- pletion of universal health insurance in 1961. Next the chapter examines European influences and models in the Japanese planning and policy process. The final section discusses lessons that Japan's historical experience may hold for other nations. Two Key Features of Japanese Health Care Financing The multi-tiered Japanese health care system4 offers varying degrees of subsi- dization according to people's economic vulnerability. In the 1920s, Japanese national health insurance focused on the formally employed, following the Bismarckian model.5 National health insurance (kokumin kenk6 hoken), organ- ized by regions and covering the self-employed and pensioners, was the central innovation in the drive for universal insurance coverage. On the payment side, the Japanese health care system reimburses providers according to a nationally uniform fee schedule, which the country introduced with the Bismarckian model. Because of this uniform schedule, people covered under any insurance scheme receive essentially the same set of treatments Providers cannot offer health services outside the fee schedule or collect charges in addition to what insurance pays for the services. (Informal payments are so Universal Health Insurance in Japan 1 89 small that they do not jeopardize the principle. )6 The fee schedule is subject to regular review and renegotiation These two key features-universal coverage through a multi-tiered insurance system combined with a uniform fee schedule-lie at the heart of Japan's suc- cessful health care system. Universal coverage allows the fee schedule to exert tremendous control over prices and provider behavior Although the fee sched- ule was an element of Japanese health insurance from the beginning, the country achieved universal coverage in four stages: legislation in the aftermath of the Great Depression, implementation of the plan, forced expansion and consolida- tion during and after World War II, and subsidized expansion to achieve univer- sal coverage. The Great Depression and the Creation of National Health Insurance Industrial and mining companies had formed mutual aid organizations by the 1880s, but Japan's first health insurance law was passed by the Diet in 1922. The country delayed enforcement until 1926 and began paying benefits in 1927. The early health insurance system, modeled after German sickness funds, covered disease and injury and provided some cash allowances but excluded dependents of the insured The system provided compulsory cover- age for employees whose annual wage was less than 1,200 yen and who worked more than 120 successive days at factories and mines covered by the Factory Act and the Mining Act Employees at any workplace with fewer than five employees were voluntarily insured. Employees made a contribution propor- tional to their wage or salary, while their employers also picked up part of the contribution. Because of the restrictions on income and size of workplace, fewer than 2 million of some 4 7 million formally employed workers were sub- ject to compulsory insurance.7 The Great Depression of 1929 hit farmers and others living in rural areas hardest. Urban workers who lost their jobs often returned to their villages, adding to overcrowding and poor living conditions. Doctors, in turn, wvere over- burdened and often unpaid, and they began to move away from the villages into urban areas in great numbers.8 Home Ministry officials who organized emer- gency assistance for the farming population recognized that sickness and death were among the leading factors in poverty. Medical spending was highest among urban dwellers, and higher among those living in villages with a resident physician than among those without direct access to a health care provider 90 1 Frameworks and Models European and Asian Approaches to Welfare and Social Policy Policymakers concluded that ability to pay and physical access, not need, deter- mined health care utilization. The rationale for expanding health insurance was thus twofold to lower the effect of health care costs on agricultural households, and to expand the presence of physicians and medical facilities in rural villages, thereby improving access to medical care. Japan expanded the health insurance system in 1934 but added only 370,000 participants to the 2 million members already covered. The Home Ministry began planning to introduce a second insurance scheme covering the rest of the 10 million people employed in the formal sector as well as many more people working in fisheries, as small shopkeepers, and in other informal sectors. To prepare the ground, officials documented the health care situation in rural villages and the effect of health spending on agricultural households. They learned of local insurance cooperatives that had existed since the beginning of the 20th century in the north of the island of Kyushu, where villagers made reg- ular payments to physicians to receive care in the event of injury and sickness. These arrangements often aimed at ensuring a living for doctors who would oth- erwise have left the community. Villagers calculated their contributions accord- ing to their ability to pay, making payments in cash or rice. Substantial co- payments lightened the burden and prevented moral hazard of unneeded access to health care. Some organizations even required members to make good any deficit in proportion to the medical treatment they had received. A second type of precedent the Home Ministry studied was medical facilities established and owned by the rural population itself to reduce health care costs. Agricultural cooperatives were operating medical facilities by the second half of the 1920s, but many did not survive because the cooperatives lacked managerial know-how and had difficulty retaining doctors. In 1928, to overcome the chal- lenges of small-scale provision of care dependent on private practitioners, a cooperative in Aomori m rural northeast Japan founded a full-scale hospital with 60 beds as well as specialty departments. Other cooperatives followed its exam- ple: by 1936, 738 cooperatives managed 2,791 beds, employing 461 physicians and providing medical services to 502,122 members.9 The Home Ministry con- cluded that the approach developed by rural people might solve the problem of reducing the financial burden of medical care. In July 1934 the ministry released an initial proposal for a health insurance system for the self-employed (mainly farmers and fishers) called National Health Insurance (NHI). The home minister noted that the plan was designed to even- tually establish a health insurance system for all Japanese. The original plan called for local insurance societies with compulsory membership, but it met with Universal Health Insurance in Japan 1 91 massive political resistance, especially from physicians who feared the bargaining power of highly organized demand. Faced with a need for health insurance schemes with voluntary membership, the Home Ministry conducted a detailed assessment of the health insurance situation in 20 European countries. Among all the foreign insurance systems studied, the policymakers found Denmark, with its noncompulsory and mostly regional health insurance societies for the general population, most relevant to the Japanese situation 10 Despite domestic precedents, the Home Ministry expected further criticism of its plans because most regions of Japan lacked familiarity with a local insur- ance system. To demonstrate and evaluate the operation of the planned system under local circumstances, the ministry launched and studied 12 health insur- ance societies in different parts of the country as pilot projects. Those organiza- tions later became formal National Health Insurance societies. In 1936, the cabinet conference formally decided to introduce the National Health Insurance (NHI) Act in the Japanese Diet During its 70th session, the Diet debated the act intensely, focusing on the position of pri- vate practitioners in relation to the new insurance societies. The law passed the lower house in 1937 after amendments opening the provision of medical care to all physicians subject to a group agreement Because the upper house could not complete deliberations, supporters reintroduced the draft law in the legislature in 1938, where it passed after brief debate. Enforcement of NHI began on July 1, 1938. Implementation of National Health Insurance The enactment of NHI marked the first time the Japanese government extended health insurance to the general population. Under the NHI Act, municipalities and employers organized three types of societies to act as insurance carriers. "Ordinary societies" covered residents of a local administrative unit; members of the same trade or business, such as barbers or grocers, organized "special asso- ciations"; and existing cooperatives formed "substitute associations." Member- ship was voluntary but could be made compulsory when two-thirds of the popu- lation concerned had joined the local society. The NHI unit insured all members of a household. The NHI Act of 1938 granted insurance carriers the freedom to set their own contribution rates and methods of collection. To encourage localities to imple- ment the scheme, the government granted each insurance carrier subsidies for 92 I Frameworks and Models European and Asian Approaches to Welfare and Social Policy the first year at a fixed rate per insured person, which gradually decreased. Under this voluntary scheme, policymakers expected communities to found some 500 local insurance societies each year and hoped to achieve reasonable cov- erage and a nationwide network of insurance societies within 20 years. The ini- tial increase in insurance societies to 2,025 during the first 4 years of operation seemed to confirm that estimate.II Although the NHI Act of 1938 laid down a broad organizational framework, it contained few provisions regarding benefits. Each insurance carrier had to pro- vide care for sickness and injury; maternal and funeral benefits were optional. In 1944-45, 73 percent of insurance carriers provided matemity benefits, and 3 per- cent paid funeral benefits. The act also left each carrier free to decide on the scope and duration of medical care Organizers expected health expenditures to drop significantly under NII for two reasons. First, NHI introduced co-payments to Japanese health insurance on a national basis for the first time. Although the administration had recom- mended co-payments at 20 to 30 percent of the cost of care, by 1941 most carri- ers had set a level between 30 and 45 percent. Second, prices set earlier by doc- tors would fall because consumption would rise, payment rates would also increase, and a markup for unpaid services that had been part of the old pricing system would no longer be needed. For example, the Okayama prefecture esti- mated the rate of unpaid bills at 20 to 30 percent at the time (whereas the rate in the United States was 10 to 20 percent) 12 Although in theory insurance societies could enter freely into contracts with providers of medical care, actual negotia- tions nevertheless tended to gravitate toward the existing fee schedule because providers would not accept lower prices and insurance societies were not willing to pay more. Insurance carriers were free to offer services and create facilities to maintain and improve health. The government actively encouraged the introduction of such services through technical and financial aid to insurance carriers. Public health nurses, for example, began to play an important role in rural areas after the introduction of the NHI scheme. According to a 1941 survey, 309 of 899 insurance carriers employed 344 public health nurses. Having their own medical facilities allowed carriers to combine insurance operations with preventive serv- ices and medical care that was largely free of the profit motive. In 1944 the gov- ernment began granting subsidies to cover part of the cost of construction of new medical facilities. Ownership and direct management of medical facilities also helped correct geographical maldistribution in the rural sector. By 1948 NHI carriers owned and ran 83 hospitals and 1,093 clinics.'3 Universal Health Insurance in Japan 1 93 Forced Expansion during World War 11 and Postwar Consolidation Humanitarian concerns had driven the advent of NHI, but when the war in Asia expanded during the 1930s, so did military preoccupation with the farming vil- lage as a source of wartime labor. The government supported its extension of health insurance coverage first to farmers, then to the entire population, with the rationale "healthy people, healthy soldiers." In 1939 the Employees Health Insurance Law extended coverage to urban white-collar workers, while the Seaman's Insurance Law extended coverage to sailors. In 1940 the national con- vention of insurance carriers formally endorsed the goal of establishing an NHI society in each municipality, thus approving universal health insurance In January 1941 the cabinet council adopted the Plan for the Establishment of Population Policy, which, among other things, underscored that the health insur- ance system must be strengthened and must cover the entire population A drastic increase in the number of insurance carriers and insured persons after 1941 stemmed from changes in the principles underlying the original scheme (see Table 5.1). In 1942 a prefectural governor's order made the scheme compulsory in areas not yet covered. The status of health care providers also Table 5.1 Insurance Carriers and Insured Persons (1938-62) Number of Number of Average number of Year insurance carrers insured persons insured persons per camer 1938 174 523,223 3,007 1940 937 3,045,046 3,250 1942 6,596 22,661,192 3,436 1944 10,474 41,161,301 3,930 1946 9,526 41,820,949 4,390 1948 5,446 25,826,890 4,742 1950 5,050 24,353,974 4,823 1952 5,008 23,088,674 4,610 1954 3,669 26,633,438 7,259 1956 2,870 30,582,065 10,656 1958 3,167 37,238,964 11,758 1960 3,599 46,171,092 12,829 1962 3,618 45,792,064 12,657 94 I Frameworks and Models European and Asian Approaches to Welfare and Social Policy changed: they now had to serve as insurance doctors by government appoint- ment. By 1943 NHI societies had been established in more than 95 percent of all municipalities in Japan. This period thus witnessed the creation of Japan's first "'universal" insurance system, even though it did not quite cover the entire pop- ulation. However, the question of universality soon became irrelevant because World War II destroyed most of Japan's health care delivery system. Many insurance carriers created under the 1942 amendments were too weak to survive the disruption of economic and social life, of more than 10,000 NHI societies begun during the war, less than half were functioning by war's end. The number of insured persons dropped sharply in 1948, and remaining insurance carriers faced difficulties The postwar withdrawal of medical providers previ- ously ordered to serve under NHI during the war also posed a serious challenge to insurance carriers. During this crisis, the government granted subsidies to cover carriers' expen- ditures for administrative personnel, public health nurses, and medical facilities When fears of famine developed, these funds included grants to purchase cows to supply milk and to produce simple pharmaceuticals. However, public subsi- dies still played only a limited role in NHI. In 1949 state and prefectural subsi- dies totaled only 572 million yen, whereas income from contributions totaled 4,196 million yen.i4 Policymakers took two steps to revitalize the NHI system after the war The first was a 1948 amendment to the NHI law making each municipal government responsible for organizing and managing its own NHI society. The prestige of local mayors with a direct role in organizing the NHI restored physicians' confi- dence in the system Ordinary insurance societies now required the consent of at least half the eligible residents, but after they were established all residents had to become members. Industrial associations-especially cooperative societies- that had helped develop the scheme no longer played any part The organizing principle had become completely regional. To stabilize the system financially, the government introduced the NHI tax in 1951. Before then, insurance carriers had individually prescribed contribution rates and methods of collection, but now the government assessed and collected uniform levies as taxes according to each municipality's number of households and insured persons. In 1952 the government introduced an incentive: reim- bursements would depend on the rate of contribution individual carriers could collect during a given period. This scheme was replaced in 1953 by a general sub- sidy for NHI equal to 20 percent of the cost of medical care borne by the insur- ance carrier However, this subsidy also depended on the financial situation of the local government, the level of protection it granted, and its performance in Universal Health Insurance in Japan 1 95 collecting contributions and controlling medical expenditures. The objective was to allow insurance carriers to overcome financial difficulties while providing an incentive for maintaining and upgrading their coverage to a reasonable level of protection. Subsidized Expansion toward Universal Coverage In 1955 the government introduced new subsidies covering 20 percent of the cost of medical care, one-third of the cost of public health nurses, and all NHI admin- istrative costs. These subsidies departed from earlier predetermined budgetary limits and strengthened the financial basis of the scheme by establishing central government subsidies for all insurance carriers. This encouraged local govern- ments that had not implemented NHI to do so, paving the way for compulsory application of the scheme throughout the country in the 1960s.15 The Advisory Council on Social Security, led by Hyoe Ouchi, president of Tokyo University, had recommended the establishment of a universal health insurance system in 1950. In November 1955 the Ministry of Health and Welfare-successor to the Home Ministry-released its Five-Year Plan for Social Security, which proposed comprehensive medical care coverage by 1960. Although the plan was not official ministry policy, it attracted a great deal of media and popular attention and triggered a public debate on universal health insurance. In January 1956 Prime Minister Ichir6 Hatoyama proposed his own plan to provide universal coverage by 1960. With 30 million Japanese still unin- sured, universal coverage was high on the political agenda during the July 1956 election, and in January 1957 the cabinet council formally stated its intention to establish a universal health insurance system within 5 years. Providers represented by the Japan Medical Association (JMA) opposed specifics in the proposed amendment, but they were not averse to the concept of universal coverage. The JMA reconciled itself to the plan because of the likeli- hood that eliminating bad debt would boost provider earnings. Also, the mass training of physicians during the war had created another problem for the JMA, because the nation saw a rapid rise in the number of physicians returning from military service. The number of doctors per 100,000 inhabitants rose by 19 per- cent from 1946 to 1955, from 89 to 106. Many physicians saw universal health insurance as a way to provide a continued livelihood for this growing group. In December 1958 amendments to the NHI Act made health insurance cov- erage mandatory for all Japanese. Many saw the 1958 changes as an entirely new National Health Insurance law. Local authorities in all cities, towns, and villages 96 1 Frameworks and Models European and Asian Approaches to Welfare and Social Policy had to establish health insurance funds by April 1961. To alleviate the financial burden of the NHI tax on lower-income groups, the government dropped uni- form levies by 40 and 60 percent in 1963 The state's financial participation con- sisted of providing subsidies, fixing the contribution rate by law, and offering grants within the limits of the national budget. As a result, national subsidies to states and prefectures in 1966 totaled 156,815 million yen, compared with income from contributions of 106,612 million yen.16 The 1958 act aimed to elim- inate discrepancies by prescribing that co-payments could not exceed 50 percent. A June 1966 amendment reduced this limit to 30 percent. The timing of those measures proved fortunate because Japan experienced rapid economic growth from 1955 to 1957 and from 1959 to 1961, so the country had enough fiscal strength to implement universal insurance rapidly. The effect of the 1958 legislation is evident in the remarkable rise in the number of insured per- sons in 1959 and thereafter The Japanese universal health insurance system was fully enacted nationwide in 1961, and the system continues to operate today. Japanese versus European Experience and Health Policies When introducing NHI, Japanese bureaucrats, politicians, and other players were keenly aware of European health insurance systems Officials took their esti- mate that leading foreign countries had achieved health insurance coverage for about 30 percent of their population as a strong indication that Japan needed to expand coverage beyond the 2 million initially insured under the 1922 Health Insurance Act. In planning the prewar drive to expand health insurance, the Home Ministry noted that in Denmark private medical practitioners had cooperated in the drive to expand voluntary health insurance in order to secure their livelihood, and pro- fessionally oriented health insurance associations had turned mostly regional by 1892. 1 Aware of the much more critical attitude toward health insurance of Japanese doctors, the ministry initially took a careful approach to extending the new NHI scheme. During the debate about the organizational structure of NHI, industrial coop- eratives repeatedly invoked the cases of European countries to support their calls for a greater role for themselves. The cooperatives highlighted the successful contribution of different types of cooperatives as carriers of health insurance in France 18 Criticizing the role of private physicians under NHI, cooperatives argued that it would only repeat in Japan the failure of the health insurance of Germany and the United Kingdom. Universal Health Insurance in Japan 1 97 As Japan switched from a voluntary system to forced development of near- universal coverage during World War II, European policies again served as important guideposts for Japanese decisionmakers. Officials referred to the fact that England and Germany had strengthened their health care systems as part of their wartime policies with expanded insurance coverage playing a central part. During the final drive for universal coverage, Japan appears to have acted largely without conscious reference to foreign models. The brief existence of near-universal coverage during the war gave Japan a structure for renewed post- war expansion, and an egalitarian public consensus allowed political actors to implement this expansion despite postwar chaos. Japan set ambitious goals and achieved universal coverage quite early com- pared with many European countries. Germany has still not achieved 100 per- cent coverage 19 France expanded its insured population significantly as late as the 1960s, when first farmers (1961) and later self-employed people outside agri- culture (1966) became part of the system. France achieved near total coverage only during recent reforms. Only in the United Kingdom did the entire popula- tion gain access to all areas of medical services with the implementation of the National Health Service in 1948 Lessons for Other Nations The Japanese experience in introducing and expanding NHI offers many poten- tial lessons for other countries First, clearly defined values and objectives are important. Once officials at the Home Ministry had developed the vision of a relief program for the farming population that would ultimately lead to univer- sal coverage, they could screen domestic and international precedents and norms effectively. Policymakers invested several years in implementing pilot projects that were financed largely by a nongovernmental third party. To gain public acceptance more quickly, officials drew preexisting organizations that expressed social solidarity into the new scheme By referencing an international precedent (Denmark) that accommodated key political stakeholders (physicians) by making their participation voluntary, the officials preemptively weakened potential oppo- sition to NHI. The second lesson is that expanded coverage should not be rushed. NHI was initially meant to grow at a slow pace, with modest financial support leading to universal coverage. The introduction of coercive elements during World War II allowed sudden expansion to near-universal coverage, but it proved short-lived. The final push for universal coverage in a democratic environment after the war 98 I Frameworks and Models European and Asian Approaches to Welfare and Social Policy required general government subsidies for insurance carriers and specific sup- port for low-income groups Although officials used coercion and subsidies tem- porarily, they still took more than 20 years to cover the entire Japanese popula- tion. The initial estimate of NHI planners proved roughly correct. Third, the stance taken by professional care providers as key stakeholders can profoundly influence the fate of a drive for universal coverage. When officials announced plans for NHI, powerful groups of physicians had already voiced opposition. Policymakers overcame this resistance only at the price of abandon- ing initial plans for compulsory health insurance. Local physicians' groups also opposed the creation of insurance societies in many places. The command econ- omy of the war years briefly eliminated physicians' resistance, as NHI forced them to serve as providers. However, this led to massive defection of physicians from health insurance as soon as the country lifted the coercive measures. What finally led physicians to accept universal coverage through NHI was the prospect of income security at a time when the profession had grown strongly in numbers. Fourth, a gradual approach combining different organizational principles and creating (temporarily) independent risk pools can draw in more social groups, finally leading to universal coverage. NHI was basically organized regionally like its Danish model, but it also initially left room for carriers organized along pro- fessional lines. After the war, risk pools gradually linked themselves into larger units, and government subsidies provided redistributive financing. The three tiers of the Japanese health insurance system therefore correspond today to three levels of subsidization by the central government. Health insurance managed by societies mostly for employees of large companies must cover benefits solely out of premium income, whereas government-managed insurance mostly for employees of small companies and NHI receive subsidies of 14 percent and 50 percent, respectively. Fifth, growing coverage of underserved groups need not lead to massively ris- ing costs. The initial introduction of NHI included cost-control measures in the form of substantial co-payments and elimination of markups for unpaid services. The Japanese experience after 1961 confirms observations from other countries that lower unit costs from co-payments and a steady income flow for medical practitioners help offset the cost of greater use of health services.20 In the Japanese case universal coverage combined with strict enforcement of the uni- form fee schedule has been critical in avoiding some of the cost inflation observed in other countries 21 Overall, the Japanese experience suggests that policymakers can effectively manage the main risks of health insurance: incomplete coverage of the population and rising costs.22 Universal Health Insurance in Japan l 99 Notes I Richard Cash, Christian Oberlander, and Akiko Matsumoto, "Lowering Infant Mortality in Japan (1915 to 1965) The Role of Government and Civil Society" (Cambridge, Mass, 2000, unpub- lished manuscript) 2 Naoki Ikegami and John Creighton Campbell, "Health Care Reform in Japan The Virtues of Muddling Through," HealthAffairs 18, no 3 (1999) 56-75 3 Angus Maddiso, Monitoring the World Economy 1820-1992 (Paris Organization for Economic Cooperation and Development, 1995), pp 20, 196-97 4 For a full discussion of the Japanese health system, see Ikegami and Campbell, op cit 5 For a detailed discussion of the Bismarckian insurance concept, see Frank A G den Butter and Udo Kock, "Social Security, Economic Growth, and Poverty Theoretical Considerations and Guidelines for Institutional Arrangements," available at 3 Sainsbury, op cit 4 Inga Persson, "The Third Dimension-Equal Status between Swedish Women and Men" in Generating Equality in the Welfare State The Swedish Experience, ed Inga Persson, (Oslo. Norwegian University Press, 1990) 5 Walter Korpi, "Faces of Inequality Gender, Class, and Patterns of Inequalities in Different Types of Welfare States," Soctal Politics 7, no I (Summer 2000) 127-91 6 Whether contributions are paid in by the employer or personally by those insured is, however, of no real significance It is still the employees who bear the cost in that their cash wages become lower in the long run than they would have been without the social insurance system See, for example, Lawrence Thompson, Older and Wiser The Economics of Public Pensions (Washington, DC. Urban Institute Press, 1998) 7 Per Johansson and Marten Palme, "Do Economic Incentives Affect 'ork Absence? Empirical Evidence Using Swedish MicroData,"3ournalofPublcEconomics 59(1996) 195-218,Jan Selen and Ann-Charlotte Stahlberg, "The Importance of Sickness Benefit Rights for a Comparison of Wages," Working Paper 1/2002, Swedish Institute for Social Research, Stockholm, available at 8. Pension experts usually distinguish between two pension plan types the defined-benefit (DB) and the defined-contribution (DC) pensions In a DB plan, a formula for retirement income based on the worker's wage and service is specified In a DC plan, benefits at retirement depend on the total contribution the worker has accumulated into the plan by retirement age 9 Ann-Charlotte Stlhlberg, "Life Cycle Income Redistribution of the Public Sector Inter- and Intragenerational Effects," in Generating Equality in the Welfare State The Swedish Experience, ed Inga Persson (Oslo Norwegian University Press, 1990) and Ann-Charlotte Stahlberg, "Women's Pensions in Sweden," Scandinavianjournal of Social Welfare 4(1995) 19-27 10 However, a family policy cannot solve the problems caused by the division of the labor force into low-wage female work and better-paid male work 11 Asa Forssell, Magnus Medelberg, and Ann-Charlotte Stahlberg, "Unequal Public Transfers to the Elderly in Different Countries-Equal Disposable Incomes," European Journal of Social Security 1, no 1 (1999) 63-89 Similar results can be found in OECD, "Maintaining Prosperity in an Ageing Society The OECD Study on the Policy Implications of Ageing," Working paper AWP 4.3 (1998) 12 Ann-Charlotte Stahlberg, "Pension Reform in Sweden," Scandinavian Journal of Social Welfare 4(1995) 267-73 CHAPTER 13 Gender, Demography, and Welfare State Restructuring in Japan by Ito Peng ABSTRACT: Gender equality was at the core of several fundamental social policy reforms undertaken by successive Japanese governments in the 1990s. The reforms expanded childcare and long-term care for the eld- erly. The reforms reflect a profound shift in policymakers' attitudes toward gender, breaking with their traditional emphasis on the model of male breadwinner and female housewife. The shift occurred because women mobilized politically during the 1970s and 1980s, and because of demo- graphic changes partly provoked by working women's reluctance to embrace married life. Japan now has the unenviable role of leading the way into uncharted waters of population aging and decline. Changing gender relations and demographic patterns are challenging the model of the male breadwinner-female housewife family, which shaped Japan's postwar welfare regime. Growing employment of married women has put a sharp focus on elder care and childcare and highlighted the difficulties women face in bal- ancing work and care obligations. The steadily declining fertility rates that so concern Japanese policymakers today are tightly linked to a gender structure that makes it impossible for people to reconcile work and family responsibilities. Many young people, particularly women, are deferring and even forgoing mar- riage and childbirth to pursue work and career, or because they fear marriage will constrain them to traditional gender roles and limit their freedom The state worries that the impact of declining fertility on aging will lead to declining eco- nomic growth, rising dependency ratios, growing care burdens for families and local governments, and eventual population decline The author thanks John Campbell and Andre Sorensen for comments on an earlier draft 215 216 i Policies to Combat Social Exclusion Responding to these issues, the Japanese government dramatically changed its social policies during the 1990s Acknowledging the reality of shifting gender relations, the state also took an active role in trying to reshape them. A vision of women as a central part of the productive labor force is replacing earlier expec- tations of women as dependent homemakers, primary caregivers, and peripheral members of the labor force, fostered by reforms in employment policies and the gradual expansion of care services. The state's aim is to free women and men from traditional gender roles and allow them to reconcile family and work responsibilities, thus encouraging higher birthrates Japan's welfare state restructuring in the 1990s reflects the collapse of old- style conservative politics dominated by the Liberal Democratic Party (LDP) and the reconfiguration of Japan's political economy.' The restructuring also reflects weakening bureaucratic control over the policymaking process stemming from growing politicization of social policy issues and the entry of participants such as women's groups into the policy debate. Ordinary Japanese now com- monly use the term "low-birth and aging society" (shoshi-koreika shakai), which did not exist before 1994, and the idea of lowv fertility and rapid demographic aging infuses nearly every social policy debate Social care policies now reflect a complex interdepartmental effort to address the social and economic challenges stemming from Japan's low-birth and aging society. The Japanese Welfare State: Factors Driving the Care Crisis Despite its expansion in the 1970s, the Japanese welfare state is still one of the leanest among the Organization for Economic Cooperation and Development (OECD) countries, ranking lust below the U.S. welfare state in total expenditures as a share of gross domestic product (GDP)-at 12 4 percent in 1980 and at 14.5 percent currently2 Pension and health care insurance accounted for 90 percent of these expenditures in 1999 3 Japan's low social security costs reflect the lim- ited public role in providing social services. Through the 1980s tight fiscal controls imposed under the state's new con- servative regime (which introduced the "Japanese-style welfare society")4 man- aged to curb growth in social security expenditures, which rose by only about 1 percentage point between 1980 and 1990 (from 12.4 to 13.6 percent of national income). This relatively modest increase, despite tremendous expansionary pres- sures from pensions and health care,5 reflected a shift in care responsibilities back to the family, rolling back many policy reforms initiated in the 1970s. The state abolished free medical care for the elderly introduced in 1973, for example, and Gender, Demography, and Welfare State Restructuring in Japan 1 217 instituted user fees. The state also "rationalized" the income assistance program (seikatsu hogo)-which had expanded modestly during the 1970s-by imposing tighter means tests. Tougher income tests similarly retrenched the child allowance program introduced in 1971, which had expanded rapidly in the first half of the decade. A special child-rearing allowance and other financial assistance to single mothers faced similar cutbacks throughout the 1980s The welfare state retrenchmcnt and policy emphasis on the family in the 1980s were not unique to Japan. Most European welfare states experienced sim- ilar policy shifts during that decade What distinguishes the Japanese welfare state from many others is Japan's much greater institutional reliance on the fam- ily. The welfare retrenchment of the 1980s clearly illustrates the state's assump- tion that the family was the appropriate place to sustain welfare cuts The results are evident inJapan's high level of co-residency and care dependency among eld- erly people and their children, the weakness of public and market care provision, and high dependency rates among unmarried adult children Co-residency between elderly people and their adult children in Japan is high compared to that of European welfare states. In 1980 nearly 70 percent of elderly people lived with their adult children To encourage and maintain co-residency, the government in the 1980s introduced policies to support three-generation households and family-based care, including tax credits for threc-generation households and allowances for family members caring for their elderly relatives Though co-residency rates have been declining since the 1980s despite these pol- icy efforts, more than half of the elderly continue to live with their children.6 In 1997, 41.7 percent of people over age 60 in Japan were living with married chil- dren, compared with 4 7 percent in Germany and 3.6 percent in the United States. Furthermore, 17.2 percent of Japanese elders lived with unmarried chil- dren, compared with 8 2 percent in Germany and 12.5 percent in the United States.7 Levels of family care are extremely high in Japan. In 1998 only 3.6 percent of elderly people lived or were cared for in institutions, even though some 13 per- cent needed such care.8 And in 1996 less than 1 percent of elderly pcople received community care, compared with 17 percent in Denmark and Sweden, 13 percent in the Netherlands, 8 percent in the Unitcd Kingdom, and about 8 percent in the United States.9 The National Basic Survey (Kokumin Seikatsu Kiso Chosa) in 1995 found that co-residing family members were caring for 86.5 percent of bedridden people, whereas families who were not living with their bedridden rel- atives looked after another 6.3 percent People who were not family members cared for only 7.2 percent of all bedridden elderly people.iO Interestingly, a recent study based on the National Household Survey (Zenkoku Setai Chosa) shows an 218 l Policies to Combat Social Exclusion inverse relationship between postretirement income and the co-residency rate, and a positive relationship between the level of disability and co-residency rate, highlighting elderly people's economic and care needs as important variables in postretirement co-residency decisions." Not only are families caring for a large proportion of elderly people in Japan, but also such care is partly preconditioned by a lack of public and institutional support. Public provisions for childcare services, in contrast, show better coverage. In Japan publicly funded centers care for about 22 percent of all preschool-age chil- dren.'2 (Though employment rates for married mothers with small children are higher in the United States than in Japan, only 1 percent of U.S. children age 0-2 years and 14 percent of U.S. children age 3-5 years are in publicly funded childcare ) Japan also compares well with continental European welfare states such as Germany and the Netherlands, where 2 percent of children 0-2 years of age are in public childcare. The Japanese figure, however, is significantly lower than the figures in Denmark (where about half of children age 0-2 and four-fifths of children age 3-5 receive publicly financed care) and Sweden (where one-third of children age 0-2 and 72 percent of children age 3-5 receive publicly financed care). 1 Public childcare in Japan faces significant problems. Childcare spaces still fall far short of needs. The number of children on official waiting lists rose from 26,114 in 1994 to 40,523 in 1997, as more mothers entered the labor market 14 Public childcare also largely serves older preschool children. Whereas 31.8 per- cent of 3-year-olds and 26.7 percent of 4-, 5-, and 6-year-olds are cared for in public childcare, the figures drop to 17.1 percent and 4.7 percent for 2-year-olds and children under 2, respectively. Economic expediency and institutional lega- cies reinforce this pattern; care for children under 2 is more expensive than care for older children because of greater care needs, higher worker-child ratios, and special care and dietary requirements. Many Japanese still subscribe to the "3-year-old myth," which admonishes mothers to stay home to care for their children during the first 3 years of their lives. Because most working women can take only up to a year of childcare leave, the lack of public childcare seriously affects anyone hoping to continue working Private sector childcare services are inadequate in Japan, as they tend to be unli- censed, uncertified, and expensive. They are also considered unsafe and are fre- quently rocked by scandals of accidental death and abuse leading to death."S Most parents avoid private sector childcare unless absolutely necessary and rely instead on family, particularly grandparents, to care for their children. The employment rate for mothers living in three-generation households with the youngest child Gender, Demography, and Welfare State Restructuring in Japan 1 219 age 0-3 is 41.4 percent, compared with 25.2 percent for mothers living in two- generation households.'6 Finally, the co-residency rate of young unmarried adult children and their parents is very high and has risen steadily. In 1996 approximately 80 percent of unmarried working adults age 20-29 lived with their parents (compared with 70 percent in 1975), whereas the proportion of those age 30-34 was about 70 per- cent (up from 50 percent in 1975). Nearly half of unmarried adults living with their parents received financial assistance from them, and most also enjoyed per- sonal services such as cooking, cleaning, and laundry provided by their moth- ers.17 Widening income disparities between young and older workers (and young adults' inability to achieve economic self-sufficiency),18 high housing costs, and lack of adequate income support and housing subsidies for young people explain cohabitation between adult children and their parents. i9 Those dependency patterns show how the family assumes a huge share of social security and personal welfare responsibilities in Japan and underscore how families' care burdens have increased rather than decreased over time. The fam- ily's capacity to care for its members reflects women's unpaid labor In Japan social policies and employment practices that discourage adult women's eco- nomic independence and marital dissolution reinforce women's family care role. The married women's tax exemption, which allows housewives to earn up to 1.3 million yen (some 20 percent of the average male annual wage) without tax lia- bility, encourages women to work only part time, whereas tax credits for three- generation households reinforce elderly co-residency and married women's care responsibilities. The lack of income assistance for divorced and unmarried single mothers, as well as punitive measures such as disqualification of divorced single mothers from income support for the first 6 months after divorce, discourage women from divorcing their husbands, particularly when their children are young A lack of social support and of effective equal employment legislation lim- its opportunities for women to achieve economic independence. For most of the postwar era, this welfare regime drew fairly wide public sup- port. Until the mid-1970s, feminists supported "housewife feminism" (shufu fem- inism) and actively pushed for social policies to strengthen full-time homemakers' positions rather than demanding gender equality.20 This stance reflected groxNing economic prosperity, except during the post-oil crisis years of 1974-75, as male wages that had expanded since 1955 relieved wives of having to work. 'he postwar Japanese welfare regime also opened opportunities for middle-class women to raise their social status as housewives,21 as women's political and social mobilization focused on improving women's social and economic status as housewives. 220 I Policies to Combat Social Exclusion However, as married women reentered the labor market after 1975 as the real value of the male wage began to decline, the issue of care for the elderly and chil- dren began to surface. By the 1980s the number of elderly people in need of care was considerable, and the lack of institutional and public home-care services had become glaringly obvious Imperatives for Change Changing gender relations in Japan were most apparent in the erosion of the male breadwinner-female housewife family, as more than half of married women were working by 1986 Families' need for a second wage earner ran counter to policymakers' expectations that the family would care for the young and the old. State assumptions about the family's social welfare role reflected the LDP gov- ernment's fixation on fiscal concerns and its clear underestimation of families' care burden. In 1981 the Second Provisional Commission on Administrative Reform (Second Rincho) sought to achieve fiscal restructuring by cutting expenditures, and social welfare was one of the first targets. Those welfare reforms provoked two different responses. One came from housewives-mostly middle-class women in their fifties-who faced the growing burden of holding a job while car- ing for aged parents and parents-in-law and often dependent children as well. Those women responded to welfare retrenchments by directly challenging the government's family policy. The younger cohort reacted to the neoconservative politics of the 1980s not only by dissociating themselves from political activities but also by consciously or unconsciously controlling their marriage and fertility patterns in ways that contributed to the later fertility crisis Fearing that the new welfare reforms would push family care burdens to the limit, in 1981 women's groups began to mobilize at the grassroots level to protest the government's top-down policymaking process. Groups such as the Women's Committee for the Improvement of the Aging Society (WCIAS) (Koreika Shakal wo Yokusuru Josei no Kai) called attention to the problem of women's double burden,22 and public support for these efforts proved surprisingly large. The WCIAS, which began as a gathering of 298 women in 1983, doubled its mem- bership to 600 within a year, and by 1989 had become a nationwide network, with regular appearances on national media The WCIAS undertook its own research and disseminated information on family care burdens This information highlighted the contradictions between the state's family policies and family realities. The evidence also highlighted the gender dimen- Gender, Demography, and Welfare State Restructuring in Japan 1 221 sions of family care Although women accounted for an overwhelming majority of caregivers (85 percent), they were more likely to be institutionalized when their turn came to be cared for, as evidenced by their high representation (80 per- cent) in public institutions.23 Men were either unwilling to assume their share of family care or were unable to do so because working conditions left them little time for family. The lack of political representation for women also underlined the need for women to engage in political action to push for their interests.24 The WCIAS brought together housewives and academics, social activists and social workers, and volunteers working with elderly people and people with dis- abilities on a common platform of care. Local chapters formed study groups and went on study tours to learn about long-term care policies and the condition of elderly people in other countries 25 Not since the 1960s had women's groups been so politically active. Their social and political impact was considerable: these groups helped change public debate and redefine women's interests regarding social welfare policies, particularly care for the elderly. Rising public and media interest in social care for the elderly helped the WCIAS gain credibility as a pub- lic voice for women and the family. As women's mobilization gathered momen- tum, academics, the media, and politicians joined to lobby for extension of social care Despite this movement, in 1982 the government enacted the Health Care for the Aged law, which replaced free medical care for those over age 70 with a co- payment system The law also introduced a cross-subsidization scheme among three public health insurance programs to cover the cost of medical care for the elderly and reduce the need to use general tax revenues to subsidize national health insurance.26 Notwithstanding these efforts, health care costs continued to rise. To raise revenue, the government introduced a 3 percent consumption tax in 1989 on the grounds that it was necessary for an aging society. Although women's groups had already severely criticized the LDP's family policy, the government chose to ignore their concerns, inviting an even larger political backlash. Voter revolt gave the LDP a humiliating loss in the Upper House election that year and gave a huge gain to the Japan Socialist Party USP), which ran on a platform of opposi- tion to the consumption tax To redress the situation, the LDP government introduced the Gold Plan (the Ten-Year Strategy on Health and Welfare for the Elderly) to justify the consumption tax and to show that it had heard public demand for more social care 27 This helped quash public anger, and in 1992 the LDP recaptured most of the seats it had lost to the JSP. Political fragmentation within the LDP in the early 1990s-which resulted in the formation of splinter groups, including the Democratic Party of Japan- 222 I Policies to Combat Social Exclusion offered another opportunity for women's and other interest groups to enter pub- lic debate and push for social care.28 Amid the economic transformation follow- ing the collapse of the bubble economy and a string of corruption scandals, the LDP gave way to a series of coalition governments.29 The first such government, headed by the reformist Prime Minister Morihiro Hosokawa of the Japan New Party, opened the policy window for the idea of long-term care insurance that had been circulating among community groups and social bureaucrats within the Ministry of Health and Welfare. Indeed, social bureaucrats had pursued long-term care insurance since the mid-1980s, closely studying Germany's debate over such insurance and forming a study group to examine the possibility of a comprehensive homecare program funded through social insurance.30 This bureaucratic interest resulted partly from pressure from the Federation of National Health Insurance Associations and major business organizations such as Keidanren (National Association of the Employers) to ease the financial burdens imposed by the cross-subsidization scheme imposed by the Health Care for the Aged Law. The Ministry of Health and Welfare was also keen to support a social insurance scheme because, unlike tax revenue, funds would go directly into its own coffers rather than to the Ministry of Finance.31 Throughout this period, women's groups continued to call for the right to social care and proposed a social insurance scheme to ensure that right.32 Thus, when the Hosokawa government expressed interest in a long-term care insur- ance scheme in 1994, social bureaucrats were prepared to seize the opportu- nity, and the window opened for women's groups to enter the policy debate. The WCIAS lost no time in linking its campaign for social care to the new pol- icy proposal. The policymaking process for long-term care insurance became even more permeable when the head of the WCIAS, Higuchi Keiko, was invited to join the Ministry of Health and Welfare's Council on Health and Welfare for the Elderly (CHWE), which approves proposals before they go to parliament. This kind of policy process is not new. Japanese policy bureaucrats have often been remarkably successful in co-opting powerful interest groups by using coun- cils, or Shingikai, to reach consensus. However, people invited to join these coun- cils tended to be well-known university professors and representatives of Keidanren or other economic and industrial leaders. The fact that bureaucrats invited the head of the WCIAS to join the CHWE attested to the group's legiti- macy and the bureaucracy's sensitivity to public opinion. The bureaucracy clearly stood to benefit from public support from women's groups. Gender, Demography, and Welfare State Restructuring in Japan 1 223 Cooperation between social bureaucrats and women's groups was most evi- dent when long-term carc insurance met unexpected opposition from conser- vative LDP members of a new coalition government in 1995 Realizing that the proposal for long-term care insurance was going nowhere, the WCIAS launched a campaign called the Ten Thousand Citizens' Committee to Realize Social Care (Kaigo no Shakaika wo Motomeru Ichiman'nin Shimin Iinkai) to galvanize public support. Drawing in support from the Liberal Democratic Party, the newly formed Democratic Party of Japan (DPJ), and other opposi- tion parties, the committee took its proposal to socialize care to the parliamen- tary floor. In 1997, when the LDP coalition government realized that it needed DPJ support to pass a health care insurance reform bill, it agreed to accept the long-term care insurance bill.33 Although this legislation did not include all the demands presented by the Ten Thousand Citizens' Committee to Realize Social Care, the WCIAS had significantly altered the course of welfare reform in Japan. Changing Gender Relations among Young People While housewives were taking on the state in fighting for social care, their younger cohorts pursued the routc of political disengagement Political partici- pation among men and women under age 40 dropped steadily throughout the 1980s. Rather than challenge structural blocks through social and political mobi- lization, younger Japanese women focused on fulfilling their personal and career aspirations, refusing to subscribe to the traditional life course The opportunities for young Japanese women in the 1980s were significantly different from those of the older cohort. expectations that women would marry and assume the role of full-time housewives had markedly declined. The Equal Employment Opportunity Law, which was introduced in 1986, did not guarantee women labor market equality; nevertheless, it gave them more employment opportunities and greater career prospects. Even so, labor market structures remained hostile to career women with fam- ily and care obligations. Japan introduced parental leave and care leave legisla- tion only in the 1990s, and companies did not integrate women into their career tracks in the 1980s despite the Equal Employment Opportunity Law. Meanwhile, women who wanted to marry found few ideal partners. Surveys show that although women look for men who will be good earners and accept more egalitarian relationships at home, most men prefer wvomen who will assume 224 I Policies to Combat Social Exclusion more traditional gender roles.34 Given such a lack of opportunity, women who wanted a career chose to postpone and even avoid marriage and childbirth, and those who wanted to marry found it difficult to find an ideal partner. The non- marriage rate for women between the ages of 25 and 29 rose from 20 percent to 50 percent between 1975 and 1995, whereas the average length of steady employ- ment for women grew from 6.1 years in 1980 to 8.2 years in 1996 35 The aggregate effect of individual decisions to postpone or forgo marriage was significant. In 1989 the total national fertility rate dropped to 1 57 per 1,000-a critical psychological point for Japanese.36 The "1.57 shock" set off a flurry of activity within the government as policymakers rushed to find solu- tions to the "quiet crisis." Social bureaucrats pushed researchers and academics to identify the causes of fertility decline and to come up with solutions to reverse or at least halt it.37 Research pointed to existing gender relations as the causes of fertility decline,38 specifically the high opportunity cost of having a child in the form of lost income39 and women's inability to reconcile family and work obligations. Gender relations were forcing young women to choose either marriage or a career, and many young women postponed the choice or chose the latter option. Social bureaucrats had assessed the European experience and were convinced that a family-work harmonization policy would be useful. The interests of the Ministry of Health and WVelfare's Child and Family Bureau, which had tried in vain to raise the profile of child welfare from another angle, intersected with those of the ministry's Population Bureau, paving the way for interbureau con- sensus. European experience shows that throughout the 1980s women-friendly welfare states such as Sweden were successful in raising or maintaining the fer- tility rate, whereas family-centered welfare states like Italy were experiencing fer- tility declines even faster than Japan's. The policy implications were obvious unlike the situation in the 1960s, child rearing and women's paid work in the 1990s were complementary rather than contradictory,40 and countries that sup- ported wvomen's work and family harmonization had a better chance of prevent- ing fertility declines. In 1994 the Japanese government introduced the Angel Plan (the Emergency Five-Year Plan in Response to the Declining Fertility). Supported by five differ- ent ministries, the plan aimed to expand childcare services and introduce employment legislation to help women harmonize work and family responsibili- ties. Thus, in an entirely different way, young women had managed to reset the state's social policy agenda in the 1990s In this case changes in individual behav- ior had become a powerful stimulus prompting the government to act. Gender, Demography, and Welfare State Restructuring in Japan 1 225 The combination of women's political mobilization and the fertility crisis sig- nificantly reshaped political discourse and reconfigured political and institutional interests in Japan in the 1990s. From a focus on strengthening the family's tradi- tional role, policy debate had shifted to how to save the family (that is, women) from collapse, and how to provide incentives for young people to get married and have children By the beginning of the 1990s, even the most conservative LDP politicians were shying away from advocating the Japanese-style welfare society. The government had also abandoned its attempt to promote three-generation households, and instead began to consider married women's employment as a norm rather than an anomaly. The fertility decline also redefined the long-term economic and social impli- cations of the aging society. Current projections estimate that the proportion of people over 65 will double from 15 6 percent to 32.3 percent by 2050. At this rate, even with higher employment rates among women and elderly people, some ana- lysts suggest that Japan will face serious labor shortages after 2005. Analysts also expect a sharp rise in dependency ratios-from I elderly person to 4 41 workers today to I elderly person to 1.7 workers by 2050.41 The government estimatcs that the social security burden will rise from 18.9 percent of national income in 1998 to 29.5-35 5 percent by 2050.42 This social insurance burden will siphon off more than 50 percent of national tax income-a figure that puts fear in the mind of many Japanese politicians, business leaders, and economists 43 Demographic changes will gradually undermine economic growth rates because labor productivity will decline and total capital assets will1 erode as more retired people begin to use up their savings Forecasts indicate that average eco- nomic growth rates will drop to about 1.8 percent in 2010 and to about 0.8 per- cent by 2025 44 High longevity and low fertility rates will worsen economic con- ditions over the next 50 years because Japan's purist (non) immigration policy seriously limits the government's options. The decline in the fertility rate and rise in one-child families will bring sharp increases in households composed of elderly couples or elderly people living alone The social impacts of these demographic changes are already apparent in rural areas, where migration of young people to large cities has aggravated popu- lation aging. By 2025 nearly 60 percent of all communities in Japan are expected to have over a third of their population over age 65, compared with 10 percent today, with obvious implications for care burdens for the family and for local gov- ernments.45 Finally, demographic projections suggest that, given current fertility rates, Japan's population will reach a peak around 2007 and thereafter decline steadily 226 I Policies to Combat Social Exclusion to about half the current level by the end of the 21st century.46 This will have serious cultural and psychological impacts, undermining Japan's status as a major economic power as well as its influence in international politics 47 Social Policy Reforms since 1990 Japan's social welfare policies in response to gender and demographic impera- tives in the 1990s shifted responsibility for personal care services noticeably, as the state took on a larger share and the market became an important player in the social care market The two major social welfare reforms-long-term care for the elderly and family-work harmonization (or kazoku to shigoto no ryoritsu)- illustrate these changes well. Social Care for the Elderly The Gold Plan, introduced in 1989, was the first comprehensive policy frame- work for institutional and community-based care for the elderly The plan not only expanded services but also shifted responsibility from central to local gov- ernments. Even though local governments were overwhelmed by demand for services almost from the start, the New Gold Plan, introduced in 1994, signifi- cantly expanded 5-year targets (see Table 13 1). The unexpected demand pro- voked much public criticism of the fact that almost all services were means tested and thus excluded most elderly people and that they were inadequate for low- income elderly people living alone or without family support. The lack of public social care resulted in high personal costs and sacrifices. In 1995 some 90,000 women left the work force to care for aging relatives, and many others tried to manage care for aging parents and elderly relatives while continu- ing to work.48 Despite the Gold Plans, female relatives still cared for 85 percent of elderly people 49 Problems with the Gold Plans help explain why women's groups supported the long-term care insurance scheme. That insurance, implemented in April 2000, transformed care for the elderly from a needs-based system to a universal rights- based scheme. Long-term insurance covers care for people over age 65 who are deemed to require it, as well as care for people age 40-64 who have aging-related disabilities such as Alzheimer's disease. The insurance covers both community- based and institutional care, with coverage based on the level of disability. Insurance contributions are compulsory for people over 40. In principle, with the introduction of this insurance, all elderly people requiring care will have the right to receive it Gender, Demography, and Welfare State Restructuring in Japan 1 227 Table 13.1 Summary of Gold and New Gold Plan Targets Gold Plan New Gold Plan Target activities or services 1989 figures target (1994) target (1999) Number of home helpers 31,405 100,000 170,000 Number of spaces for day services 4,274 10,000 17,000 Number of beds in short-stay centers 1,080 50,000 60,000 Number of community care support centers n a 10,000 10,000 Number of visiting nurse stations n a n a 5,000 Number of spaces in special chronic care homes 162,019 240,000 290,000 Number of elderly health care institutions 27,811 280,000 280,000 Number of spaces in care houses 200 100,000 100,000 Number of elderly welfare centers n a 400 400 Number of new community careworkers n a n a 200,000 Number of new nurses and nursing careworkers n a n a 100,000 Number of new occupational and physical therapists n a n a 150,000 n a = not available Source Ministry of Health and Welfare, Elderly Welfare Bureau Family-Work Harmonization Policy In tandem with policy reforms in long-term care, the family-work harmonization policy proceeded rapidly in the 1990s Like the Gold Plans, the Angel Plans aimed to broaden public support for families with young children by expanding childcare and other services, and by overhauling the public childcare system (see Table 13.2). In addition to expanding childcare services, the Angel Plan broadened the child allowance from 3 to 6 years and raised the income cutoff so that some 80 percent of families with small children qualify.50 New employment laws imple- mented since 1994 also help working mothers. Paid parental leave introduced in 1994 allows parents to take up to I year of unpaid childcare leave with job guar- antees In 1998 legislation added an income replacement equivalent to 25 percent of salary and the costs of social insurance, and in spring 2000 the income replace- ment was further raised to 40 percent. Family care leave introduced in 1996 allows workers to take up to 3 months of unpaid care leave, and in 2000 an income replacement of 25 percent was added. Together these reforms have significantly expanded social care and support for families with small children and reflect efforts to encourage women's labor market participation They also reflect public recognition of the enormous care 228 I Policies to Combat Social Exclusion Table 13.2 The 1994 Angel Plan and 1999 New Angel Plan Target activities or services 1994 1999 2004 target Number of childcare spaces for children age 0-2 years 470,000 580,000 680,000 Number of extended-hour childcare centers 2,530 7,000 10,000 Number of childcare centers operating during weekends and holidays n a 100 300 Number of temporary childcare centers n a In 450 local In 500 local communities communities Multifunctional childcare centers 200 1,600 2,000 Childcare support centers for stay-at-home mothers 354 1,500 3,000 Temporary childcare support for stay-at-home mothers 600 1,500 3,000 After-school programs for children in elementary school 5,220 9,000 11,500 Treatment and counseling centers for infertile couples n a 24 47 n a = not available Source Ministry of Health and Welfare, Heisei 12-nendo Kosei Hakusho(1999) burdens on women imposed by existing gender relations. The result was that social welfare expenditures nearly doubled between 1990 and 1998-from 4,799 billion yen to 8,323 billion yen, compared with an increase from 3,999 billion yen to 4,799 billion yen between 1981 and 1990 51 Expanding Social Care through Devolution and Deregulation The expansion of social care occurred along with devolution of state welfare responsibilities and deregulation of social services.52 Administrative responsibility for the Gold Plans, the long-term insurance scheme, and the Angel Plans all devolved to local governments, while the central government redefined itself as the coordinator and planning body within the new institutional arrangement. Although local governments now theoretically have more autonomy to respond to local social welfare needs, they are also subject to tremendous fiscal pressures because they cannot raise local taxes without central government approval The expansion of social care responsibilities without the ability to raise their own rev- enues has pushed many local governments to near bankruptcy.53 For local govern- ments in rural areas, where economic growth is negative and demographic aging is more advanced than in urban centers, devolution comes as a harsh measure. Gender, Demography, and Welfare State Restructuring in Japan 1 229 Deregulation of welfare services has also put pressure on public sector service providers. The 1996 child welfare reform replaced the system (sochi seldo) that gave the state total control over placing children in public childcare centers with an individual contract system. This allowed parents to choose among childcare centers and services, and forced public childcare providers into market competi- tion. Outcomes have been mixed. Although more childcare centers are providing longer hours of service, many state-run childcare centers have been forced to close and others to use part-time and contract workers to cut personnel costs, raising concerns about the quality of care.S4 Long-term care insurance operates in a similar manner. Although local gov- ernments are responsible for administering insurance, private, voluntary, and other tertiary agents provide care services on a competitive basis Since its implementation in 2000, market competition has not worked well, and elderly people and families do not necessarily receive more or better care. Criticisms focus on computerized methods of measuring disability that are inaccurate and screen out many people who need care, huge regional variations in services that contradict the principle of a universal right to care, and a 10 percent surcharge on services blamed for "care-shy" behavior among users.S5 The valuation of care services under the long-term care insurance scheme is also seriously flawed. Because the government presets the costs of all services pro- vided under the insurance, care providers set their rates accordingly. However, because the price of home-help services (including cleaning, cooking, shopping, and laundry) is about half that of physical care services, users tend to rely on home helpers to provide physical care while they clean the house. As a result of deregu- lation and changes in government roles, most careworkers have been laid off from local government jobs and often work for lower pay in the private sector.56 In sum, although decentralization and deregulation of social care can allow a better mix of services that meet individual needs more efficiently, such reforms raise questions regarding the capacity of local government to provide care and the wisdom of shifting public and private boundaries. Lessons from the Japanese Experience Japan's policy reforms of the 1990s suggest some interesting lessons. First, expe- rience clearly shows the critical role of gender relations and demographic changes in shaping social policy today Before 1990 policymakers focused on industrialization and urbanization as the cause of disintegrating families and 230 I Policies to Combat Social Exclusion communities, and policy responses aimed to reduce such fragmentation by rein- forcing traditional gender roles. The new perspective, embodied by a recent Ministry of Health and Welfare white paper,57 puts the blame for family and community disintegration squarely on traditional gender relations and seeks to redress them by reforming social and labor market policies This suggests an important shift in the state's position in relation to women and the family. Second, the political and demographic imperatives facing the Japanese wel- fare state illustrate what happens when traditional care and welfare responsibili- ties overburden families. The policy response suggests a dynamic relationship between the Japanese welfare state and its people, and calls attention to the importance of examining the interaction of social, economic, and political fac- tors in policy change. This experience is not Japan's alone. Similar shifts in gender relations and demography are occurring in other welfare states in the East and West. As the postindustrial economy draws more women into paid employment, the state will have to pay critical attention to tensions between changing gender relations and institutional arrangements that reflect the postwar male breadwinner-female housewife family model. Such tensions are particularly apparent in a family- centered welfare regime such as Japan's, but other conservative welfare states, including Germany, Italy, Portugal, and Spain, share similar concerns. Fertility rates in those countries are among the lowest in the European Union, suggesting that young people in those countries may also be realizing that they cannot afford to have children. Throughout its development, the Japanese welfare state has learned and adapted ideas from western welfare states. Japan is now in the unen- viable position of leading the way into uncharted waters regarding population aging and decline. The actions of the Japanese welfare state in the face of gender and demo- graphic pressures, although clearly significant, have yet to show the desired results in reversing or even slowing fertility decline. Policy reforms have largely focused on relieving women of undue care burdens, but evidence suggests that the basic structure of gender relations remains unchanged and that women have not altered their fertility decisions. The state continues to define the prob- lem in terms of adverse effects on fertility rather than in terms of gender inequality. More positive efforts to facilitate changes in gender relations are still needed. Efforts to combine expansion and devolution of the welfare state, while cer- tainly not unique to Japan (consider, for example, welfare reform in the United States, which expands services at the state and local level), merit more research and debate. In Japan researchers often discuss the two issues separately. But Gender, Demography, and Welfare State Restructuring in Japan 1 231 together they increase the complexity of the challenge, and the two processes have interacted both positively and negatively. For example, although expansion and devolution produce more services that respond better to community needs, those trends can also lower the quality of care and exacerbate regional dispari- ties. And ironically, devolution and deregulation under Japan's expanded system of social care, which seeks to encourage wvomen to work by relieving them of care burdens, appear to have produced a newv system that employs women as low- wage, part-time, and contract workers. Notes I T J Pempel, Regime Shift Comparative Dynamnic oJ'the apanese Pohlical EronomiyR (Ithaca, NY Cornell University Press, 1998) 2 Between 1970 and 1980, social security expenditure rose sevenfold, from 3,524 billion yen to 24,763 billion yen, or from 5 7 percent to 12 4 percent of national income 3 National Institute of Population and Social Security Research (2000) 4 The conservative welfare regime of the 1980s is often referred to as the "Japanese-style welfare regime" (Nihnngata Fikinshi Shakai Regtime), a term first coined by the ruling conservative Liberal Democratic Party in 1979 in its policy paper calling for welfare policy rcview In the 1970s it became the LDP's platform to steer the Japanese welfare statc toward retrenchment and attack (XVestern-stN le) welfare state expansioni Redefining Japan as a welfare society, not a welfare state, the policy called for a new regime based on "individual self help and mutual aid between fami- lies, neighbors, and local community, and on the selective provision of public welfare by efficient state in accordance with the principle of the liberal cconomic society " See Economic Planning Agency, Shmn-Keizaishakai 7-kanien Keikaku (New Economic Society 7-Year Plan) (Tokyo Economic Planning Agency, 1979), p I) 5 Pressure on pension and health care spending stemmed from the maturing of pension schemes and population aging The proportion of people over age 65 rose from 9 1 percent in 1980 to 12 1 percent in 1990 See AMinistry of Health and Welfare, Ileivet l0-ieendo Kosei lIaktusho (1998 White Paper on lIea lth and Welfare) (Tokyo Gyosei, 1998) 6 Ministry of Health and Welfare, Ileisei 12-nendo Kose Ilakusho (2000 Whitte Paper on llealth and Welfare) (Tokyo Gyosei, 2000) 7 Office of the Prime Minister, Anniual Report on the Family Intcome and Eapenditure Survey (lokyo Statistics Bureau, Management and Coordination Agency, 1997) 8 Campbell and Ikegami argue that if we account for "social hospitalization," the level of institu- tionalization for elderly people would come closer to 6 percent This underscores the limited options for institutional care for the elderly Because public institutional carc is scarce and means tested, and because private institutional care facilities are lacking and expensive, elderly people in Japan must rely on their families much more than those in other countrics Social hospitalization is therefore a logical option for middle-class elderly people who cannot rely on their families for care See John C Campbell and M Ikegami, "Long-Tcrm Care Insurance Comes to Japan," Health Affairs 19, no 3 (2000) 26-39 A survey by the Association of WVomen and Youth indicates that a majority of people claimed that thev hospitalized family members at some point to meet their care needs See Fulin Shoseinen Kyokai, Shigoto to Kaigs to no Ryoritsu ni Kansuru Chosa 232 1 Policies to Combat Social Exclusion (Survey on IYork and Care Harnnoization) (Tokyo Fujin Shoseinen Kyokai, 1996) The national survey also found that about 41 percent of people have used private hospitals and that 37 percent of people have used public hospitals to meet family care needs, vwhereas about 11 perccnt of people have used home helpers and 3 6 percent have used public institutions for the aged See Ministry of Labor, Hatarakujose ?iojttszjo (Tlhe Condition of Working Women) (Tokyo 21-seiki Shokugyo Zaidan, 1998) 9 Diane Gibson, Aged Care Old Polcjes, New Problems (Singapore, Cambridge, U K Cambridge University Press, 1998) 10 Ministry of Labor (1998), op cit 11 Fumio Funaoka and Mitsuaki Avusawa, "Koreisha no Dokyo no Kettelyoin no Bunseki kazoku no seikatsujokyo to hosho kino" ("Analysis of the Deciding Factors for Co-residency for the Elderly People"), in Kazoku-Setas no Henyo to Se:katsu Hosho Kino (Family and Household C/ia uges and the Social Secirity Function), ed National Institute of Population and Social Security Research (TIbkyo University of Tokyo, 2000) 12 Ministry of Health and NVelfare (1998), op cit 13 Marcia K Meyers, Janet C Gornick, and Katherin E Ross, "Public Childcare, Parental Lcave, and Employment," in Gender and Wfe/fare State Regimes, ed Diane Sainsbury (Oxford, England Oxford University Press, 1999), pp 117-46 14 Ministry of Health and XVelfare (1998), op cit Unofficially, some 150,000 children are waiting for childcare space, according to Prime Minister Jun'ichiro Koizumi's campaign speech on July 14, 2001, NHKNational News 15 For example, see the Japan Times, "'Baby Hotel' Foibles Vex Working Mums Shortage of Trustworthy Day Care Leads Some to Abandon Plans to Have Kids," June 7, 2001, p 3 16 Ministry of Labor (1998), op. cit , and Nobuhiko Maeda, Shigoto to Kate,se,katsu no Chowa Nihon, Oranda, America no Kokusai Hikaku (WVork and Family Life Harmonizationl A Japan, Netherlands, and US Coinparison) (Tokyo Nihon Rodo Kenkyu Kiko, 2000) Other studies note similar patterns 17 Ministry of Hlealth and Welfare (1998), op cit 18 Kazuo Seiyama, "Shotoku Kakusa wo Do Mondai ni Suruka Nenreisonai Fubyodo no Bunsekikara" ("How to Problematize Income Disparity From the Analysis on Interage Group Inequality"), Kikan Kakei Keizai Kenkyu (journal of Family Economic Researc/i) 51 (Summer 2001) 17-23 19 Mami Iwakami, "Shoshi-Koreika Shakai no Oyako Kankei 30dai wo shoten ni shite kangaeru" ("Parent-Child Relationship in the Low Fertility and Aging Society vWith a Focus on People in Their Thirties"), Seikatsu Kyodo Kumia: Kenkyu 2 (2000) 19-26, and Michiko Miyamoto, "Ban-kon, Hikon Sedai no Chokumen suru mono Parasite Single no Airo" ("The Issues Faced by the Late Marrying or Nonmarriage Generation 'I'he Path of the Parasite Single," Kikan Kakei Keizai Kenkyu 47 (Summei 2000) 28-35 20 Sakiko Shiota "Gendai Feminism to Nihon no Shakai Seisaku, 1970-1990" ("The Contemporary Feminism and Japan's Social Policy, 1970-1990"), inJoseigaku to SeiyjiJssen (Political Iniphcations of Womnen's Studies), ed Society for WNFomcn's Studics Japan (Tokyo Keiso Shobo, 1992), pp 53-70 21 Before the mid-1950s employment rates for married women in Japan were high, most worked on farms or in family businesses Throughout the era of high economic growth, many women still had to work for a living Single mothers, women in low-income houscholds, and women with family businesses continued to work, poor women bore a heavy burden from the lack of welfare provisions, even though the popular image of married women was that of full-time housewives Gender, Demography, and Welfare State Restructuring in Japan 1 233 22 Keiko Higuchi, "Kaigo no Shakai-ka wo Koshitc Susumeta" ("How XWc Promoted Socialization of Care"), in Clh,,ki dle Mirn,A Minna de AMltoro 7oseiga Suummeru Kaigo no Shakaika (Everyone in the Community Takes Care of the Elderly WAomen Promoting Socializatioll of Care), ed Keiko Higuchi (Kyoto Minerva Shobo, 1997), pp 136-70 23 Ibid 24 Koreika Shakai vVo Yokusuru Josei no Kai, ed , Tomodaore ka Tomodachi Shakal he (Together We All Collapse or Together WVe Support Together) (TokNo Akashi Shoten, 1998) 25 lbid 26 Health insurance in Japan consists of three main programs employee insurance for workers and their depeindent family members through large, medium, and small firms, schemes organized by mutual aid societies for employees in public service, private schools, seamen, and dependent fam- ily members, and national health care insurance for the self-employed, retired, and those who do not fall into the first two categories Not surprisingly, since employee health insurance does not cover most rerired workers, they automatically sN^ itch to national health care insurance upon retirement Although the government subsidizes about half the cost of national health care insur- ance from general tax revenues, the high proportion of elderly people in this insurance scheme has produced a serious financial crisis 27 Campbell, for example, argues that it was crucial for the government to introduce a welfare reform policy for the aging society, as only then would voters support any new tax measure See John C Campbell, Howr Polocies Change. Thie Japanese Government and the Aging Society (Princeton, NJ Princeton Uisiversity Press, 1992) 28 Murase Mlikiko Lto, "The Establishment of Long-Term Care Insuirance," in Powet Shujiles and Polity Protesses Coalition Go,vern,ment in Japan tit the 1990, ed Hidec Otake (Tokyo Japan Centre for International Exchange, 2000), pp 21-50 29 Penipel (1998), op cit 30 Eto (2000), op cit 31 Campbell (1992), op cit 32 Murase Alikiko Eto, "Public Insolhement in Social Policy Reform Seen from the Perspective of Japan's Elderly-Care Insurance Schemc," Journal of Social Policy 30, no 1(2001) 17-36 33 For a more detailed discussion, see Eto (2001), op cit , and Kunihiko Ushiyama "Shakaiundo no Atarashii Tenkai to Seisakukatei kaigohoken nio scidoka to Kaigo no Shakaika wo Miotomeru Ichiman'ninshimin linkai" ("New DcOclopment in Social Movement and Social Policy Long- Term Care Insuran,c and the 'T'cn 'fhousand Citizens' Committee to Realize Social Care Campaign") in Shakai Undo Kenk)lu no Shindoko (New Developienits in Sowial AMovement Research), ed Shakai Undoron Kenkyukai (Tokyo Scibundo, 1999) 34 MImisti y of Health and Welfare (1998), op cit 35 Ministry of Labor (1998), op cit 36 The figure 1 57 is critical becausc it marks the historical low Fertility rates had previously dipped sharply and temporarily to 1 58 per 1,000 woomen (total fertility rate) in 1968 bccause that year was considered a particularly ominous year for childbirth, symbolized by the fire-breathing horse of the Chinese astrological calendar 37 1 wvas a member of the Ministry of Health and Welfaie's Special Research Commission on Issues of Family and Low Birth (Katei-Shussanl Alondai Sogo Chosa Kenkynu), which looked into this problem in 1992 38 Eiko Nakano and Yoshikazu WVatanabe, "Mikon Danjo no Kekkonkan" ("The Views of Marriage among Unmarried Youths in Contemporary Japan"), Jtnko Moudal Kenkyul (Journal of Population Problems) 50, no 3 (1994) 18-32, and Makoto Atoh, Shigesato Takahashi, Eiko 234 I Policies to Combat Social Exclusion Nakano, Yoshikazu Watanabe, Hiroshi Kopima, Ryuichi Kaneko, and Fusami Mito, "Dokushin Seinenso no Kekkonkan to Kodomokan" ("Attitudes toward Marriage and Fertility among the Unmarried Japanese Youth"), Jinko Mondat Kenkyu (journal of Population Problems) 50, no 1 (1994) 29-49 39 Katsura Maruyama, "The Cost Sharing of Child and Family Care Leave," Review of Populaton and Social Pohly 8 (1999) 49-74 40 OECD, Caring World The New Social Polc)y Agenda (Paris OECD, 1999) 41 Ministry of Health and Welfare (1998), op cit 42 Ibid , and National Institute of Population and Social Secunty Research (2000), op cit 43 Interestingly, national polls indicate that Japanese people appear to prefer increases in special insurance and fiscal contributions if they are assured that social welfare benefits will continue 44 Ministry of Health and Welfare (1998), op cit 45 Ibid 46 Ibid 47 Noriyuki Takayama, The Morning After in Japan Its Declining Population, Too Generous Pensions and a Weakened Economy (Tokyo. Maruzen, 2000) 48 Statistics Bureau, Management and Coordination Agency, Employment Status Survey, (Tokyo MCA, 1997) 49 Ministry of Health and Welfare (1998), op cit 50 The previous child allowance had a fairly low income cutoff, and as a result only about 60 per- cent of all families with children under the age of 3 qualified 51 National Institute of Population and Social Security Research (2000), op cit Social security expenditure in Japan is broadly divided into pension, health, and social welfare, this amount only accounts for the social welfare portion of the total social security expenditure Although social welfare accounted for just 11 5 percent of total social security expenditures in 1998-with pen- sions and health care insurance accounting for the other 88 5 percent-social welfare expendi- tures rose faster 52 Currently the cost burden ratio between the central and local governments for most social care services is 50 50 53 In 1999 total local government debt was estimated at 176 trillion yen, or 35 4 percent of the gross domestic product A growing social movement during the 1990s, led by social policy experts, economists, and community activists, called for greater local autonomy 54 Ito Peng, "A Recent Childcare Reform in Japan," in Family Pohcy and Childcare, eds Thomas Bole and Arnlaug Leira (London Routledge, 2000), pp 175-205 55 Long-term care msurance allows individuals to "purchase" a designated amount of care services each month, depending on their level of disability Insurance (government) presets the cost of each service However, users must pay 10 percent of this cost, the logic being that this makes serv- ices more transparent The 10 percent surcharge has forced people with low incomes to limit the use of care services The government's own survey has shown that demand for care services was much lower than initially expected because people shied away from using them because of the financial burden 56 Before long-term care insurance, local governments employed most careworkers in Japan Interviews with local governments and social welfare associations in Yokohama Cit) and in Iwate and Hyogo prefectures, and with careworkers in the Yokohama chapter of the local government employees' union, May-December 2000 57 Ministry of Hlealth and Welfare, Heisri Il-nendo Kosei Hakusho (1999 White Paper on Health and Welfare) (Tokyo Gyosei, 1999) BOX 3.3 Social Exclusion Unit of the United Kingdom Government by Tara Karacan Remedies to issues of social exclusion are as varied as its definitions Continental European welfare states have long wedded economic poli- cies to social policies, whereas Anglo-Saxon countries traditionally conceived of poverty primarily as a social problem to be addressed out- side the market. For example, the United Kingdom has a Social Exclusion Unit (SEU) incorporated in the Cabinet Office under the deputy prime minister It was set up by the prime minister in December 1997 to help improve government action to reduce social exclusion by producing "joined up solutions to joined up problems."'' It works mainly on specific projects, chosen by the prime minister following consultation with other ministers and suggestions from interested groups The SEU is staffed by a mixture of civil servants from a number of government departments and external secondees from organizations with experience in tackling social exclusion. The SEU works on issues that affect a range of government departments, though it does not dupli- cate work being done elsewhere It also participates in work that has a close bearing on social exclusion elsewhere in government. The SEU works on specific projects and has reported on five main areas: * Youth 16-18 years old who are not in education or employment * Teenage pregnancy * People who sleep in public places * Truancy and school exclusion * The National Strategy for Neighborhood Renewal The unit is currently working on four projects: * Reducing reoffending rates among former prisoners * Working with young runaways * Raising the educational attainment of children in care * Addressing transport and social exclusion continuied on Pne.% t page 236 I Policies to Combat Social Exclusion BOX 3.3, continued What is defined as social exclusion in one country may be defined simply as poverty in another country, and so on. For example, ethnic exclusion is a major issue in many countries, but it is not on the agenda of the UK Social Exclusion Unit. Because of the varied character of social exclusion, policies to deal with social exclusion clearly cannot be generalized. 1 See for an introduction BOX 3.4 The French Health System: Where It Stands and Where It Is Heading by Louis-Charles Viossat Rated number one worldwide in the 2000 World Health Organizationz Report, I the French health system would appear to offer excellent cause for patient-and government-satisfaction. Indeed, the French population enjoys one of the longest life expectan- cies worldwide and has excellent access to medical services and pharma- ceuticals, both in ambulatory care and hospitals. This is because of universal health insurance coverage, more than 100,000 medical practi- tioners and specialists, and an extensive and well-developed network of private and public local and university hospitals. Also of great impor- tance are the high quality of medical training, the education system over- all, and interesting innovations in evaluation practices. The truth, however, is that French health system reform has been a headache for every government since the end of the 1970s. It is worth recalling that conservative governments repeatedly lost general elec- tions-in 1981, 1988, and 1997-after implementing health reform plans that caused doctor and nurse strikes and demonstrations in the streets of Paris. The government that was elected in May and June 2002 is, once again, faced with the underlying difficulties of health system reform. To deal with this challenge, the new health minister, Jean-Francois Mattei, a pro- fessor of medicine, has adopted a multipronged approach that is based on the following: * Reinforcing the importance of public health. The French system is rich on curative care but poor on preventive care. This move- ment will be put into effect through a 5-year health program and priority legislation to be voted on in 2003. * Decentralizing health care providers and programs. The govern- ment plans to pass a law that will create 22 regional health agen- cies, to be in charge of hospital and ambulatory care planning, financing, and supervision. continued on next page 238 I Policies to Combat Social Exclusfon BOX 3.4, continued * Improving the governance of health insurance funds that are, as in Germany, managed by representatives of employers and employees * Pushing for quality and excellence everywhere in the health sys- tem by accreditation, evaluation, and training programs. * Reforming hospital management by streamlining hospital organi- zation and procedures and unifying financing techniques. 1 World Health Organization, World Health Report 2000 iGeneva World Health Organization, 2000), available at , last accessed October 10, 2002 BOX 3.5 Seminar on Social Capital, Social Exclusion, and the East Asian Crisis, Manila, November 5-7, 2001 This seminar was held in Manila at the headquarters of the Asian Development Bank (ADB), not long after September 11, 2001. This colored discussions of social exclusion, inequality, and poverty. The pro- ceedings were among the most thought provoking and engaged among the project seminars, and a central conclusion vas that such topics have urgent national, regional, and global dimensions. Ambassador Howard Q Dee (of the Philippines) and Colin Moynihan (of the United Kingdom) focused on exclusion as a phenomenon that could affect majorities of societies, with the critical barometer being youth and their sense of inclusion, commitment, and hope. Social exclusion is a specific problem for specific groups but can also apply much more broadly to the tenor of a society and to one country or society relative to others. An introduction by Akira Seki of the ADB highlighted the impor- tance of socially inclusive development, combining effective macro- economic governance with pro-poor, sustainable economic growth. Economic growth, he said, can rcduce poverty only when it forms part of a comprehensive national strategy that includes the development of human capital, provision of basic social services, removal of gender-based obstacles to inclusion, and expansion of social protection. Anita Kelles- Viitanen of the ADB highlighted the importance of Europe's social pol- icy experience in addressing social capital and social exclusion. The seminar was organized around seven topics. Jan Breman (Uni- versity of Amsterdam) focused on two villages in rural Java, Indonesia. When the urban economy crashes, many workers become much poorer than the "myth of the peasant society" would lead us to believe, because there is no longer any land to go back to and no agricultural work to compensate for the loss of urban jobs. Gareth Api Richards (see Box 3.1) focused on the relevance of social capital in the field of development assistance, and participants agreed that proper concep- tualization and measurement of social exclusion and social capital can continted on next page 240 I Policies to Combat Social Exclusion BOX 3.5, continued add to understanding of poverty and enhance the effectiveness of social policy. The second theme, regional strategies to fight social exclusion, drew on a presentation by Gabriella Battaini (see Chapter 27). European wel- fare states face serious challenges: aging, declining interest in collective action, and a trend toward fragmentation-all feeding the fear of a two- tier economic system. To address these challenges, public authorities could promote social partnerships, facilitate access to basic social rights, and fight exclusion on the basis of gender and race. Discussion focused on how to implement rights. Participants underlined the constructive role of the European Commission in peer-reviewing social policies at the European level. The third topic focused on the urban poor through the work of Vincent Delbos on urban policies in Europe (see Chapter 10) and of Soren Villadsen on the role of the state in providing housing for the poor (see Chapter 9). The case of Hong Kong figured prominently although self-reliance is at the heart of its welfare programs, 50 percent of its people live in public housing, and the government offers "interim hous- ing arrangements" for the elderly and devotes 15 percent of public expenditures to social housing. The fourth topic wvas the rural poor, grounded in Luis Frota's paper on social assistance for aging rural households (see Chapter 17). Discussion focused on whether the universal, pay-as-you-go pension system in Europe is suitable for East Asia The politics of pension reform, especially in Scandinavian countries, where farmers' parties have actively participated in the policymaking process, was highlighted. The fifth topic, social exclusion and gender, was based on a paper by Ann-Charlotte Stahlberg on the gender impact of the welfare state (see Chapter 12). Discussion turned on the role of women in societies in change. In the transition from public to private sector-dominated economies in East Asia, women and low-skilled workers are the first to be laid off and the last to be hired Pensions and social security schemes can play a critical role in preventing serious gender inequality. Women still benefit less from such schemes, even though in Europe they repre- sent 70 percent of retirees. Seminar on Social Capital, Social Exclusion, and the East Asian Crisis, Manila, November 5-7, 2001 1 241 BOX 3.5, continued The sixth topic, the informal sector, was based on a paper by Jacques Charmes (see Chapter 18). Discussion highlighted the elusiveness of the notion of the informal sector, accentuated by the urban bias of policy- makers and the formal sector bias of economists and statisticians. The seventh topic compared social exclusion in Europe and East Asia, with presentations by Isabel Ortiz of ADB on East Asia and Colin Moynihan on Europe. A broad conclusion was that many social protec- tion instruments today are not effective, owing to their limited coverage and generally low levels of public expenditure. Discussion focused on the need for greater coordination among government policies and a larger role for the private sector. PART IV Facing Aging and Providing Social Security Introduction This section addresses European and East Asian experiences in designing and revamping social security systems in the face of rapid social change. Pension reform is a particularly critical issue because rising life expectancy poses a fun- damental challenge to the basic design of modern welfare states Although indus- trialized economies face the most immediate challenges, the Republic of Korea and Taiwan, China, are already well into the demographic transition, and the challenge looms on the East Asian agenda In Chapter 14 Frank A G den Butter and Udo Kock analyze experiences with two types of social security systems in Europe-social insurance and redis- tribution-as well as the complex trade-offs among social security, equity, and economic growth. Although social security systems help overcome income inequality, for example, they can also hamper economic growth. However, social security systems can also foster economic growth if they enable unemployed workers to search for jobs that better match their skills. Hence designers of social security systems need to seek to balance equity and efficiency, with policymakers in developing countries especially avoiding too-generous systems that create a "social security trap " Box 4 1 is a thumbnail sketch of the history of social secu- rity in Europe In Chapter 15 Meesook Kim recounts the Republic of Korea's relatively fast development of a comprehensive social security system in the face of a bewilder- ing array of social changes. Today the Korean system includes work injury com- pensation, health insurance, public pensions, and unemployment insurance, and the country also ensures a minimum income to all its citizens. Although many of those provisions are nearly universal, the author contends that the Republic of 245 246 1 Facing Aging and Providing Social Security Korea must further extend those protections to address high unemployment, continued rapid aging amid dissolving family networks, and possible reunifica- tion with the impoverished Democratic People's Republic of Korea. In Chapter 16 Mauro Mare and Giuseppe Pennisi recount efforts to advance critically needed pension reform in Italy. After inconclusive debate in the 1980s, a foreign exchange and financial emergency in 1992 prompted major changes that continued under pressure from the impending European monetary union. Such experiences show that advocates of challenging reforms must act decisively given a window of opportunity, the authors say, and that pension reforms quickly reach a standstill if they do not form part of broader efforts to reform labor mar- kets and the welfare state. In Chapter 17 Luis Frota shows how countries have confronted the challenge of providing social insurance to agricultural households given rural aging. He recounts how European nations with large agricultural sectors have provided incentives to lure new generations to remain on farms, extended agricultural pensions to farmers who also hold jobs in other sectors, and boosted agricul- tural pensions up to those of other workers. He also examines East Asia's recent experience in extcnding social protection to rural communities experiencing out-migration, and he holds that fostering democratic participation in decision- making can help ensure that no farmers are left behind In Chapter 18 Jacques Charmes charts the changing views of public authori- ties toward the informal economy. In the early 1970s many analysts considered that sector a relic, destined to quickly disappear as countries embraced market and monetary economics. However, economic shocks in the 1980s prompted sig- nificant growth in the informal sector, which absorbed the resulting labor sur- pluses. Jacques Charmes shows that East Asian countries experienced an anom- alous effect during the recent crisis. Because informal sectors in those countries are tightly linked to export markets, they could not absorb displaced workers (many of them women), who instead returned to the countryside. The significant share of subcontracting work in the informal sector makes it an important engine for growth; yet countries will have to confront major questions regarding how to design social security systems to cover families who work in it. Finally Box 4 2 recounts the highlights of the February 2002 seminar in Rome/Caserta, which focused on the linked challenges of policymaking for the social sectors and managing reforms in social security regimes. BOX 4.1 A Brief History of Social Security in Europe by Frank A. G. den Butter and Udo Kock We can distinguish three stages in the evolution of European social secu- rity systems. In the first stage, charity was the main source of social protection for the poor. In the second stage, in the aftermath of the Industrial Revolution, coun- tries introduced social insurance schemes to cover the social risk of old age, occupational disability, and illness of workers in particular industries. Halfway through the 20th century, most such schemes were expanded to cover unemployment risk and all workers. In the thirdi stage, after World War II, prevention of social risks became important, and countries expanded social protection to cover almost all aspects of occupational and private life. In the 1970s many countries also introduced or extended early-retirement schemes in response to rising unemployment. Influenced by Keynesian ideas, social security policy had become a tool of macroeconomic policy. Western European countries differ with respect to the speed of this historic process, depending on the degree and types of industrialization, the level of economic development, and changing social conditions. However, by the end of the 1960s all countries had developed a compre- hensive system of social security. CHAPTER 14 Social Security, Economic Growth, and Poverty by Frank A. G. den Butter and Udo Kock ABSTRACT: This chapter reviews both the literature and experience concerning social security systems in Europe. The focus is on two distinct but closely related approaches to social security-social insurance and redistribution-and their implications for economic welfare and economic growth The chapter highlights the need for social security systems to bal- ance equity and efficiency by making complex tradeoffs In modern industrial countries, social security arrangements purport to protect workers and their families from extreme income losses, and should, therefore, enhance economic welfare for risk-averse individuals Solidarity is central to the philosophy of social security: it addresses the reality that private companies can- not insure against some risks, such as cyclical unemployment. This chapter reviews both the economic literature on social security systems and practical experience in building them. The focus is on the complicated rela- tionship among social security, economic welfare, and economic growth The chapter makes special reference to the European experience because social secu- rity systems in Europe are much more elaborate and diverse than those in Japan and the United States The aim is to draw policy lessons that might help devel- oping countries build and reform social security schemes, especially East Asian countries, which are experiencing relatively fast catch-up with the industrial world. The caveat is that we do not address risk management tailored to social structures in developing countries. 249 250 i Facing Aging and Providing Social Security Discussions of social security hinge on three major attributes. efficiency, equity, and administrative feasibility We focus mainly on efficiency, but policy- makers and analysts in Europe also heavily debate equity issues such as inequal- ity and social integration.2 We do not treat administrative feasibility because it is a rather technical issue with many legal and administrative implications. Pension schemes are also an important component of social security systems, but that topic deserves separate treatment (see Chapters 16 and 17). The next section describes the main characteristics of two concepts of social security in use in Europe: the Beveridge concept and the Bismarck concept. Then we apply aspects of economic welfare theory to social security arrange- ments. Later we address experience with actual social security arrangements in Europe and the problems policymakers encounter when designing and imple- menting them. We conclude with lessons about the design and implementation of social security schemes. Beveridge versus Bismarck Two social security concepts emerged in Europe during the 20th century: the insurance concept and the redistribution concept (see Box 4.1 for a brief history of the development of social security in Europe). Bismarck introduced the insur- ance concept in Germany in the second half of the 19th century when he designed the first social protection schemes for workers. British reformer Lord Beveridge developed the redistribution concept during World War II. The insurance concept focuses on insuring workers against income loss in the event of unemployment, disability, or retirement-the system smoothes lifetime income. Both contributions and benefits depend on earnings, and most programs are financed out of premiums and managed jointly by unions and employers. Redistribution programs, by contrast, cover all citizens Benefits are meant to be the last resort for workers without enough unemployment insurance, for households with no source of income, and for the working poor. The key focus is on poverty relief: benefits are means tested and provide only a minimum income guarantee. There is no link between contributions and benefits because they are financed through general taxes. Public administrative bodies usually administer the programs, which have few eligibility rules, although the means test is typi- cally less strict for unemployed workers than for persons without employment history. The difference between the two types of programs can be characterized as solidarity among insured workers and solidarity among citizens (see Table 14.1). Social Security, Economic Growth, and Poverty I 251 Table 14.1 Characteristics of Social Security Concepts in Europe Pure insurance (Bismarck) Pure redistribution (Beveridge) Main goal To guarantee socioeconomic status To guarantee income at subsistence level Eligibility Depends on contributions Is independent of contributions Expected benefits Match contributions Are means tested (contributions are income tested) Type of benefits Depends on previous wage Is means tested and at a flat rate and contributions Financing Is through premiums Is through general tax revenues Administration Is private Is public Focus Is labor market Is citizen's rights Examples Germany United Kingdom The savings concept represents a third alternative.3 Compulsory savings pro- vide social protection for individuals, not only for retirement pensions but also for contingencies such as unemployment, disability, and ill health. Benefits depend on the accumulation of an individual's compulsory savings. The typical case is the scheme in Singapore. All European social security systems also provide some universal benefits' tax-financed benefits for specific contingencies that do not require either a con- tribution or a means test. They can include child benefits and the flat-rate retire- ment pensions in countries such as the Netherlands and Sweden, the National Health Service in the United Kingdom, and family support and child benefits in many other European countries. We expect countries in which the redistribution concept dominates to spend a relatively large portion of revenues on social assistance and family allowances, and a relatively small portion on social insurance. The reverse pattern applies in countries in which the insurance concept dominates. Thus we can conclude that the insurance principle dominates in Germany and to a lesser extent in the Netherlands, whereas the redistribution principle dominates social security in the United Kingdom and to a lesser extent in Denmark and Sweden (see Table 14.2). Interestingly, although the insurance system provides relatively high bene- fits to a limited group, while the assistance concept grants relatively low benefits for a large group, overall spending as a percentage of gross domestic product does not differ much between the two systems. In practice, of course, many European social security programs combine ele- ments of both systems, and all countries have faced trade-offs. The concepts and 252 1 Facing Aging and Providing Social Security Table 14.2 Characteristics of Social Security Systems: Sources of Revenue and Type of Benefits Sources of revenue' (%j Spending by type of benefit' (%) Insured Social Social Family Country persons Employers Stateb insuranced assistance allowance Denmark 1960 na na na na na na 1970 141 89 758 694 185 66 1980 18 59 902 683 242 30 1985 31 78 861 682 257 18 1990 46 51 878 626 286 33 19958 108 95 737 na na na France 1960 189 689 111 na na na 1970 189 684 103 na na na 1980 210 534 241 626 158 106 1985 230 503 237 657 132 112 1990e na na na 719 46 119 Germany 1960 259 444 250 667 66 20 1970 289 423 251 708 47 26 1980 340 342 289 754 41 50 1985 363 347 264 783 49 33 1993 368 323 276 817 64 26 1995c 298 381 298 na na na Netherlands. 1960 409 403 122 644 48 130 1970 385 429 123 694 48 103 1980 332 333 246 737 62 75 1985 395 318 163 765 35 72 1990 430 177 244 774 27 51 1993 460 180 214 775 22 48 Social Security, Economic Growth, and Poverty 1 253 Sources of revenuea (%°/) Spending by type of benefit' (%c Insured Social Social Family Country persons Employers Stateb Insuranced assistance allowance Sweden 1960 205 110 669 713 122 107 1970 114 277 546 802 147 51 1980 10 459 453 779 168 52 1985 1 3 364 522 845 03 153 1991 1 7 397 494 842 06 129 1993 1 0 43 0 56 1 83 8 1 5 12 8 United Kingdom f 1959-60 20 0 17 9 58 7 71 0 10 8 5 4 1969-70 199 249 530 705 150 54 1979-80 15 6 26 2 55 0 67 3 15 6 7 8 1984-85 183 235 555 588 245 70 1991-92 156 255 498 611 228 57 1993-94 140 228 547 578 262 56 n a = not available a Includes medical care, sickness benefits, unemployment insurance, retirement pensions, employment injury benefits, family benefits, maternity benefits, invalidity benefits, and survivors' benefits Totals do not add up to 100 percent because revenues from capital income and other sources have been omitted b Includes special taxes allocated to social security and participation of other public authorities c Total spending does not add up to 100 percent because benefits for pubic employees and war victims have been omitted d Includes public health services e The definition changed, so no consistent data were available f Reference period is April-March Source lntemational Labor Office (1992). available at , Nicholas Barr, 'Economic Theory and the Welfare State A Survey and Interpretation," Journal of Economic Literature 30, no 211992) 741-803, and authors' calculations 254 I Facing Aging and Providing Social Security content of particular social security schemes have also changed according to the political and economic context, and because policymakers continue to fine- tune programs to suit economic and labor market conditions. Nevertheless, the classification of social security systems turns out to be remarkably stable over time. In fact, many countries have recently sharpened their profiles.4 The United Kingdom has moved even further away from the insurance principle unemployment insurance contributions have related to earnings since 1992, although the program pays a flat-rate benefit. France, Germany, and Spain, meanwhile, have strengthened the link between benefits and previous employ- ment and contributions. Considering Efficiency and Equity Economists do not agree on whether a comprehensive system of social security and cuts in inequality from redistribution are an impediment or a spur to eco- nomic performance.5 According to one argument, social security expenditures are thrown into a leaky bucket because the welfare loss of those who pay the social security premiums is larger than the welfare gain of those who benefit.6 Others hold that inequality is harmful to economic growth' and that the metaphor of the leaky bucket wrongly assumes that we live in a perfect world with complete infor- mation and well-functioning markets. Those analysts advance the irrigation function as an alternative to the leaky bucket. According to that theory, workers with some economic security will be more eager to search for new jobs, and job destruction and creation can proceed at a good pace at the macro level, enhanc- ing productivity and economic welfare. The irrigation function underscores the desirability of social security sys- tems in newly developing countries that, owing to technological catch-up with the industrial world, are witnessing fast economic growth It is especially impor- tant for workers in those countries to spend time searching for good jobs and establishing good matches rather than earning a subsistence living in the infor- mal sector. How Can Social Security Enhance Welfare? Formal modeling exercises show that the limited U.S. social security provisions give the United States lower unemployment than in Europe, but that European social security systems are more beneficial to economic welfare.8 Social Security, Economic Growth, and Poverty 1 255 Skill-biased technology shocks are the most important source of the differ- ence between these two regions. These shocks imply that educational differ- ences become more prominent as the labor market demands more capabilities and skills 9 The effects of skill-biased technology shocks differ according to whether countries have extensive or limited social security systems. A generous social security system makes both employers and employees more choosy in establishing a good match. This mechanism gains importance when the demand for skills grows more heterogeneous, and leads to longer unemployment spells when social security is good. This explains why unemployment is higher in Europe than in the United States. But the quality of the matches increases and productivity is higher in the European situation because companies better exploit workers' skills 10 The trade-off reflects the equilibrium search theory of the labor market. The same models indicate that social security enhances the welfare of low-paid work- ers as well as the unemployed. The models do not reckon with risk-averse behav- ior, but a good social security system can enhance welfare by negating such behavior (The models also do not allow for job mobility, in that they assume that only unemployed and not employed workers search for new jobs.) Although the models analyze the effects of differences among social security systems in the industrial world, the arguments about why a good social security system may enhance productivity and welfare also apply to developing countries undergoing fast economic development and extensive structural change. Analyzing Reforms The traditional aims of social security are to protect people from the financial consequences of unemployment, disability, retirement, and other social risks, and to reintegrate workers into the labor market. Analysts should use those two main goals to assess proposals to reform or implement social security pro- grams. l The first key question relates to the coverage and scope of a social security program or reform: who is entitled to benefits? Analysts often pose this ques- tion as a trade-off between general and targeted policies. The advantage of tar- geted policies is that they directly address the social risk of a narrowly defined group. A disadvantage is that people outside the target group have an incentive to adjust their behavior to become entitled to the benefit. The earned-income tax credit is an example of a targeted tax policy aimed at increasing the supply of low-wage workers. The disincentive of this policy is that high marginal tax 256 1 Facing Aging and Providing Social Security rates in phase-out range give workers disincentives to invest in human capital. The policy also distorts the relative prices of labor. General social security policies do not face these problems because their wider scope reduces people's incentives to change their behavior to become entitled to a benefit. However, general policies suffer from a large deadweight loss: a large number of recipients who are entitled to the benefit do not need it. A good exam- ple is the general income tax credit for workers, which was introduced under the Dutch tax reform and became effective in 2001. The credit aims to boost labor supply and cut unemployment. The deadweight loss arises because current workers are also entitled to the tax credit, without creating additional labor sup- ply Government budget constraints will limit the tax credit, further restricting its impact on labor supply and unemployment. On the positive side, the policy measure does not distort the labor market. Once policymakers have determined a social security program's coverage and entitlement criteria, they have to enforce those criteria. Applications have to be evaluated and entitlement decisions made. This gatekeeper function determines who will receive a particular benefit and what its level and duration will be The gatekeeper function is usually more important for targeted policies because those programs give people incentives to change their behavior and apply for a benefit, although their personal characteristics have not changed. The gatekeeper function always involves social costs. A strict application of the rules implies that the chance that the system will admit someone who is not entitled to benefits is small. This is analogous to the error of the second kind in statistical testing. On the other hand, the error of the first kind-the proba- bility that the system will not admit someone who is entitled-is rather large. The opposite holds for a generous application of admission rules. Policymakers can try to enhance the discriminatory power of the gatekeeper function by reducing the asymmetric information between the gatekeeper and the individ- ual who seeks entitlement, but doing so will seriously enlarge the costs of gate- keeping. Those first two criteria in analyzing social security reform-the trade-off between general and targeted policies, and the gatekeeping function-determine the number of people entering the system. The third criterion-incentives- determines outflow from the system. That is, social security programs for workers who are unemployed, have disabilities, or are ill should contribute to rapid reinte- gration in the labor market. The system can accomplish this in many ways, such as by integrating active labor market policies into the social security program. Social Security, Economic Growth, and Poverty 1 257 Designing Incentives If properly designed, social security programs can minimize their adverse effects on the labor supply. However, many social security programs in Europe cause severe incentive problems in the labor market. The magnitude and form of these problems depend on the scheme's institutional characteristics For unemploy- ment compensation programs, these characteristics include the following * Benefit level * Benefit duration * Contributions (in the form of taxes or premiums) * Entitlement conditions (such as through layoffs rather than voluntary quits and other employment history) * Job search conditions * Job acceptance conditions (people may reject some unsuitable job offers) * Means test (including other household income and housing property) * Household circumstances (such as number of children) The traditional negative incentive effect is that higher benefit levels and longer benefit duration may mean that workers are less willing to search for jobs or accept job offers An increase in benefit levels will also increase the value of leisure, and the worker will reduce his or her search intensity, cutting the out- flow rate from unemployment.'2 High benefit levels certainly undercut workers' incentives to accept low-paid jobs, because their net gain in income will be small In many European countries, supplementary means-tested benefits increase this problem. In Germany, the Netherlands, and the United Kingdom, unemployed workers receiving unem- ployment assistance are exempt from local taxes and are entitled to discounts for some educational and health services. In some countries rent allowances are linked to unemployment assistance If unemployed workers accept a job, they will lose most of these additional benefits. This creates the "unemployment trap," by which workers remain unemployed voluntarily. On the other hand, unemployed workers who arc not entitled to unemploy- ment insurance benefits will raise their search intensity when unemployment benefits become more generous, to find employment and hence qualify for unem- ployment benefits. The magnitude of this entitlement effect depends on the enti- tlement conditions. 258 I Facing Aging and Providing Social Security Whether a country's social security system is mainly insurance based or redis- tribution based determines the incentive effects of unemployment compensa- tion.'3 Insurance-based social security systems, such as in Germany, tend to exclude long-term unemployed workers from insurance. Because these unem- ployed workers usually receive means-tested unemployment assistance or a fam- ily allowance, the unemployment trap applies in particular to them. In countries such as France, Germany, and the Netherlands, unemployed workers who reach the end of the maximum period of unemployment insurance benefits join the stock of unemployment assistance beneficiaries and face a sharp decline in benefit levels. This induces them to lower their "reservation" wage- the wage at which they will accept a new job-and increase their job search activ- ities, raising the outflow from unemployment Empirical studies have found a large rise in the outflow rate of unemployed workers toward the end of the bene- fit period.i4 However, higher benefits extend the duration of unemployment only slightly,15 because other policies counteract them. For example, Sweden intro- duced active labor markct policies in the late 1960s and early 1970s while extend- ing unemployment compensation. 16 Practical Experience: The Spiral of the Wedge and the Supply Effect Social security influences tax and premium rates, the development and pattern of labor force participation, and labor productivity.i7 The interaction of those macroeconomic variables can cause a negative spiral High average and marginal tax and premium rates create a wedge between the gross and net wages of work- ers, inducing a decline in labor force participation and the number of hours worked, especially for women earning a second family income. Low labor force participation, in turn, provides a small base through which to finance benefits. A downturn of the business cycle will require a relatively large rise in premiums and taxes, inducing unions to demand higher wages. This reduces labor demand, shrinking the premium and tax base even further and increasing demand for social security benefits. High wage costs reinforce this mecha- nism-sometimes referred to as the "social security trap"-because employers have an incentive to increase productivity by investing in labor-saving technol- ogy and other innovations. A major engine behind this negative spiral is that supply, as well as demand, determines the impacts of social security The supply effect reflects the fact that Social Security, Economic Growth, and Poverty 1 259 people become more aware of their legal right to demand social security benefits, so more eligible people actually reccive them. One investigation of such a supply effect in the Netherlands showed a sharp rise in the number of people who received benefits from 1970 to 1990,18 even though the system did not add major new provisions.19 Strengthening eligibility standards and applying stronger gate- keeping can decrease this effect, as can admonishing people not to abuse the sys- tem and providing training and work experience rather than just financial com- pensation (workfare instead of welfare). Because the indirect effects of the supply effect on labor participation are much higher than the direct effects, such insti- tutional changes are essential to reducing the negative spiral and curbing it, espe- cially in developing countries that are building a social security system. To our knowledge, no comprehensive studies have examined the supply effect. However, a general comparison among countries can shed light on these relationships. If we take the United States as a benchmark, for example, Japan has high labor participation and low social security expenses and tax and pre- mium rates, combined with low productivity. In France and the Netherlands, and to a lesser extent in Germany and the United Kingdom, labor participation is low, whereas social security expenditures and income and productivity levels are high (see Table 14.3) Other Influences on the Economic Impact of Social Security The impact of a social security system depends on circumstances in other parts of the economy and society Economic growth, low unemployment, and high labor force participation may accompany a generous social security system, but if the macroeconomic, social, or technological environment changes, the same sys- tem could seriously frustrate economic development. Many analysts maintain that economic turbulence caused by technological change and global competition underlies high and persistent unemployment in Europe, given a generous system of social security that reduces work and job search incentives. The developers of the social security systems in the 1950s and 1960s, a time when structural change and labor market dynamics were more modest, did not foresee those problems. Today workers lose skills when displaced and while unemployed, owing to the transition from a manufacturing to a service economy, economic globalization, and new information and communication technologies This causes long-term unemployment because reservation wages are high relative to productivity. Table 14.3 The Supply Effect of Social Security (1993): Labor Participation, Social Security Expenditures, and Productivity and Wage Cost Elasticities Wage cost elasticities with respect to social security contributions Social security Taxes and GDP per GDP per hour Income taxes Participation expenditure (as a premiums (as a worker (index, worked (index, Employer's and worker's rate (labor years) percentage of GDP) percentage of GDP) United States =100) United States = 100) contnbutlons contnbutions Netherlands 51 27 55 90 106 France 56 24 47 94 97 0 40 0 40 United Kingdom 60 16 46 75 80 0 25 0 25 Germany 62 15 37 82 84 100 1 00 United States 65 13 32 100 100 0 00 1 00 Japan 67 12 33 75 63 Source Organization for Economic Cooperation and Development (OECD), The OECD JobStudy(Paris OECD, 1995f, OECO, LaborForce Statzstics(Paris OECD, 1995), and K van Paridon, 'The Crucial Importance of High Labor Force Participation The Dutch Case and Lessons for Europe,' in Unemployment in Europe, ed Michael A Landesmann and K Pichelmann, (London Macmillan, 2000), 1 8-204 Social Security, Economic Growth, and Poverty 1 261 However, economic turbulence can also be seen as a permanent shock that increases the value of the right match between jobs and workers.20 Generous benefits allow workers to search longer for better matches, which increases the equilibrium unemployment rate. Mismatch declines, but at the expense of an inefficient level of investment in search Changing social and demographic patterns also influence the long-run impact of social security programs. Individualization and the growing number of divorced couples put more pressure on noncontributory programs such as fam- ily allowances and social assistance In the United Kingdom policymakers con- sidered social assistance a temporary provision in 1948, because they expected that eventually everyone would be self-supporting through work or unemploy- ment insurance l A social security system where benefits are highly dependent on labor market status, such as the German system, has trouble coping with changing social patterns outside the labor market. Structural changes in the employment pattern of workers, such as the sharp rise in the share of part-time workers in the Netherlands, also put pressure on existing social security pro- grams, because unemployment and occupational disability programs often exclude workers with intermittent incomes. Despite those changes, the number of disability beneficiaries in the Netherlands has exploded since the 1970s and now exceeds 900,000-far beyond expectations when Parliament unanimously passed the disability act in 1967. One study concludes that up to 50 percent of workers receiving disability benefits are in fact unemployed.22 Because disability benefits are more generous than unem- ployment benefits, a high share of workers with disabilities raises wage costs and increases the government's budget deficit. If employment outflow is skewed toward disability benefits, it also makes the unemployment level artificially low. This generates biased and unintended labor market signals that could raise wages beyond economically sustainable levels. Yet for obvious political reasons, curbing this "curse of a good act" is extremely difficult. This problem is not unique to the Netherlands. One study suggested that the unemployment rate in Norway also significantly affects the disability program.23 Conclusion: Providing a Trampoline Instead of a Hammock The lesson for newly developing economies is that there is no such thing as a "natural next step," uniform blueprint, or optimal system in developing formal social security programs Social security systems should be tailor made, adapted to the social structure and labor market in each country. 262 1 Facing Aging and Providing Social Security Social security systems may combine elements of the insurance concept and the redistribution concept The system will always have to be based partly on solidarity, especially with respect to the poor, who should receive an above- subsistence income guarantee. On the other hand, the system should contain enough incentives to avoid moral hazard and frce rider behavior. It is essential that incentives in the system encourage, not discourage, labor participation. The redistribution inherent in social security systems is bound to diminish income inequality. Because of the trade-off between equity and efficiency, this may hamper economic growth On the other hand, social security, owing to its irrigation function, can also foster economic growth because it allows the unemployed to search for good and productive job matches. Hence the design of a social security system should seek a balance between the negative equity- efficiency trade-off and social security's irrigation function A major lesson, especially suggested by the Netherlands experience, is that policymakers should not overdirect the initial design of the social security sys- tem toward provision of benefits. The mere availability of these benefits can evoke a supply effect strengthened by the negative wedge spiral: individuals' entitle- ment to social security benefits causes a negative externality for the employed, because they must pay higher social security premiums. Newly developing coun- tries should avoid setting up too generous and passive systems- they should pro- vide a trampoline instead of a hammock Notes 1 For more discussion see, for example, World Bank, World Development Report 2000/2001 Attacking Poverty, (Oxford, England Oxford University Press, 2000) 2 See Nicholas Barr, "Economic Theory and the Welfare State A Survey and Interpretation," 3ournal of Economic Literature 30, no 2 (1992) 741-803, section 5, for a brief assessment of the distributional effects of social security in a number of Organization for Economic Cooperation and Development (OECD) countries 3 Sara Connolly and Alistair Munro, Economies of the Public Sector (London Pearson Education, 1999) 4 Gunther Schmid and Bernd Reissert, "Unemployment Compensation and Labor Market Transitions," in International Handbook of Labor Market Policy and Evaluation, ed Gunther Schmid and Bcrnd Reissert (London Edward Elgar, 1996), pp 235-76 5 See, for example, H Borstlap, "Labour Market, Social Protection and Economic Performance," in Economic Science Art orAsset2 The Case of the Netherlands, ed Peter A G van Bergeilk, A G van Bergeijk, A L Bovenberg, E E C van Damme, and J van Sinderen (Rotterdam, Nether- lands OCFEB, 1996) 6 Arthur M Okun, Equality a,id Efficiency The Big Trade-off (Washington, DC Brookings Institution, 1975) Social Security, Economic Growth, and Poverty 1 263 7 T Persson and G Tabellini, "Is Inequality Harmful for GrowthP" American Economic Review 84 (1994) 600-21 8 D T Mortensen and C A Pissarides, "Unemployment Responses to Skill Blased Technology Shocks The Role of Labour Market Policy," Economic Journal 109 (1999) 242-65, and R Marimon and F Zilibotti, "Unemployment versus Mismatch of Talent Reconsidering Unemployment Benefits," EconomicyJournal 109 (1999) 266-91 9 See, for example, S Machin and J Van Reenen, "Technology and Changes in Skill Structure Evidence from Seven OECD Countries," Quarterly Journal of Econontics 113, no 4 (1998) 1215-44, and E Berman,J Bound, and S Machin, "Implications of Skill-Biased Technological Changes International Evidence," Quarterly Journal of Economics 113, no 4 (1998) 1245-79, for empirical investigations of the importance of these shocks 10 R Marimon and F Zilibotti, op cit. 11 See also Scienitfic Councilfor Government Policy, Continued Growth of Labor Force Participation (in Dutch), Report to the Government no 57 (The Hague, Netherlands Sdu, 2000), and F A G den Butter and U Kock, "More Labor Participation with Individual S avings Accountsp" (in Dutch), Openbare Uitgave 32 (2000). 200-12 12 See E Karni, "Optimal Unemployment Insurance A Guide to the Literature," World Bank mimeo (January 1999), for an overview of the theoretical literature on optimal unemployment insurance 13 See also Schmid and Reissert, op cit 14 G J Van den Berg, "Nonstationarity in Job Search Thcory," Review of Economic Studies 57 (1990) 255-77 15 R Layard, S Nickell, and R Jackman, Unemployment Macro-Economic Performiance and the Labour Market (Oxford, England Oxford University Press, 199 1) 16 A B;orklund and B Holmlund, "Effects of Extended Unemployment Compensation in Sweden," in Tihe Political Economiy of Social Security, ed B A Gustafsson and N Anders Klevmarken (Amsterdam North Holland, 1989), 165-83 17 See K van Paridon, "The Crucial Importance of High Labor Force Participation The Dutch Case and Lessons for Europe," in Unemployment in Europe, ed M A Landesmann and K Pichelmann, (London Macmillan, 2000), 188-204 18 F A G den Butter, "Supply of Social Security as a Cause of Low Labour Participation in the Netherlands A Cliometric Analysis," (Free University Rescarch Memorandum, Amsterdam, 1993),p 56 19 R F M Lubbers, "De economische politiek in Nederland vanaf de jaren '60" (Economic policy in the Netherlands from the 1960s), in Het sociaal-economisch beleid in de tweede helft van de twinztigste eeuw, ed J van Sinderen (WVolters-Noordhof, Netherlands Groningen, 1990), pp 14-19 20 See the model by Marimon and Zilibotti, op cit 21 Nicholas Barr, The Economics of the IVelfare State, 3d ed (Oxford, England Oxford Ulliversity Press, 1998), p 242 22 L J M Aarts and Ph R de Jong, Economic Aspects of Disabihty Behavior (Amsterdam. North Holland, 1992) 23 E Bowitz, "Disability Benefits, Replacement Ratios and the Labor Market A Time Series Approach," Applied Economics 29 (1997) 913-23 CHAPTER 15 Social Security and the Social Safety Net in the Republic of Korea by Meesook Ktm ABSTRACT. Social security in the Republic of Korea has developed rap- idly in all three fields: social insurance, public assistance, and social serv- ices Public expenditures have risen while coverage has gradually expanded to become almost universal in some areas. However, public expenditures are still below those of other Organization for Economic Cooperation and Development (OECD) countries, and the Korean social security system faces tough challenges ahead Although the Republic of Korea has developed an ever more elaborate social security system, successive governments have left major responsibility for pro- tecting against risks to individuals, households, and the market Until recently, social security and welfare goals were subordinate to economic growth. Social security became an important policy issue during the 1997-98 crisis, as unem- ployment soared. With unemployment came other social problems rising poverty, family dissolution, homelessness, and child neglect As a result, Korea acted to confront the urgent and overriding challenge of strengthening the social security system by earmarking higher budgets for the welfare system and reorganizing administrative structures. This chapter introduces the overall Korean social security system-which includes social insurance, public assistance, and social services-and its limita- tions. The chapter compares Korea's social welfare expenditures and programs with those of other OECD countries, and it concludes with a summary of the future social security challenges facing Korea. 265 26f6 Facing Aging and Providing Social Security Social Insurance Components of Korea's Social Security System Social security in Korea includes four social insurance components: work injury compensation, health insurance, public pensions, and unemployment insurance (see Table 15.1). The country laid the groundwork for its social insurance system over a short period of time. Work injury compensation insurance, introduced in 1964, is the oldest main component. Health insurance has expanded since 1977, and the country established a public pension program in 1988 Unemployment insurance, the newest social insurance element, was adopted in 1993 Work Injury Compensation Insurance Work injury insurance is financed exclusively by contributions from employers; employees are not required to make contributions.' The government covers the administrative costs of the system. Beneficiaries and Benefits The range of beneficiaries of work injury insurance has expanded. In 1964, one year after the workplace insurance law was enacted, it covered only miners, assembly workers, and workers at firms with more than 500 employees Amendments to the law in 1972 extended insurance to workplaces with 30 or more employees. The government further extended insurance in 1992 to cover workplaces with five or more employees, and after the financial crisis, it extended Table 15.1 Types of Social Insurance in Korea, 2001 Types Recipients Number of participants Ministry, year established Work injury All workplaces 9 50 million workersa Ministry of Labor, 1964 compensation insurance Health insurance All people Over 96% of population Ministry of Health and Welfare, 1977 National pension All people 16 million Ministry of Health and Welfare, 1988 Unemployment Almost all 9 27 million workers Ministry of Labor, 1995 insurance workplaces a Figure is from 2000 Source Ministry of Labor, Report on Work Injury Compensation Insurance in 2000(Seoul Mtnistry of Labor, 20011, Ministry of Labor, White Paper on LaborlSeoul Ministry of Labor, 2001), and Ministry of Health and Welfare, White Paper on Health and We/fare (Seoul Ministry of Health and Welfare, 20011 Social Security and the Social Safety Net in the Republic of Korea 1 267 coverage to all workplaces. As of 2000, beneficiaries totaled 9.50 million workers at 706,231 workplaces nationwide.2 Benefits include sick leave, a sickness com- pensation pension, and a disability benefit.3 The insurance covers treatment of work-caused diseases and injuries, compensates for income losses, and helps pre- vent work-related injuries The insurance also covers occupation-related diseases that develop over time.4 Future Tasks Korea's work injury insurance is widely criticized for its stringent eligibility cri- teria and narrow range of coverage Moreover, its benefit level is too low to pro- tect the economic stability of injured workers. Because the prevention of work injury is even more important than its treatment, raising benefit levels alone is not sufficient: the country needs to pursue full-scale prevention strategies to reduce the possibility of injury and foster employers' and employees' awareness of workplace security. Health Insurance Health insurance is generally designed to reduce the financial onus imposed by accident, disease, delivery, and death. When Korea enacted its health insurance law in 1963, coverage was too narrow to function as a social security scheme. Only in the 1970s, when the country was in the throes of economic development, did health insurance become a social issue. Witnessing the need, President Chung Hee Park turned his attention to developing a social welfare program that included health insurance.5 Against this backdrop Korea adopted a national health insurance system in 1977 that encompassed firms with more than 500 employees as well as medical assistance for the poor In 1989, 12 years after its inception, the system was extended to cover everyone, including rural residents and the urban self-employed.6 Benefits and Types Health insurance benefits consist of cash payments and in-kind benefits. The lat- ter include medical consultations, drugs and other therapeutic materials, med- ical and surgical treatments, hospitalization, operations, and other services such as nursing care and transportation.7 To prevent the unnecessary use of health care services and resources and save on public health care costs, the insured make co-payments when receiving medical services. For example, patients pay 20 per- cent of hospitalization fees. Cash benefits reimburse insured people and their 268 1 Facing Aging and Providing Social Security dependents for medical care and delivery costs, and also provide a fixed amount for funeral expcnses.8 Three health insurance schemes used to correspond to different target groups government employees, private school employees, military service employees and their dependents, rural and urban self-employed individuals, and employees of industrial and commercial companies In 1998 the country merged the first and third schemes, and in 1999 it incorporated all three into a single sys- tem under the National Health Insurance Act Contributions from insured indi- viduals, employers, and the government finance the system. In 2000 Korea separated prescription services from dispensation services for the first time, to prevent the overuse and misuse of drugs and drug-related acci- dents.9 Eight months later the National Health Insurance Corporation con- fronted a serious financial imbalance owing to growing fees required by medical service centers The Korean government had to devise a coping mechanism to reduce the financial burden on medical insurance. The government took a loan from the banks, raised health insurance fees, drew tobacco tax money, raised drug prices and medication examination fees, and limited the services covered by the insurance to 365 per year. Challenges Ahead Korea has achieved near-universal health insurance within a short time frame. The country must now focus on the quality as well as the quantity of health care. Three issues should receive priority. First, people should pay insurance fees in proportion to their income to intensify the income redistribution role of the health insurance system 10 Second, because existing schemes do not cover numerous services, future tasks include extending health services to examination of, treatment of, and rehabilitation from chronic disease. Health insurance should cover not only curative treatment of disease and disability but also pre- vention. Finally, the government needs to secure the health insurance system's finances by minimizing the provision of unnecessary health services by health centers. In the long run Korea must develop avenues for delivering cost-effective health services by connecting them to health centers and by connecting personal services to social welfare centcrs.II National Pension Insurance Schemes Korea first adopted a national pension system in 1988 by amending the unworkable, ineffective 1973 pension law. Before 1973 the pension system covered only government employees, military personnel, and private school- Social Security and the Social Safety Net in the Republic of Korea I 269 teachers The national pension scheme (NPS)-the centerpiece of public pen- sions in Korea-now covers everyone, including farmers, fishers, and the self- emploved The system provides the insured with financial security against aging, disability, and death. National Pension Beneficiaries National pension participants fall into two groups workplace-based participants and residential area-based participants. Since 1991 the scheme has included employees in businesses with five or more workers, and rural dwellers have been able to participate since 1994. As of 1998 the scheme included 5.06 million workplace-based participants and 2.11 million regional participants."2 In 1999 the country extended the scheme to the urban self-employed, employees at workplaces with fewer than five workers, and part-time workers The scheme does not automatically cover full-time housewives and people under 23 years of age without an occupation, but they can participate voluntarily. The average income replacement ratio used to be 70 percent for those with 40 or more years of participation, but amendments reduced the benefit to 60 percent, although people now qualify for benefits after 10 years rather than 15.13 Recipients receive pension benefits for life So that the scheme can achieve financial stability, the minimum age for receiving pension benefits will rise from 60 to 65 in 2033. Challenges Ahead Challenges include the level of benefits, fund security, coverage, and links to other pension plans. The pension fund is on the verge of exhaustion because its benefit levels have been too generous compared with low contribution rates. This imbalance needs to be redressed so future generations do not carry the burden of financing the present generation's pension fund. The current pension fund is likely to be depleted by 2030 owing to accelerating population aging Although the government recently took countermeasures to secure the fund by lowering benefits and raising the entitlement age, it still needs to find ways to manage the fund to make it secure and profitable. And even though coverage has expanded to include rural residents, self-employed people, and part-time workers, the sys- tem needs to expand further to include housewives and spouses working with their self-employed partners. Finally, because the present pension scheme is not linked to public pension schemes for government employees, military personnel, and private schoolteachers, insured people who move across schemes are likely to lose their entitlement if they do not fulfill the minimum contribution period for either scheme. Combining contribution periods would increase the portabil- ity of pensions 270 1 Facing Aging and Providing Social Security Unemployment Insurance Of the four types of social insurance, unemployment insurance is the newest; it was started in 1995, 2 years after the enactment of the Unemployment Insur- ance Act, when unemployment was fairly low. Unemployment insurance was designed not only to secure income for workers during unemployment but also to promote employment through job training and human development. The insurance now covers all full-time workers-9.27 million employees from 1,208,000 workplaces as of 2001. Unemployment benefits provide 50 percent of average income before unemployment, with a minimum of 250,000 won and a maximum of 900,000 won per month.14 The insured must have worked at an insured company for at least 6 months to be eligible for benefits, and may receive benefits for 2 to 7 months.15 Public Assistance Schemes Korea officially established its public assistance system in 1961, when the liveli- hood protection law took effect. Benefit recipients were initially to include two types: home and institutional care recipients, and self-support care recipients. Home and institutional care recipients were those unable to work, including the elderly, children, people with disabilities, and those cared for at welfare institu- tions. Self-support care recipients were those able to work but lacking enough resources to live. The scheme provided livelihood aid, maternity aid, and burial aid only to home care and institutional care recipients,'6 but the rest of the bene- fits were provided to both recipient groups The livelihood protection system was widely criticized because of its low benefit levels, unreasonably narrow selection criteria, and other structural prob- lems. In 1999 the Basic Guarantee Law replaced the Livelihood Protection Act within a context of rising demand for public assistance after the economic crisis. The unemployment rate rose sharply during the crisis, from 2.6 percent in 1997 to 6.8 percent in 1998 and to 6 3 percent in 1999, and the poverty rate doubled. Korea's social safety net was clearly too weak to respond effectively: the liveli- hood protection system covered only 60.4 percent of the absolute poor. Because it also did not cover self-support care recipients and the low-income unem- ployed, they suffered most. The Basic Guarantee Law was the first legal mecha- nism to guarantee a "national minimum" and self-sufficiency for the poor. Those eligible for benefits fall into three categories The first includes elderly persons who are unable to support themselves, children under 18, expectant Social Security and the Social Safety Net in the Republic of Korea 1 271 women, and people unable to work owing to disease or mental or physical dis- ability. The second category includes those who have no one to support them. The third group includes people whose family per capita income and household property fall below a certain level. (In 2000, to be eligible, a recipient had to report family income of 930,000 won or less, and household property valued at 32 million won or less for a family of four). The system includes seven types of protection: livelihood aid, housing aid (newly added), medical aid, educational aid, self-support aid, maternity aid, and burial aid. The new law has lifted demographic eligibility criteria, and anyone who fits the family income and property criteria is entitled to aid. Because the new law now includes housing aid for the first time, the Korean welfare system protects every aspect of basic living. The new law clearly strengthens the Korean social safety net. The system pro- vided over 1.5 million people in poverty with livelihood aid in 2000, against only 0.5 million people in 1999, when the income guarantee was not yet in effect (see Table 15.2). Moreover, the overall benefit level per beneficiary has risen. The number of homecare recipients has grown by only 9.0 percent, but the number of self-support care recipients, who were not eligible for livelihood aid under the old law, has risen by as much as 40.9 percent. The Korean government has markedly expanded the budget for basic livelihood aid. In 1997 the per capita basic livelihood aid budget was 639,000 won, but in 2001 it was 1.98 million won-an increase of as much as 210 percent 17 Although the effects of the basic livelihood guarantee have yet to be evalu- ated, several areas for improvement can be identified. First, since the poverty line is based on living costs in small and medium-size cities, the system excludes people from larger cities where living standards are higher 18 The system also excludes some unprotected poor, elderly, people with disabilities, and children whose family income is above the poverty line but who live a very poor life. The Table 15.2 Livelihood Aid Beneficiaries 1997 1998 1999 2000 Total beneficiaries of public assistance (A) 1,410,000 1,470,000 1,920,000 1,510,000 Number of persons covered by livelihood aid (B) 370,000 440,000 540,000 1,510,000 B/A (percent) 26 2 29 9 281 100 Source Mee-Gon Kim, "Basic Livelihood Guarantee System as a Social Safety Net" (paper presented at a workshop on Securing Social Safety Nets, Korea Institute for Health and Social Affairs, Seoul, 2001), p 77 272 I Facing Aging and Providing Social Security government needs to readjust the minimum cost of living according to region, household size, and type of household. For instance, the cost of living for a fam- ily with a member with disabilities must be higher to reflect its medical fees The welfare administration system also needs improvement. Local govern- ment offices now lack enough personnel to satisfy all the need, as of 2001, one public employee dealt with 150 households in Korea, whereas in developed coun- tries one public employee deals with 100 households.19 This work overload means that the poor cannot get enough support or social services adapted to their type of household. Medical Assistance Those who are eligible for public assistance also receive medical assistance. This system was first adopted in 1961, when the livelihood protection law was enacted, but has since been managed separately for financial reasons. The system is more closely linked to the health insurance system, unlike in other countries, where medical assistance programs are typically part and parcel of public assistance. To be eligible for benefits, individuals must demonstrate that they live below the poverty line. Beneficiaries may receive one of two kinds of medical assistance, depending on the degree of their poverty. First-class assistance, for those in the lowest-income class, provides various medical services, including hospitalization and outpatient services, free of charge. Second-class medical assistance, provided to people from the second lowest-income class, offers medical services for a small charge For example, recipients pay 20 percent of hospitalization fees and 1,500 won per visit for outpatient services.20 Most recipients of medical assistance are those who have chronic diseases and those who are very poor and receive no family support. In particular, first-class medical assistance recipients tend mostly to be the severely sick or elderly people without a family. Those recipients need to consult with someone who can link them to other welfare services. Social Services Social services were devised to protect the socially disadvantaged, including the elderly, children, those with disabilities, single parents, and the homeless. Un- like social insurance and public assistance, social services provide clients with Social Security and the Social Safety Net in the Republic of Korea 1 273 professional social work Although this gives the socially disadvantaged added protection, benefit levels are so limited that the system covers only people with the lowest income, leaving out most socially disadvantaged persons. Welfare Expenditures Before the crisis, social welfare expenditures in Korea were low compared with those of other OECD countries. In 1990 the country spent 8 trillion won-only 4.52 percent of its gross domestic product (GDP)-on social security, according to OECD estimates (see Table 15.3). But in 1997 social security expenditures rose to 6 65 percent of GDP (30 trillion won),21 and in 1998 they grew dramatically to 11 09 percent of GDP.22 Those increases reflected growth in budgets devoted to social assistance and unemployment insurance. Spending on health, which includes health insurance and medical assis- tance, constituted the largest share (37.0 percent) of total social security expen- ditures, followed by unemploymcnt (34 9 pcrcent), pensions (18 0 perccnt), compensation for work injuries (3 7 percent), social services (3.7 percent), and public assistance (2.7 percent). Korea's social expenditures were lower than those of some OECD countries in every category except work injury However, expenditures on health and pensions are projected to reach those of developed countries in the near future.23 Table 15.3 Relationship between Social Security Expenditures and GDP, 1990-98 Year Rate (%) 1990 4 52 1991 4 28 1992 4 63 1993 471 1995 523 1996 5 47 1997 6 65 1998 11 09 Note This mformat on is based on OECD estimates Source Korea Institute for Health and Social Affairs, Health and Welfare Indicators in Korea ISeoul Korea Institute for Health and Social Affairs. 2000). pp 436-37 274 1 Facing Aging and Providing Social Security Social security expenditures in Korea will inevitably rise, owing to the accel- erating process of population aging and growing demand for social welfare. The underprivileged and the elderly will call for more government protection as the economy grows. In addition, already plunged into the morass of family dissolu- tion and divorce fueled by the growing labor force participation of women, the elderly and children are bearing the brunt of the changing social mores that weaken the sense of familial responsibility Those factors will make welfare for children, the elderly, and family even more important in the years to come. Growing demand for social welfare will require the Korean government to raise social security expenditures. Social welfare areas in urgent need of expansion include public assistance and health A reasonable estimate of the proper ratio of gross expenditures on social security to GDP is 15 percent.24 The Korean gov- ernment will need to draw on the financial resources of both the public and the voluntary sectors to achieve that level Future Challenges for the Korean Social Security System The Republic of Korea has recently faced a bewildering array of social changes: population aging, the adverse effects of the 1997-98 economic crisis, and the high expectation of unification with the Democratic People's Republic of Korea. All this has called attention to the necessity of strengthening social security and the social safety net. The Republic of Korea will have to meet the welfare and health needs of its rapidly aging population. Over 90 percent of its elderly suffer from chronic diseases, and their financial situation is much worse than that of the younger generation.25 Because of the diminishing role of the family as an informal care mechanism, the buck of protecting the elderly has passed to society as a whole. The economic crisis brought an unemployment rate that is still lower than the OECD average but much higher than the precrisis level. Before the crisis the Republic of Korea was approaching full employment, with an unemployment rate of 2.0 percent, but its advancing economy is likely to face higher unemploy- ment rates. Social protection for the unemployed will thus become an ever more important issue, and the country will need to establish a broad social safety net to protect its population. The country also needs to prepare to raise the living standards of people in the Democratic People's Republic of Korea before taking further steps toward Social Security and the Social Safety Net in the Republic of Korea 1 275 unification, because the poverty level is much higher and the gross national prod- uct is much lower there than in the Republic of Korea. A unified Korea would require a welfare system buttressed by higher expenditures and an appropriate combination of social safety measures to protect needy people in the Democratic People's Republic of Korea. Existing welfare expenditures need to grow, and public sector participation in financing welfare funds needs to be strengthened. The current private sector contribution rate in Korea-26.5 percent in 1996-is much higher than that in other OECD countries (1993 rates- Germany-4.6 percent, Sweden-4.5 per- cent, the United States-3.0 percent, and the United Kingdom-1 0 percent).26 Meanwhile, the private sector should continue to help strengthen the social wel- fare system through volunteer activities and donations. In sum, in the midst of globalization and torrential social change, the Korean social security system needs to adopt new strategies and raise budget allocations to enhance programs, benefit levels, and coverage Notes I Yong-Ha Kim, J Suk, and S Yoon, Efficient Management of Social Insurance (Seoul Korea Institute for Health and Social Affairs, 1996), Kyung-Suk In, The Korea Welfare State Ideals and Realites (Seoul Nanam Publishing, 1998) 2 All data in this paragraph are from the Ministry of Labor, Report on Work Injury Comnpensation Insurance in 2000 (Seoul Ministry of Labor, 2001). 3 In, op cit 4 Ibid 5 Eunyoung Choi, J Kim, and W Lee, Health Care System in Korea (Seoul Korea Institute for Health and Social Affairs, 1998). 6 In Hyop Chang, Hye Kung Lee, and Jungsoo Oh, Social Welfare (Seoul Seoul National University Press, 1999) 7 Editing Committee of the White Paper on Welfare Reform, Welfare Reform iit Korea toward the 21st Century To Enhance the Quality of Life in the Globalization Era (Seoul Ministry of Health and Welfare, 1998) 8 Choi et al , op cit 9 Editing Committee of the White Paper on Welfare Reform, op cit 10 Mandoo Kim and H Han, Modern Social Welfare (Seoul Hong-Ik Jae, 2000) II Ibid 12 Suk-Myung Yoon, "National Pension," in Goal Setupfor Social Security Development and Current Issues (Seoul Korea Institute for Health and Social Affairs, 1998) 13 All data are from Chang et al , op cit 276 1 Facing Aging and Providing Social Security 14 Ibid 15 National Social Welfare Education Committee, Introduction to Social Welfare (Scoul National Social Welfare Education Committee, 2001) 16 Ministry of Health and Welfare, The Analysis of Livelihood Protection Recipients (Seoul Ministry of Health and Welfare, 1997) 17 Data are cited in Mee-Gon Kim, "Basic Livelihood Guarantee System as a Social Safety Net" (paper presentcd at a workshop on Securing Social Safety Nets, Korea Institute for Health and Social Affairs, Seoul, 2001) 18 Neng Hoo Park, "Productive Welfare " Health and Welfare Forum 60 15-25 19 Mee-Gon Kim, op cit 20 Ministry of Health and Welfare, White Paper on Health and Welfare (Seoul Ministry of Health and Welfare, 2000) 21 Byug-Ho Choi and K Ko, "Social Security Expenditure in Korea and Ways to Improve Its Level," Health and Welfare Forum 49 48-58 22 Korea Institute for Health and Social Affairs, Health and Welfare Indicators in Korea (Seoul. Korea Institute for Health and Social Affairs, 2000) 23 Choi and Ko, op cit 24 Ibid 25 Kyung-Hee Chung, Y Cho, Y Oh, J Byun, Y Byun, and H Moon, A National Survey on the Elderly Life a,id Their Welfare Need (Seoul Korea Institute for Health and Social Affairs, 1998) 26 W Ademan and M Einerhand, "The Growing Role of Private Social Benefits," Occasional Papers on Labour Market and Social Policy, no 32 (Seoul Korea Institute for Health and Social Affairs, 1998) CHAPTER 16 Italy's Pension Reform Process: Where Financial and Political Realities Meet by Mauro Mare and Giuseppe Pennisi ABSTRACT In Italy, various attempts at pension reform, necessitated by dramatic demographic change, were unsuccessful until 1992, when an external crisis forced action Political factors also played a vital role in suc- cessive reform measures, but their influence means that the process is far from complete. Despite significant differences in socioeconomic develop- ment, Italy's pension reform experience offers useful indications for East Asian countries that face the complex challenge of designing and develop- ing their own pension systems. Few policy areas present such complex political challenges. The interplay of events during economic crises pre- sents important opportunities. The Italian pension system of the late 1980s was a highly fragmented, pay-as- you-go scheme financed by payroll taxes and employers' and workers' contribu- tions The system linked benefits largely to earnings in the latest or best years of working life.' The 25 percent of the gross domestic product (GDP) that Italy devoted to public welfare expenditures was broadly in harmony with the European average. However, the over 100 different public pension schemes that absorbed 60 percent of welfare expenditures and approximately 13 percent of GDP were way out of line. The ratio of payroll taxes and contributions to wages and salaries was twice that in France and Germany and four times that in the United Kingdom, exert- ing a major effect on Italian labor costs and the competitiveness of Italy's com- panies Transfers from the active population to pensioners-the resources that active workers funded above those that pensioners had paid in-totaled about 2,500 trillion lire Such intergenerational inequity was socially unacceptable 277 278 I Facing Aging and Providing Social Security Organization for Economic Cooperation and Development (OECD) and International Monetary Fund (IMF) estimates indicated that without major reforms annual public expenditures on pensions would reach 21 percent of GDP by 2030 and that public pension-related debt would reach six times GDP by 2050, jeopardizing public and private spending on other social and economic activities. Econometric studies showed that such an imbalanced pension system would have severe implications for labor and capital markets and that this bur- den was a major cause of the Italian economy's sluggish growth. The successive and often tortuous measures that Italy undertook to reform its pension system2 can shed light on the complex political and economic reali- ties that countries in East Asia and other regions will face as they pursue their own pension systems and create a wider social safety net. Creating the Monster The grave problems confronting Italy's pension system had evolved over less than a century, as Italy moved from an agrarian to a postindustrial, high value- added, service-oriented economy. Each stage of economic growth and transfor- mation saw the introduction of a different pension system, and cumulative errors had created the severe financial and economic problems of the late 1980s. Italy's pension system developed hand in hand with interwar industrialization. As people moved from rural areas to towns and from agriculture to manufactur- ing, the aged could no longer rely on the extended family for support. The solu- tion was compulsory savings schemes that would provide an income when workers were no longer employed. Employers were required to contribute to these schemes because the consensus was that they should help support former employees after relying on the employees' human capital and skills during their working lives. The pension system developed into a fully funded occupational scheme simi- lar to those evolving around the same period in France and Germany.3 In paral- lel, labor legislation and practices provided-first de facto and later de jure- lifetime employmcnt in the growing Italian manufacturing sector.4 Pension mechanisms gradually extended to employment outside manufacturing and the civil service (the first sector to adapt the idea to its own needs), as commerce, banking, and self-employed workers and professionals, plus, naturally, farmers developed their own fully funded occupational pension schemes. Those schemes worked relatively well before World War II, sometimes even providing a surplus to the general budget because of greater-than-expected financial yields.5 Italy's Pension Reform Process Where Financial and Political Realities Meet l 279 However, World War II and its aftermath brought high inflation, turmoil in financial markets, and extremely low yields for occupational pension funds invested in real estate and government bonds. To avoid sudden impoverishment of those already on pensions, the scheme gradually became a mixed system with both fully funded and pay-as-you-go features. The system was still structured along occupational lines and mostly financed by payroll taxes and by employer and employee contributions. However, the system made growing claims on gen- eral taxation to help offset the financial and economic effects of the war. Haphazard changes-plus each occupational category's scramble to obtain a better pension and the shortsighted view that current payroll taxes and contri- butions could always be manipulated to meet the pension bill-helped create a serious "pension maze" from 1965 to 1969.6 This maze entailed serious inequities between workers of the same generation but belonging to different occupational segments and even more severe disparities between workers of dif- ferent generations. Despite these problems, in the final years of its "economic miracle" in the late 1960s Italy had a young and expanding labor force, high multifactor productiv- ity, a low unemployment rate (around 3 percent), and sustained GDP growth of about 5 percent per year Political parties and trade unions thought that a general overhaul of the system would help "ensure that, after 40 years of work and con- tributions, workers would be entitled to a pension based on 80 percent of the average wage of their last three years in employment."7 Thus, legislation intro- duced in March 1968 aimed to establish "the most advanced pension system in the world,"8 and a bill filed in April 1969 introduced a general old-age pension for all Italians over 65 with no other income. In the case of the general old-age pension, benefits were indexed to wage increases rather than to the cost of living so that pensioners would profit from rising productivity. The scheme allowed early retirement after 35 years of servicc-even sooner for working mothers Because the system based benefits on earnings in the last or best years of work- ing life and not on payroll taxes and contributions, the system was termed a pay- as-you-go, earnings-related system. Although it retained an occupationally based system, the legislation was designed to compensate for inequities within the same generation and thus temper political tensions and allow older workers to reap the benefits of the "economic miracle."9 However, the system could meet these objectives only through a series of technical corrections to the combination fully funded, pay- as-you-go scheme, and the changes took on a highly political cast. Workers' and employers' contributions were no longer seen as a means of financing the 280 1 Facing Aging and Providing Social Security system but as a levy on employers, often viewed as "exploiters" of the working class. Benefits were no longer linked to financial yields but to the broader aim of maintaining a standard of living that would favor workers with less success- ful working careers. Within a few years of the pension system overhaul, changes in demography and the labor market meant that the new system was not financially sustainable: the ratio between the active and nonactive population dropped from 4.6 to 1 in the 1950s to 1.2 to 1 in the 1990s. The system also included built-in incentives that would create new inequities and encourage evasion of payroll taxes and con- tributions: employers and workers would tend to avoid payments except during the years used to compute benefits. Those inefficiencies and inequities worsened as Italy's economic and social structure changed from large manufacturing groups to burgeoning small enterprises in both industry and services and as the labor market evolved from reliance on lifetime employment to mobility from firm to firm, sector to sector, and location to location. Why Reform Attempts Failed Reform from 1978 to 1992, which aimed at remedying the most unsustainable, distorting, and inequitable features of the 1965-69 system, offered a sobering object lesson. Ministries of labor and social security, which were motivated by far-sighted visions, rather than finance ministers, who tended to be preoccupied with shorter-term financial issues, generally took the lead, and the larger unions also played constructive roles. However, the negative outcomes of these efforts prove a basic theorem of neoinstitutional economics faced with abrupt and far- reaching changes, "old institutions" become more rigid and "path dependent"' 0 until a drastic exogenous determinant breaks them up. 1l During the years 1978-80, after a major study of the pension system by an independent committee that included representatives of trade unions and employers, Labor MinisterVincenzo Scotti attempted to raise the retirement age gradually to 65 years for men and 60 for women and to harmonize the nearly 120 "social security regimes." He also tried to modify the mechanisms allowing retirees to accumulate pensions and wages, to define new rules for self-employed workers, and to introduce incentives for private pension funds and other individ- ual retirement systems After 2 years of negotiations, however, the small and fis- cally sensitive Republican Party blocked this "rationalization," arguing that tran- sition costs would be too high because the program improved benefits for retirees Italy's Pension Reform Process Where Financial and Political Realities Meet 1 281 at the lowest income levels-a concession that Scotti had made to win support from unions and the left-wing opposition. In 1980 Labor Minister Italo Foschi raised the pension ceiling, streamlined payments, and established new procedures for computing supplementary pen- sions, but did not address the key issues of financial, economic, and social sus- tainability In 1982 Labor Minister Michele Di Giesi again tried to harmonize the 100-plus regimes and increase the pensionable age to 60 years for women and 65 for men. A bill drafted after consultation with unions and employers' associa- tions sailed through the Council of Ministers, but Parliament debated the dis- crimination that the measures would produce between different categories and generations of workcrs and dissolved before voting on critical sections of the bill. During the years 1983-87 dynamic and energetic Labor Minister Gianni De Michelis charted a great and general pension reform by drafting four successive bills that would raise the pensionable age and pension ceiling, establish new pro- cedures for calculating benefits, and create fiscal incentives to promote private pension funds; however, Parliament acted only to increase the pension ceiling and lengthen the period for computing average wages to 10 years. Labor Minister David Donat Cattin attempted similar reforms again in 1990, as did his succes- sor Franco Marini, a former trade union leader. The failure to approve similar proposals year after year cannot be laid solely at thc door of resistance to change. In the 1980s Italy's central policy focus was developing a noninflationary growth path after the inflationary no-growth path of the 1970s 12 This was not only a requisite for participation in European exchange rate agreements but also had high priority because of growing unem- ployment. Policymakers thus focused mostly on indexing of wages and salaries to inflation, and benign neglect of pension matters was often a means to forging agreement on those issues The 1992 Financial Crisis and Pension Reform A financial and foreign exchange crisis in 1992 proved to be a turning point in Italian development, abruptly shifting priorities to controlling public expendi- tures and reforming factor and product markets, including pension systems. 13 The Italian crisis had important similarities to the East Asian crisis of 1997-98. In Italy a pegged European exchange rate had not brought convergence of monetary and economic policies, creeping overvaluation of the exchange rate had undermined export competitiveness, and the economy relied on a high level 282 I Facing Aging and Providing Social Security of short-term financing in the international market. A series of judicial investiga- tions and trials among the Italian ruling class only reinforced market skepticism of the capacity of Italy's often inexperienced leadership to come to grips with the requirements for European monetary union. When Italy asked for a suspension from the exchange rate mechanism and let its currency float, the Italian lira depreciated by nearly 30 percent. Against this background the Italian government enacted a supplementary budget in 1992 designed to boost revenues and reduce expenditures to stem the looming crisis. Those measures prepared the ground for major pension reform, and Parliament gave the government authority to change the system without negotiating with employers and trade unions. Pension reform went hand in hand with reforms in other key elements of the social safety net, such as public health services and the labor market. However, the crisis-driven new policies lacked the long-term vision required for a sustainable, effective, and efficient new system. Thus, the 1992-93 "Amato reform" did not change the basic design of the pay- as-you-go system but revised benefits downward by raising the pensionable age to 65 for men and 60 for women and extending the minimum number of contribu- tory years from 15 to 20. The new system also based benefits on participants' entire working life rather than on the last 10 years and harmonized the yield on contributions to 2 percent a year, with a few notable exceptions for powerful occu- pational groups. Pensions were indexed to the cost of living rather than real earn- ings, with the possibility of future adjustments if higher productivity yielded great differentials between real earnings and inflation. Contributions were set at 26.5 percent of earnings for employees and 15 percent for the self-employed (and two-thirds for employers). The new indexing measure proved to be the most effective in reducing future pension expenditures. 14 However, the Amato reform also drastically heightened intergenerational inequities by sharply differentiating between participants con- tributing for fewer than 15 years and other workers. More significantly, the reform did not tackle the basic conceptual flaws of the pay-as-you-go scheme. Although it was saluted as the definitive pension reform-a standard for all European and non-European countries1 5-thc new government had to introduce a temporary freeze on implementation the very next year. The government again made pension reform a central item on its political agenda after the 1994 elec- tion, but the governing coalition dissolved around the proposal to abolish senior- ity pensions and drastically revise the benefit computation. Meanwhile, a new financial crisis was looming, caused by downward pressures on the exchange rate and upward pressures on interest rates 16 Those market pressures finally con- vinced policymakers to overhaul the system Italy's Pension Reform Process Where Financial and Political Realities Meet 1 283 The 1995-97 Reform and Other Options The 1992 reform was intended to improve the financial sustainability of the pen- sion system by abolishing indexation to real wages and extending the period of earnings on which pensions were computed. In contrast, the 1995-97 reform was aimed at stabilizing the impact of pension expenditures on GDP, boosting the efficiency of labor markets by reducing distortions and the "tax on labor," and enhancing the fairness of the overall pension system The 1995-97 reform significantly changed the basic functioning of the pen- sion system The government now strictly ties benefits to contributions paid dur- ing the working career (although the new rules apply only to contributions made after 1995). The reform thus shifted the system from a defined-benefit to a defined-contribution scheme. The pension is established by multiplying the bal- ance in each individual's account by an age-related conversion coefficient to cre- ate an annuity This coefficient can be modified every 10 years to reflect changes in life expectancy, rates of GDP growth, and earnings assessed for social security contributions. Workers may elect to begin receiving pensions between 57 and 65 years of age. Although the reforms of the last 5 years have introduced important changes in the Italian pension system, they will fully affect only people who began work- ing after 1995. The system still needs major structural reform that fully accounts for changing demographics and labor market conditions, stabilizing the ratio of expenditures to GDP. For example, the system could index benefits to increases in GDP rather than the retail price index. To equalize the rates of return for dif- ferent individuals, new rules could apply benefit computation to everyone, even those who are now exempted. Those measures would leave the basic design of the 1995-97 reform largely unchanged. Another possibility is to move to a mixed-pension system by allowing people to shift some contributions to private and occupational pension plans. Those options would strengthen property rights and sharply limit political interference in the system. A two- or three-pillar system would also diversify the economic, demographic, and financial risks of accumulated pensions and increase workers' commitment to the rules of the game Fiscal and monetary incentives could encourage people of retirement age to continue working and thus curb the irreg- ular economy, boosting employment and national income. Another possibility is a fully funded government-run pension system pat- terned after a compulsory provident fund. But that option has major flaws because it would require a 70-year transition period and initially higher overall contributions. A more significant flaw is that, at the eventual steady state, such a 284 I Facing Aging and Providing Social Security system would pose enormous corporate governance issues, because the new fund would become a major stockholder in all Italian large and medium-size firms All these reform options are based on moving from a one-pillar, pay-as-you- go-system to a two- or three-pillar system with a greater role for fully funded mechanisms. East Asian countries are now attempting to develop pension sys- tems anchored on fully funded mechanisms However, the conventional wisdom concerning the advantages and costs of such systems seems to be flawed Can a fully funded system guarantee higher real returns? The most frequently cited advantage of a fully funded system is that it offers a return on investment in the capital market. Between 1960 and 1990, most pay-as-you-go systems offered much higher returns to the first generation because they had not reached maturity and benefited from high GDP growth rates and rising contributions. Recent studies show that pension funds invested in financial markets are more likely to provide high real rates of return over the long term.'7 However, the advantage tends to fade if we also consider the taxes that the generation in tran- sition to the new system will have to bear.'8 Will a fully funded system improve economic welfare for society as a whole? Some analysts have built models that generate efficiency gains from a transition to a funded system,i9 whereas others maintain that a welfare-improving transi- tion to a funded system is impossible.20 Gencrally, the latter is more plausible. Advantages for future generations tend to be somewhat offset by costs to present generations A more interesting question is whether full funding can increase national savings, the stock of capital, and growvth in financial markets, and thus boost national income. If the pie becomes bigger through growth of financial markets and the national income, this could relieve distributional conflict between generations. Does full funding insulate the system against demographic shocks? An aging population means that pension funds will have to sell their assets to a smaller active population, undercutting the price of the assets and the resources avail- able for pensioners. Lessons for Reform The Italian experience shows how pension reforms quickly reach a standstill if they are not part of a broader reform of the welfare state and labor markets. Attention to pension reform in Italy was essential, but it diverted focus from tax and transfer mechanisms, which target poverty and people of working age. Since Italy's Pension Reform Process Where Financial and Political Realities Meet l 285 1995 modest attempts to remedy this situation have been unsuccessful, and little has been learned because a well-functioning monitoring and evaluation system did not exist. "Less pension, more welfare" is a call for a broad and deep revamping of social policies22 that must include labor market reform to enhance flexibility, decentral- ized wage bargaining, and a breaching of barriers between the formal and informal employment sector.23 Only such broader adjustments in social policy can prevent pension reforms from coming to a halt, given other elements of the welfare state. The recent white paper from the Italian government on labor policies augurs wel.24 If the Italian pension system moves gradually from an essentially one-pillar, pay-as-you-go system to a two- or three-pillar system with a growing fully funded leg, individual savings accounts should integrate the fully funded element with unemployment insurance-an approach that might also prove useful in Asia 25 Pension and welfare systems result from a delicate balance of economic, social, and political power. Shocks may forestall reforms, but they can also accelerate them. Students of market-supporting institutions know that several large shocks are sometimes needed to bring change.26 Crises sometimes provide policymakers and politicians with opportunities to undertake bolder institutional reforms Timor Kuran shows how crises can break through "public lies" in economic and social policies, especially in sensitive sectors such as pension and welfare sys- tems.27 Dani Rodrik shows how crises forestall conservative attempts by "political losers" to avoid change in obsolete institutions and can be a lever to create high- quality institutions.28 In Italy, as in other European countries, the 1992 foreign exchange and finan- cial crisis and the path toward European monetary union gradually changed unions' outlook, which was often a stumbling block in pension, welfare, and labor market reform This change entailed a move from a corporatist and conservative posture favoring "old institutions" of the "old economy" to broader, more socially responsible attitudes geared toward easing the transformation to a flexi- ble, high value-added economy.29 Despite these changes, "nurturing" reform remains a difficult process. Momentum for definitive, longer-term pension reform in Italy has gradually wvaned. Policymakers took no substantive action after the 2001 checkup, and none of the possible models, nor any combination, is truly on the table. However, advo- cates of reform could seize the recent world economic slowdown and fear of a new crisis as an opportunity. For East Asian countries in particular, the recent crisis offers opportunities for reform advocates to establish an efficient, effective, sustainable, and equitable social safety net. 286 1 Facing Aging and Providing Social Security Notes 1 M Ferrera, II welfare state in lItalia (Bologna, Italy Il Mulino, 1984), and M Ferrcra, Le trappole del welfare (Bologna, Italy II Mulino, 1998) 2 For an exception, see G Cazzola, Lafabbnca delle pensions (Rome Ediesse, 1992) 3 P Flora and A J Heidenheimer, eds , Lo sviluppo del welfare state in Europa e in America (Bologna, Italy II Mulno, 1983) 4 Giuseppe Pennisi, La guerra des trentenn Italia e nuove generazioni (Rome Ideazione Editrice, 1997) 5 M A Coppini, Le ragiont dello stato sociale (Rome Hediese, 1994) 6 0 Castellino, 11 labirmto delle pensiont (Bologna, Italy II Mulino, 1975) 7 Maurizso Ferrera, Modelli di Solidarseta (Bologna, 11 Mulino, 1993) 8 Act 238, March 18, 1968 9 Ferrera, II welfare state in Itah/a, op. cit 10 Douglass C North, Institutions, Institutional Change, and Economic Performance (Cambridge, England Cambridge University Press, 1990) 11 Albert 0 Hirschman, Comefarpassare le riforme (Bologna, Italy II Mulino, 1990), and Albert 0 Hirschman, Retoriche dell'intransigenza (Bologna, Italy II Mulino, 1991) 12 A Graziani, Lo sviluppo dell'economia italiana dalla ricostruzione alla muneta unica (Torino, Italy Bollati Bonnghieri, 1998) 13 A Monorchio, ed., Lafinanza pubblica dopo la svolta del 1992 (Bologna, Italy II Mulno, 1996) 14 Ministero del Lavoro e delle Politiche Sociali, Venfica del sistema prevsdenzsale as sensi della legge 335/1995 e successivi provvedinenti nell'ottica del/a competltivitd, dello svsluppo e dell'equstd (Rome Commissione Brambilla, 2001) 15 Instituto Nazionale della Previdenza Sociale, Lepensiont domanm (Bologna, Italy II Mulmo, 1993) 16 M Tivegna and G. Chiofi, ANews e dinamica des tassi di cambio (Bologna, Italy II Mulino, 2001) 17 J Siegel, "The Shrinking Equity Premium Historical Facts and Future Forecasts" (mimeo- graph, 1999), P Jorion and W Goetzman, "Global Stock Markets in the Twentieth Century," working paper no 7565, National Bureau of Economic Research, Cambridge, Mass, 2000, and W Goetzman and P Jorion, "A Century of Global Stock Market," working paper no 5901, National Bureau of Economic Research, Cambridge, Mass, 1997 18 The same story applies if we finance the payment of existing pension rights by issuing public debt Someone has to pay the interest related to this debt, and this is the same as new taxes 19 M Feldstein, Privatizing Social Secunty (Chicago- University of Chicago Press, 1998) 20 P Diamond, "Towards an Optimal Social Security Design," working paper no 4, Cerp, Torino, Italy, 2001 21 Or if elderly people demand more goods than are currently produced by active people-pen- sioners' desired consumption exceeds desired savings by workers-this will cause price inflation, reducing the purchasing power of pensioners' annuities 22 Tito Boeri and Roberto Perotti, "Less Pensions, More Welfare," (paper presented at the Innocenzo Gosparini Institute of Economic Research conference "Pensions davvero una ven- fica>," Rome, September 28, 2001) 23 Pennisi (1997), op cit. Italy's Pension Reform Process Where Financial and Political Realities Meet 1 287 24 Ministero del Lavoro e delle Politiche Sociali, Libro bianco sul mercato del lavoro in Italia, "Proposte per una societa attiva ed un lavoro di qualita," Rome, 2001 25 Joseph Stiglitz and J Yun, "Integrating of Unemployment Insurance with Pension through Individual Savings Accounts" (mimeograph, 2001) 26 Raghuram Rajan and Luigi Zingales, "Financial Development and Growth," American Economic Review 88, no 3 (1998) 27 Timor Kuran, Private Truths, Public Lies The Social Consequences of Preference Falsification (Cambridge, England Cambridge University Press, 1995) 28 Dani Rodrik, "Institutions for High Quality Growth What They Are and How to Acquire Them," working paper no 7540, National Bureau of Economic Research, Cambridge, Mass, 2000 29 G Bertola, T Boeri, and G Nicoletti, Velfare and Employment in a United Europe (Cambridge, Mass MIT Press, 2001), and T Boeri, A Brugiavini, and L Calmfors, The Role of the Unions in the 21st Cenltury (London Oxford University Press, 2001) CHAPTER 17 Social Insurance for Aging Rural Households: A Comparative Perspective by Luis Frota ABSTRACT This chapter reviews the development of old-age social insurance schemes for farmers in six European countries. Declining numbers of agricultural workers have obliged European governments to provide special incentives to renew farming communities and to expand pensions for elderly agricultural households These efforts have met con- siderable success, although pensions are still lower in rural areas than else- where. East Asian countries face similar challenges as they expand social insurance in agricultural communities. Europe's agricultural landscape has experienced dramatic changes during the last 40 years, as the rural population, the number of farms, and farm employ- ment have dropped significantly. Agricultural employment declined by 30 per- cent over the past 12 years, and only 10 percent of the total population is rural today compared with 40 percent in the 1950s. East Asian countries, too, have experienced rapid dislocation of agricultural populations. Consequences are par- ticularly severe for older farmers This chapter examines the evolution of social insurance schemes for the rural elderly in six European countries-Austria, Finland, France, Gcrmany, Greece, and Poland-and contrasts their experience with that of seven East Asian coun- tries-China, Indonesia, Republic of Korea, Malaysia, the Philippines, Thai- land, and Vietnam. The chapter suggests ideas for future research and action based on the European experience. 289 290 1 Facing Aging and Providing Social Security Changes in Europe's Rural Economy Europe's Common Agricultural Policy, established by article 39 of the Treaty of Rome in 1957, aims to stabilize incomes and provide equitable living standards for the agricultural population. Productivity increases are the main achievement in 1998 one German farmer could feed 124 consumers-about seven times as many as in 1960. Rising productivity has meant that average incomes of farming households are now comparable to national household averages However, dis- parities in farm incomes have grown. Studies show that persistent low agricul- tural incomes are related to farm size; smaller farms run by the elderly tend to have low incomes.' Farmers with low income tend to seek new income sources, so part-time farming, multiple jobs, and occasional work are on the rise in Europe. These new work patterns are emerging as profitable new uses of land bring new wealth to rural settings. European countries with civil law traditions-including the six that are the focus of this chapter-have sought to increase the average size of family farms as well as to raise productivity (see Table 17.1). As a result, the average size of German farms increased from 9.3 hectares in 1960 to 33.4 hectares in 1998, and the average Finnish farm size grew to 23.7 hectares in 1997. Greece and Poland are exceptions, with more than 50 percent of their farms still below 5 hectares. These countries made family farming, land planning, and spatial planning pol- icy priorities, along with agricultural modernization, during the postwar period.2 Countries with common law traditions, such as Denmark, the Netherlands, and the United Kingdom, subject land use and its transfer to market rules and less regulation, which has discouraged traditional family farming. Civil law coun- tries, in contrast, provide substantial guarantees to farmholders, allowing them to pass on leased property to heirs, for example, encouraging long-term invest- ments.3 Ownership and direct management-the norm in 1950s Europe-still constitute the bulk of agricultural activity, but leasing represented 65 percent of cultivated area in France in 1997, compared with 50 percent in 1970. France, Germany, and other countries have introduced special provisions to facilitate family continuity on farms and smooth transitions in rural areas. The French civil code, modified in 1980, allows family members 18 years of age or older to claim accumulated revenues as an inheritance share, on the basis of their participation in farm work if that work was not previously remunerated.4 Finland and France have provisions favoring new young entrants to farming and grant- ing them loans and assistance, thus minimizing the financial sacrifice that young entrants must make to gain access to land. Social Insurance for Aging Rural Households A Comparative Perspective 1 291 Table 17.1 Number and Area of Holdings in 1997 Percentage of holdings Farm size Austria Finland France Germany Greece Poland 0-5 hectares (%) 37 2 83 263 312 76 2 553 5-10 hectares (%) 18 9 159 92 14 6 142 25 5 10-20hectares(%) 224 300 111 170 66 151 20-50 hectares (%) 172 36 8 23 5 23 0 26 3 7 > 50 hectares (%) 41 88 299 142 04 04 Total hectares 100 0 100 0 100 0 100 0 100 0 100 0 Average size (hectares) 163 23 7 417 321 43 80 Source Statistical Office of the European Communities, "Agriculture in the Union, Economic and Statistical Information" (Luxembourg Eurostat. 20001, data for Poland from "Agricultural Census of 1996" IWarszawa 1998), 1 5-16 Despite these efforts, rural aging problems loom even larger when family ownership is high because a significant proportion of agricultural households are elderly and the average number of family members per household has dropped, both reflecting the urban exodus of the younger generation. A particularly acute problem is elderly households with a single head of household, often female European demographic figures show that aging has progressed more rapidly in rural than in urban areas and that pension demands are likely to become more acute in light of sharply dropping birthrates. A declining active rural labor force and rising numbers of elderly farmers over the past five decades pose special challenges for social protection of the elderly population Until the mid-1990s, dependency ratios rose because of growing numbers of pensioners, a shrinking active labor force, and early retirement schemes Although farmers' retirement has since stabilized and began to drop in the mid- 1990s, the active labor force continues to decline (see Table 17.2). Table 17.2 Old-Age Dependency Ratios in Selected European Countries Austria Finland France Greece Poland (1991) (%) (1996) (%) (1990) (%) (1991) (%) (1996) (%) Urban (% 33 4 290 30 4 27 9 23 2 Rural (%) 29 6 344 396 40 6 30 4 Source United Nations Department of Economic and Social Affairs, Statistical Office, United Nations 1997 Demographic Yearbook (New York United Nations, 1999) 292 1 Facing Aging and Providing Social Security In most European countries, farmers have remained a powerful social and political force even as agricultural employment has dropped. Farmers have long worked cooperatively to insure themselves against risks, exemplified by accident mutual funds in both civil law and common law countries. In contrast to com- mon law countries, European countries with civil law traditions instituted dis- tinct social security systems for farmers after World War II, reflecting relatively strong political representation of farmers How Agricultural Insurance Evolved in Europe Over the last 40 years, rural social security systems have focused on two comple- mentary tasks to ensure a steady rise in agricultural households' living standards and to protect the older farmers during the reforms designed to bring this about Over this period Europe has seen complex reforms in rural insurance schemes, the achievements of those reforms need to be read against the schemes' objec- tives and their degree of maturity. Figure 17.1 summarizes the evolution of rural pension schemes. Most European countries have harmonized rules for these systems with those of other professional groups; this convergence allows countries to take into account new social risks, such as atypical working careers. Workers typically ben- efit from individualization of rights and greater possibilities to transfer them Agricultural insurance schemes reflect differing philosophies as well as vary- ing circumstances and historical paths. For example, France has focused more explicitly on the problems of poor older people in rural areas, whereas in Germany the modernization of agriculture was the principal policy priority. After unsuccessful efforts to introduce a voluntary pension scheme before World WVar II, French farmers after 1945 resisted government attempts to imple- ment a nationwide farm pension system because they were powerful and would also be net contributors to the scheme.5 Nevertheless, the French government moved to introduce a basic agricultural pension scheme in 1952, reinforced by a contnbution-based system in 1955. In 1990 the farm pcnsion scheme underwent a major overhaul, with contributions now paid on the basis of reported income instead of imputed income based on region and type of crop, making the system more comparable to other professional schemes. Family-based coverage of farm- ers is meanwhile diminishing as working women demand individual rights. These successive reforms have brought minimum agricultural pension levels to national levels and good returns on investment, especially for lower-end Figure 17.1 Development of Agricultural Pensions in Selected Countries in Europe Proportional Supplementary contributory Separate body Individual pension pension SVB women's pension 1958 1971 1974 1992 Austria * * Basic Professional universal proportional pension pension 1956 1970 Finland A A Division of Women access Proportional rights between proportional pension men and women pension 1954 1992 1999 France p* 0 * 0 Flat rate Reform of the pension contribution basis -Reform of pension's calculus 1952 Old-age pension 1990 -Individualization of women's pension 1957 1995 Germany Proportional Flat rate contributory pension pension 19S1 1998 Greece Rights for Separate body Farmer's old age spouses and for farmer's old-age pension (ZUS) family aides pension (KRUS) 1978 1983 1990 Poland 1950 1955 1960 1965 1970 1975 1980 1985 1990 15 20 25 20 2 ts1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 2020 294 1 Facing Aging and Providing Social Security contributors. Agricultural pensions are nonetheless still below the national aver- age. Equity between contributors is also a concern because the current system is not fully proportional to incomes. Larger-income earners criticize the system for its complexity and the possibility of political manipulation. In postwar Germany rapid demographic changes called for swift policy inter- vention in rural areas. Policymakers opted for an unfunded pension scheme, seeking to shorten the time lag between contributions and entitlements. Early attempts to extend coverage through private voluntary insurance proved insuffi- cient because only large farmers applied for the scheme. The Old Age Pension Law went into force in 1957, and in 1972 older farmers became formally eligible for a full pension Those reforms have reduced the average duration of farm- holders' contributions and produced gender equity. However, the scheme has accentuated inequalities between high-performing large farmholders and average- performing smaller farmholders because it provides relatively low, basic, flat rate benefits Farmers can seek private insurance and supplementary coverage upon transmission of the farm. The Finnish welfare state introduced a flat rate universal pension in 1956 that benefited farmers-especially small farmers, who were well represented in the Central Party (Keskustapuole-Kesk), an old agrarian party. A scheme for farm- ers established in the 1970s has achieved full individualization of rights, high lev- els of equality (higher contributions for higher incomes), rights equivalent to those of other professional groups, and stability and simplicity of rules. In Austria in the 1950s, as family succession on farms began to change, a pen- sion scheme was seen as a way to complement farm owners' obligation to take care of retired farm people and to meet appropriate retirement standards. A 1958 law instituted supplementary old-age insurance for farmers, Austria introduced a contributions-based social insurance scheme in 1971, and women gained full pension rights in 1992. Greece introduced its Agricultural Social Insurance Organization system in 1961, which included old-age health care and insurance against diseases and bad weather. However, this system was marked by low coverage, noncompliance, and low pensions (although it did provide important supplementary income for very low end income earners). A major overhaul in 1998 integrated two separate schemes a flat rate pension and a contributions-related fund created in 1987. Finally, Poland first introduced agricultural social insurance in 1978 under the ZUS (main statutory social insurance scheme) general scheme, and a 1983 amendment extended coverage to spouses and family aides. Only in 1990 was a separate scheme for farmers introduced. The Polish scheme is said to provide fairly generous coverage; its benefits are rather low but are above poverty levels. Social Insurance for Aging Rural Households A Comparative Perspective 1 295 However, the scheme is administratively heavy, and its financial sustainability is in question (complianice is an important issue). Characteristics of Old-Age Agricultural Social Insurance By the early 1970s, after the creation of the Finnish professional pension in 1970, Europe's agricultural old-age pensions were fully elaborated. Still, debate about the adequacy of old-age benefits for farmers gained momentum from the mid- 1980s. In 1986 the European Parliament called on the European Commission (EC) to examine specific agricultural social security schemes.6 The EC lacked a juridical basis for action, but 6 years later an EC regulation sought to provide additional resources to enable older farmers to leave their farms. These early- retirement schemes have helped accelerate reform and change all over Europe. European old-age agricultural schemes have many similarities and have con- fronted common problems. Coverage and Requirements Family coverage is the general rule. Spouses are automatically covered even if they do not work on the farm, but other family members must work on the farm from age 18 (16 in Poland) Children are normally covered. Gender Issues To a significant extent, agricultural social security was designed for a male bread- winner family model that no longer holds. Since women's life expectancy is higher than men's, the social risks facing rural women are great and growing. In Germany the pension scheme covered less than half of female-headed farms because the farms were smaller and women's labor, though significant, generally was not visible. In traditional social security settings, women depended on (usually male) farmholders' capacity to pay their contributions. A complete transition to indi- vidual rights thus needs to ensure that whoever pays women's contributions does not have the power to decide whether it is appropriate to do so Finland has achieved this by introducing a mandatory split of rights, unless both members of a couple decide otherwise. That is, women are automatically considered cohold- ers of farms. In contrast, Polish female farm workers' situations have deterio- rated in the past 10 years. Women have reportedly withdrawn from labor markets 296 1 Facing Aging and Providing Social Securty both voluntarily and involuntarily, which may increase their vulnerability to poverty. Retired women receive only 74 percent of men's benefits. Austria sought to address gender bias in agriculture in the 1990s by giving women equal rights to a contributions-related pension.7 In 1999 France granted spouses the rights already granted to family aides. Today women in France can divide retirement rights, contribute to a proportional pension on the basis of the minimum income scale, or cohold with their husbands. Minimum Farming Area A minimum farming area is usually part of the benefits eligibility formula. France and Germany establish a minimum area deemed sufficient to allow sub- sistence, and the social security scheme automatically covers farms above half the minimum area. The minimum social threshold is around 25 hectares in France and around 4-5 hectares in Finland and Germany. Poland specifies a minimum of I hectare. Austria and Greece do not specify minimum farming areas Instead, those countries' schemes automatically cover all farm households whose main activity is agriculture. Greece defines main activity as time spent or main income source and covers people living in areas with fewer than 5,000 inhabitants regard- less of the origin of their income, except for salaried workers already covered by the main regime 8 Austria defines a threshold of 1,526 euros in annual farm worth for social security purposes All contribution formulas today are based on proven income, except in Ger- many and Poland. Germany's contributions are flat rate, with a contribution allowance for incomes below 30,000 deutsche marks (15,339 euros) per year. Around 50 percent of the insured benefit from this allowance Before 1995 the sys- tem estimated agricultural income based on land size and other regional factors, but it has used fiscal income since 1995. Poland at first implemented an income- related contribution based on farm size or income from certain products. However, this scheme generated opposition from farmers because their benefits did not relate directly to their contributions Today contributions are flat rate but are considered very low, seen as an "eligibility criterion" rather than as a true contribution. Compliance Issues Compliance issues have arisen in collecting contributions. For example, Greece tried to set up a pension system in which farmers would contribute up to one- third of the costs However, sluggish progress in collecting contributions and Social Insurance for Aging Rural Households A Comparative Perspective 1 297 disproportionate administrative costs led to failure. Compliance problems in more-developed agricultural schemes are usually associated with underreport- ing of income, a common feature with other self-employed activities, which results in mistrust from other professional groups. In Germany less than 10 per- cent of contributors experience payment difficulties, most stemming from excep- tional diseases or other risks as well as chronic losses characteristic of some crops and farm sizes. The government may exempt farmers from contributing if they experience exceptional, large income downturns. In France, Finland, Germany, and Greece farmers are eligible for old-age benefits at age 60. In Austria and Poland men are eligible at age 65, whereas women can claim a pension from age 60. The schemes usually have a vesting period of 15 years or more. France grants the basic flat rate pension after 15 years of contributions but requires 37.5 years for a full income-related pension, whereas Poland requires 25 years for full entitlement. Most countries, except Finland and Greece, require farmers to leave the farm on retirement Minimum Guarantees Countries vary in their minimum guarantees. Germany and Poland offer flat rate pensions, while Austria, Finland, and France offer income-related pensions with a guaranteed minimum. Austrian and Finnish pensions are easy to compute, with each year of income yielding a percentage of future pensions, but the French sys- tem is more complex and less proportional because it awards points for different income scales This system improves low-end pensions at the cost of high-end earners, who are demanding a second pillar to top up the mandatory scheme The German system does not guarantee a minimum benefit if the contribu- tion record is incomplete. Austria and France both offer a compensatory allowance that covers the difference between a national minimum and actual benefits In France, old pensioners with short contribution histories and those who contributed when the scheme was less generous used to be overrepresented in this national minimum scheme Polish benefits are aligned with minimum pensions for other workers, but farming pensions may fall below poverty thresh- olds as average workers' wages and benefits grow, and farmers are left behind.9 Finnish farmers tend to be overrepresented in the universal basic pension because they have lower incomes and shorter contribution periods than people in other professions. These characteristics of old-age agricultural insurance schemes inevitably challenge interprofessional equity. For example, German farmers pay 80 percent 298 1 Facing Aging and Providing Social Security of the contributions of other workers for an equivalent pension. Almost 50 per- cent of farm pensions are subject to contribution allowances that raise benefit- contribution ratios On average, French farmers will have benefit-contribution ratios equivalent to those of other workers as new regulations reach maturity in 2028, but low-income earners will be relatively better off. The contribution- income ratio for Finnish farmers seems comparable to that of other self-employed groups and categories of workers. Because farmers now rely mostly on basic pen- sions, they tend to have better benefits than other groups, but this difference will become marginal as farmers' incomes and years of contribution grow. Financing Governments pay a substantial share of agricultural social security budgets: about 70 percent in Austria and Germany, about 75 percent in Finland, and over 90 per- cent in Greece and Poland. France is an exception: direct state financing repre- sents only 30 percent of the agricultural social security budget; 58 percent comes from "demographic compensation" by other professional schemes. State support compensates for differences in income per capita among different sectors. Agricultural social insurance schemes display different demographics from those of general schemes. The number of pensioners has begun to decrease, and there is no "baby boom" effect on the horizon, so dependency ratios have stabilized. However, a significant generation gap still demands provisions for intergenerational equity that also recognize that elderly people today con- tributed to the system during a time of inadequate income levels and insuffi- cient coverage. The sustainability of farmers' pensions remains under debate, notably in Germany Thus the systems will have to address the needs of a "new breed of farmers" while continuing to care for elderly farmers. Governance of Pension Systems Farmers' associations have generally been deeply involved in designing and run- ning old-age insurance schemes. Farmers elect representatives to these associa- tions, except in Greece and Poland where governments nominate candidates. In Austria the powerful chambers of agriculture, which farmers elect at the provin- cial level, until recently appointed the representatives. In Germany farmers elect regional representatives to a general assembly every seventh year, and those Social Insurance for Aging Rural Households A Comparative Perspective 1 299 regional representatives elect the presidents of the administrative boards of 20 regional social unions. France maintains a similarly decentralized system, in which 76,000 elected peers (one for every nine insured people) participate in decisionmaking at both the local and national levels. These farmer social security associations enjoy some genuine autonomy at regional or provincial levels in Austria, France, and Germany. In Finland the national professional institution has a lighter structure and is composed of the central agency. Though power rests at the central level, elected peers play an important role at the local level, as they are scattered throughout the country. Poland sees the strengthening of worker involvement in managing pension insti- tutions as a prerequisite for further sustainable development of the scheme Most schemes are still organized around the national structure, which plays an impor- tant operational and strategic role, pooling funds, guaranteeing uniformity and concerted action, and helping spread best practice (see Table 17.3). Old-Age Social Insurance for Farmers in East Asia Demographtc trends in East Asia are less stark than in Europe, but East Asian countries face similar aging and a changing balance between rural and urban areas (see Table 17.4). The old-age dependency ratio is likely to double within the next 30 years.t0 In 1985 East Asia represented 28 percent of the world's pop- ulation aged over 60; by 2025 this figure will be 58 percent. In China, aging in regions with low economic growth is three times the national rate l East Asian farmers are predominantly nonwage workers Only around one- quarter of registered holdings belong to those who engage fully in agriculture, family members living on the other three-quarters of registered holdings, which are small, rely on a combination of income sources or live in great distress. 12 The shift toward export agricultural production, capitalistic land holdings, and higher internal mobility is integrating East Asia's agriculture into the global market. Notwithstanding potential benefits, the growing market character and globaliza- tion of agriculture are likely to increase economic vulnerability Economic inse- curity will particularly affect older farmers who use their plots of land as old-age security, because their plots are often too small and unproductive to remain inter- nationally competitive. The 1997-98 Asian crisis highlighted these issues. The crisis reversed rural-urban migration patterns. urban workers returned to their villages and plots of land, thus increasing the number of informal farmers. A more lasting Table 17.3 Administration, Governance, and Risk Coordination of Selected National Agricultural Social Insurance Agencies Risk integration Representatives from the Elected Role of the central One-stop Agriculture Mutual funo' profession administrators Decentralization or national level shopping insurance Austria (SVB) Yes Yes Administration elected Unified system Strategic, pool of funds Yes No indirectly by chambers of agriculture Finland (MELA) Yes Yes Administrators National body Operational role Yes No elected France (MSA) Yes Yes Peers elected at Significant operating Strategic, pool of funds Extended No (until 1996) regional level autonomy at the department level Germany (GLA) Yes Yes Peers elected at Significant regional Strategic, pool of funds Yes No regional level operating autonomy Greece (OGA) No Yes No elected Network of local High level of centralization No (many different Yes administration agencies (agriculture (state agency) delivery channels) credit bank, rural cooperatives, and others) Poland (KRUS) No Yes No, representatives National body with Centralized activity Yes No chosen by the own regional network government a Policyholders' contributions pay for the administration of mutual organizations, which are nonprofits managed by farming peers Note SVB = Sozialvericherunganstalt der Bauern (Austria), MELA = Maatalonsyrirttlien Elaikelaitos (Finland), MSA = Mutualite Sociale Agricole (France), GLA = Gesamtverband der Ladwirtschaftlichen Alterskassen (Germany), OGA = Office General d'Assurance des Agrtculteurs (Greece), KRUS = Kasa Rolniczego Ubezpiezenia Spolecznego (Poland) Social Insurance for Aging Rural Households A Comparative Perspective 1 301 Table 17.4 Old-Age Dependency Ratios in Rural and Urban Areas in Selected East Asian Countries China Indonesia Korea Malaysia Philippines Thailand Vietnam (1990) (1995) (1995) (1991) (1990) (1990) (1992) Rural(%) 141 130 290 121 105 113 145 Urbanl%) 117 91 100 87 87 90 140 Source Ronald Skeldon, Ageing of Rural Populations in South-East and East Asia Rome FAO, 1999) consequence came with pressures for financial stabilization that led to general budget cuts. This shifted the burden of social security-particularly in countries without comprehensive social security coverage-to private and professional organizations Available evidence on land size and the extent of subsistence agriculture in most East Asian countries tends to underscore that the potential for capital accu- mulation for a large part of the farming population is very limited. Without such accumulation many older farmers will likely have to continue to work to make ends meet For example, 44.5 percent of rural male Thais over age 60 worked in 1997, while only 29.4 percent of their urban counterparts remained active.13 Private landownership is now widespread in East Asia after successful land reforms, though inequality in access to land persists in some regions Tenancy has been declining in favor of salaried work, though diverse tenure forms remain. Land lease increasingly tends to take the form of stable and long-term contracts. Aging of the rural population has a gender dimension in East Asia, because women tend to outlive men-further increasing their vulnerability to poverty. For example, in Thailand in 1993 women averaged smaller agricultural holdings, except in the 20-24 age category '4Thailand, the exception, is interesting because it illustrates a paradoxical characteristic of Thai society In keeping with matri- lineal transmission of property, the elder daughter receives part of her parents' property when she marries, but women may be "ashamed of taking and manag- ing their husbands' due ",5 Another reason for women's greater poverty is that they are less likely to remarry after separation or the death of their spouses. The percentage of widowed-divorced-separated persons is higher for rural women than rural men in Thailand i6 In East Asia community cohesion tends to be stronger in rural than in urban areas Farmers' organizations play important roles in many settings. Korea's diverse and specialized farmers' organizations, for example, have helped bring 302 1 Facing Aging and Providing Social Security modern techniques to farmers. Malaysia presents a similar pattern of diverse farmers' cooperatives and associations, with the addition of a central farmers' organization that covers around 80 percent of farmers.17 Mutual fund arrange- ments have helped build local solidarity in Indonesia, though these grassroots farmers' organizations provide weak protection for local populations.'8 The pro- liferation of voluntary farmers' organizations in the Philippines highlights a lack of organizational capacity beyond local boundaries. In China collective farming included all farmers for many years. As independent agricultural activities grew from the mid-1980s, provision of welfare became more problematic. The new rural welfare system, still state organized, is decentralized and participatory, in keeping with the tradition of local participation in rural provinces. Vietnam's once highly centralized farming system appears to be decentralizing rapidly. Traditional family support remains a primary source of social security in rural areas. In China, for instance, 1987 data show that children supported around 61 percent of people age 60-74 in rural counties, compared with 21 per- cent and 16 percent in towns and cities However, with rapid demographic change, such informal protection mechanisms are eroding. Table 17.5 shows sharp differences between Korea and France in living patterns of the elderly, reflecting a lack of means for subsistence other than family support in Korea High levels of co-residence and material support from the younger generation remain the norm in the Philippines, Thailand, and Vietnam (see Table 17.6).19 Korea and the Philippines maintain mandatory schemes for farmers covering retirement and survival risks, whereas China and Vietnam have developed volun- tary schemes. Malaysia allows voluntary participation in its Provident Fund but is studying the possibility of extending mandatory coverage under a special fund for farmers. Indonesia has no equivalent formal social security provision (see Table 17.7). Table 17.5 Living Arrangements of the Elderly (Over 65 Years) Living in Lone Elderly Living with children, institutions (%) elderly (%r couple (%r relatives, or others (%P France 5 00 32 2 54 3 13 4 Korea 0 29 9 6 13 3 77 1 a Elderly people living in domestic households Source Didier Jacobs, Social Welfare Systems in East Asia A Comparative Analysis Including Private Welfare (London Centre for Analysis of Social Exclusion. London School of Economics, 1998), data are from 1995 Social Insurance for Aging Rural Households A Comparative Perspective 1 303 Table 17.6 Prevalence of Family Support for the Elderly (Age 60 or Older), percent Vietnam Red River Ho Chi Minh Support parameters Philippines (%J Thailand (%M Delta (%) City (%) Co-residencea 70 6 72 5 74 0 85 9 Quasi-residenceb 53 2 64 7 - - Monetary support 'Any amount 86 6 881 Substantial amount 46 6 69 1 - - Material support 88 7 89 2 83 9 72 2 Social contact Monthly or more 57 8 756 - - Weekly or more 72 3 883 - - a Among respondents with 1+ adult b Among respondents with 1+ non-co-resident child Source Robert Holzmann, op cit See also Association Internationale de la Secruit Sociale, Urban-Rural Differences in China (Geneva Association Internationale de la Stcuritd Sociale. 20001. p 50 Ideas for Further Work and Analysis Despite significantly different circumstances in East Asia and Europe, the European experience suggests priorities for East Asian countries in confronting the challenges of aging rural populations.20 First, effective social security schemes need to rest on formalized farming practices. Such schemes require strong and clear property rights, simplified administrative procedures, recognition of existing informal rights, and nation- wide databases on rural populations and their income. Second, a sound, effective, and equitable contribution base is essential, with the foregoing characteristics crucial for establishing such a base. Third, a priority for reform is to help low-income earners gain access to social insurance. This effort needs to include support for individuals' contributions; progressive schemes based on income, especially for countries with large income disparities; and provision of in-kind benefits for elderly people to complement cash transfers, such as affordable housing for retired elderly farmers. Women's needs require particular attention. Fourth, empowering farmers' organizations and thus improving farmers' par- ticipation in decisionmaking is essential for timely reform. Such democratic processes can help ensure that no elderly farmers are left behind 304 1 Facing Aging and Providing Social Security Table 17.7 Main Features of Rural Social Protection in East Asia Formal social insurance coverage Safety nets (social assistance type) Retirement Survival Invalidity China Emergency social relief, Voluntary farmers' funds cover - - emerging local rural around 83 million rural residents assistance Indonesia Emergency assistance None (disaster relief, social unrest) Korea Public assistance Yes Farmers, fishers, and the Coverage Yes programs low flat rate rural self-employed pay 3% of includes benefits for needy elderly their earnings (scheduled to rise survival, and other in-kind to 9% by 2005) The government risk assistance pays for administrative costs and subsidizes premiums for low- income farmers and fishers Malaysia Minimal provisions from The Provident Fund is voluntary the Ministry of National for the self-employed Extension Unity and Community of the scheme to different self- Development employed groups, including farmers, is under study Such a scheme would be set up through partnerships with farmers' organizations Philippines In-kind assistance (Social Yes, the general statutory Yes Yes Welfare Department) scheme covers all self-employed Thailand Subsistence allowances None to indigent elderly in rural areas, low coverage, restricted access Vietnam In 1993 land distributed to Voluntary scheme is managed by elderly for subsistence the Vietnam Farmers' Union farming Annual contributions with a value of 80 kg of rice are paid over 20 years, which yields a pension entitlement of about 6 kg of rice per month Sources- Association Internationale de la Secunite Sociale, Urban-Rural Differences in China (Geneva Association Internationale de la Secunte Sociale, 2000), Didier Jacobs, Social Welfare Systems in East Asia A Comparative Analysis Including Private Welfare (London Centre for Analysis of Social Exclusion, London School of Economics, 1998), Intemational Labor Organization (2000), Robert Holzmann, "The World Bank Approach to Pension Reform," International Social Secunty Review 53,no 1(2000) 11-34 Social Insurance for Aging Rural Households A Comparative Perspective 1 305 Finally, national organizations will need to pool funds from regional organi- zations and ensure regional compensation, with contributory flexibility at the local level. Countries need to develop a clear rationale for those measures The challenge is to preserve a high degree of regional autonomy while ensuring national coordination so the whole effort reaps economies of scale, shares know- how, and attains geographical equity These goals will require a sound, integrated information system and clear rules assigning responsibility to different levels of government Notes I John Bryden, Structural Changes in Rural Europe (Aberdeen, Scotland Arkleton Centre for Rural Development Research, University of Aberdeen, 2000) 2 Neil Ravenscroft and research team members, Private Sector Agricultural Tenancy Arrangements in Europe Themes and Dimensions-A Critcal Review of Current Literature (Madison, Wis Land Tenure Center, University of Wisconsin-Madison, 1999) 3 A market for "usufruct rights," crcated with transmission of production quotas under the European Common Agriculture Policy, has become a complementary, informal source of capital 4 '[he same right has been extended to spouses under the 1999 Agriculture Guidance l aw (Lol dorientation agricole) If the family member does not take over the farm, the law allows purchase of rights to a pension 5 Peter Baldwin, The Poltics of Social Solislarity Class Bases of the European IVellare State. 187i-1975 (Cambridge, U K Cambridge University Press, 1992) 6 European Parliament, Docutnetnts de Seance, 1983-1984, Rapport sur la nicessitd d'instaurer un nou- veau regime de retraite agricole (Commission des affaires sociales et de l'emploi du parlement europ6en) 7 E Talos and C Badelt, The Welfare State Between New Stimuli and New Pressures Austrian Social Policy and the EU," J7ournal of European Social Pohcy 9, no 4 (1999) 351-61 8 With expansion of off-farm work, agricultural activities escape traditional definition. To what extent should nonagricultural activities be included under agricultural social security coverageP The boundaries of coverage for nonfarm work in rural settings will have a substantial effect on the number of insured people In France the informal definition of agricultural activity is more comprehensive than the legal and fiscal one Yet general legal rulings tend to be more restrictive than specific social security schemes would wish In 1999 France simplified the rules for entitle- ment, establishing that nonfarm workers may draw a pension from the scheme under which they havc worked longer Their full income, including nonagricultural activities, would count toward the agriculture entitlement However, the farmer can be exempted from social security contribu- tion in agriculture if it is not his or her main activity The same holds true for Greece, though for residents in very small towns social security coverage is compulsory Given widespread part-time work, Germany needs to ensure that part-time agriculture work is covered if nonagriculture work is not sufficiently rewarding Under current legislation, if total income exceeds 640 deutsche marks, the farmer may be exempted from social security coverage In Austria insurance for part- time workers became compulsory in January 2000 306 I Facing Aging and Providing Social Security 9 Agnieszka Chlon, Marek G6ra, and Michal Rutkowski, "Shaping Pension Reform in Poland Security through Diversity," Social Protection Discussion Paper No 9923 (Washington, D C World Bank, 1999) 10 Jill Armstrong, with Louis-Charles Viossat, Towards an East Asian Social Protection Strategy (Washington, D C World Bank, East Asia Human Development Unit, 1999) 11 Association Internationale de la S&urite Sociale, Urban-Rural Differences in China (Geneva Association Internanonale de la S&urite Sociale, 2000), p 50 12 Frithjof Kuhnen, Synthesis of Current State of and Trends in Land Tenure and Land Policy in Asia (Deutsche Gesellschaft fur Technische Zusammenarbeit, 1996) 13 Sutthichai Jitapunkul and Srichitra Bunna, Aging in Thailand (Bangkok, Thailand Thai Society of Gerontology and Geriatric Medicine, 1998) 14 Libor Stloukal, "Rural Population Aging in Poorer Countries. Possible Implications for Rural Development," SD Dimensions (May 2001) 15 Kuhnen, op cit 16 SutthichaiJitapunkul, op cit 17 Association Internationale de la S&urite Sociale, op cit 18 John Ingleson, "Mutual Benefit Societies in Indonesia," International Social Security Review 46, no. 3 (1993) 69-77 19. Robert Holzmann, "The AVorld Bank Approach to Pension Reform," International Social Securnty Review 53, no 1 (2000) 11-34 20 For a discussion of limitations on cross-country transfers of experience and what can be learned from policy experiences in different national settings, see Ehtisham Ahmad, Jean Dreze, John Hills, and Amartya Sen, Social Security in Developing Countries (Oxford, England Clarendon Press, 1991) CHAPTER 18 Social Policy and the Informal Sector by Jacques Charmes ABSTRACT The informal sector, often underestimated in assessing national economies, plays a particularly important role during periods of crisis. This chapter reviews the historical evolution of the links between the informal sector and social security systems. The chapter also highlights significant differences in the East Asian sector's response to economic crisis compared with that of other regions. Although informal activities account for a significant share of gross domestic product (GDP) in many countries, knowledge of the informal sector's size, dynamics, and links with formal sectors of the economy tends to be partial and patchy.' The informal sector is still widely-and wrongly-interpreted as syn- onymous with illegal or underground activities Yet in reality more and more industrial wage employment has become informal and thus precarious, tempo- rary, and often home based, especially for female workers. Global economic turbulence has added a new dimension to this growth, as countries expect informal sectors to absorb labor and support people during tough times. The East Asian 1997-98 crisis put a new spotlight on this phenom- enon. This chapter builds on empirical evidence about the dynamics of self- employment and the informal sector in creating jobs and generating income. Views of the Informal Sector The views of public authorities regarding the informal sector-or the "tradi- tional sector," as it was formerly known-have changed dramatically In the early 307 308 I Facing Aging and Providing Social Security 1970s, when belief in Western models of modernization and industrialization was widespread, many policymakers and analysts considered the informal sector a relic, destined to quickly disappear as countries embraced market and monetary economies. Although most economic activities, including agriculture, were tra- ditionally informal, rapid progress in building states and judicial systems was supposed to encourage people to shift rapidly to more formal means of earning livings and running businesses. The international oil shocks and debt crises of the 1980s, however, prompted significant and lasting growth in the informal sector, which absorbed labor sur- pluses in countries with no unemployment benefits. The informal sector allowed formal workers to survive when real wages dropped dramatically. Economists and policymakers thus changed their views of the sector, seeing it now as pro- viding temporary positions for rural migrants whose ambition was to seek jobs in the formal sector,2 they regarded underemployment in the informal sector more favorably than open unemployment After these crises, national and international institutions began to explore two types of policies toward the informal sector. Policies of the first type aimed to encourage income-generating activities, especially survival activities by the poor, to alleviate poverty and vulnerability Policies of the second type were designed to support microenterprises, which employ permanent wage workers, to sustain job creation and investment. Those policy initiatives posed significant questions for economic develop- ment theory and policy Why should the state and international institutions support activities that are growing naturally? Would such intervention disturb subtle mechanisms? Would government do better to address unemployment among retrenched workers of the formal sector, as well as young unemployed graduates, more directly? Policymakers concluded that direct support for the informal sector was necessary because its growth related closely to the market distortions targeted by structural adjustment policies.3 East Asian countries rarely participated in these debates as they focused on industrialization policies. However, the role of subcontracting in the region has influenced thinking about the informal sector Economists have long seen small subcontracting firms-often in industrial districts or clusters-as fueling indus- trial growth by providing export-oriented firms with labor-intensive raw materi- als.4 For example, industrial districts in northern Italy provided an effective response to prolonged economic crisis because they continued to create jobs and improve productivity even when Italian unemployment rates reached high levels.5 Informal family- and ethnic-based networks in the Philippines and Indonesia resemble such industrial clusters and help keep small firms alive. In fact, many Social Policy and the Informal Sector 1 309 analysts see the role of subcontracting in fostering links between the informal and the formal sectors as the key to East Asia's successful industrialization Extending Social Protection to the Informal Sector: The European Experience In Europe, "out-work" intensified during the 19th-century Industrial Revolu- tion, especially in rural areas. European countries expanded their social security systems, and especially the right to employment benefits, only gradually, as infor- mal workers became part of the formal wage sector.6 At the beginning of the 20th century, a legal framework emerged to control the sweatshop system that employed women working at home. More than 40 years intervened between the first laws forbidding any wage differential between homeworkers and industrial workers and contemporary laws stipulating that homeworkers have the same legal and regulatory rights as wage employees. When countries adopt and enforce such new laws, the number of out-workers drops dramatically, as recently observed in Ireland. Thus out-work declined con- tinuously until recently, when the number of employees equipped with comput- ers began to grow. The extension of social security systems to self-employed workers remains a challenge. European governments, long convinced of the major role of micro- and small enterprises (termed the "MSE sector")7 in creating employment and building social capital and networks, have designed specific policies to support it, such as temporarily exempting the sector from fiscal and social regulations Microcredit policies have also helped the unemployed create their own jobs, underscoring the importance of the informal sector in alleviating poverty Analysts have focused on two major roles of the informal sector in social pro- tection: the role of informal firms in absorbing labor and the role of households in meeting basic needs. The former focuses on the informal sector's counter- cyclical role. The share of the informal nonagricultural labor force rises rapidly in most regions and countries during economic downturns, although it does not drop as quickly during economic upturns because of its labor-absorbing role. When crises occur and inflation rises sharply, households of the self-employed in the informal sector, as well as in agriculture, who do not consume imports and can adjust prices to reflect cost and demand, are supposed to support family and community members who have lost jobs or face straitened circumstances. When that occurs, overall living standards drop and the number of households living under the poverty line rises, because many households have lost their main 310 X Facing Aging and Providing Social Security source of income even while they must support more members. In some devel- oping countries the share of private household transfers of income has reached 10 percent to 25 percent of GDP8-a share comparable to that of public trans- fers in European countries.9 Thus households seem to come to the rescue when the state is unable to pro- vide social benefits. But the reverse is also true: if the state establishes a universal system of social protection, households can restrict their private transfers to the optimum required to maintain social relationships and social capital. This raises the question of whether and when the state should provide social protection for all its citizens. The 1991 International Labor Conference debated the dilemma posed by the informal sector: "whether to promote the informal sec- tor as a provider of employment and incomes; or to seek to extend regulation and social protection to it and thereby possibly reduce its capacity to provide jobs and incomes for an ever expanding labor force."'l The World Trade Organization is debating the same issues, largely in the context of unfair competition. In many developing countries, social protection systems cover self-employed as well as wage employees. However, many self-employed workers remain unreg- istered, and the informal sector remains outside the social security system. Because formal social protection is often limited to health services and family allowances, and because benefits are uncertain, with retirement pensions appear- ing especially distant, many wage employees in the informal sector are reluctant to accept deductions from their earnings for these ephemeral benefits. Strict enforcement of social security laws can therefore mean retrenchment for employ- ees in the informal sector who work for wages. The Informal Sector in East Asia In contrast to Europe, industrialization in East Asia and other developing regions appears to rely heavily on the informal sector because it allows low pay, child labor, and flexibility. The contribution of the informal sector to GDP in East Asian countries is actually much smaller than employment figures would suggest (see Table 18.1). This probably reflects the fact that most labor force statistics do not document the importance of home work well, and production statistics do an even poorer job. 1 " In 1999 Thailand found that homeworkers (80 percent women, and 90 per- cent in manufacturing) constituted only 2 percent of nonagricultural employ- ment. In contrast, a 1993-95 Philippine survey showed that homeworkers (80 percent women) accounted for 14 percent of manufacturing output because the survey included the secondary activities of women farmers. Social Policy and the Informal Sector 1 311 Table 18.1 Percentage of Informal Sector in Sectoral GDP in East Asia Indonesia /%) Korea N%) Philippines (%°/) Sector 1997 1998 1990 1995 1996 1997 1998 Agriculture 78 0 79 2 95 4 95 2 64 3 65 3 63 6 Industries 15 2 16 7 12 6 8 9 33 4 32 3 35 2 Trade 74 9 77 2 48 8 42 4 50 6 51 0 51 4 Services 28 6 30 5 26 8 23 4 22 6 221 22 3 Total nonagriculture 30 2 31 6 22 8 16 6 31 7 30 8 31 9 Total GOP 254 255 210 155 252 251 265 Sources- Calculations based on Statistics Indonesia IBPS/SDN), Gross Domestic Product Indonesia 1997 and 1998, Philippine National Statistical Coordination Board INSCB), Economic and Social Statistics Office, NationalAccounts of the Phi hppines, 1990-98 Those workers offer enterprises flexibility during periods of growth as well as decline, but the drawbacks for households can be severe, since social protection systems do not cover them. They have obtained benefits only through mutual soctal insurance schemes and grassroots organizations, such as those that pro- moted recent International Labor Organization (ILO) resolutions on homework- ers like the Declaration on the Fundamental Principles and Rights at Work and Its follow-up, adopted in 1998. Despite their reliance on the informal sector, East Astan countries have gen- erally ignored its counter- and pro-cyclical contributions. However, when unem- ployment rates rose during the 1997-98 crisis, these countries expected the infor- mal sector to function as a safety net by preventing many retrenched workers from falling into poverty (see Table 18.2). East Asta thus turned out to be a special case; the "survivalist" component of the informal sector was already saturated and could not absorb retrenched workers Table 18.2 Trends in Labor Force Indicators Indonesia (%) Philippines (%) Thailand (o) Indicators 1997 1998 1999 1997 1998 1999 1997 1998 1999 Participation rate 66 3 66 9 67 2 65 5 66 0 n a 717 70 6 n a Unemployment rate 4 7 5 5 6 3 7 9 9 6 n a 0 9 3 4 n a Underemployment rate 12 0 9 3 12 6 20 8 23 7 n a n a n a n a Share of agriculture 40 7 45 0 43 2 40 4 39 9 41 0 50 3 51 3 48 5 Share of informal sector 45 8 42 8 n a 40 3 n a n a 25 2 24 6 26 5 n a = not available Source Author's calculations based on countries' national surveys and records 312 1 Facing Aging and Providing Social Security during the crisis, especially when purchasing power in urban areas fell suddenly and dramatically. Subcontracted by formal firms through intermediaries and then laid off, many workers returned to their previous status in the countryside as farm- ers, family workers, and housewives. These workers are mostly female and under- take a high share of home work as a secondary activity, with agriculture as the main activity (in this case the return to farming did not imply any return migration) In Indonesia, for example, although the mformal sector did not respond significantly, the share of agriculture in total employment grew from 40.7 percent to 45.0 per- cent between 1997 and 1998. The Philippines is an exception a late 1997 drought stemming from El Nifio and excessive rainfall from La Nimia in late 1998 and 1999 prevented such an absorption and aggravated the effects of the crisis. Such behavior is probably specific to the region, because the informal sector was so tightly linked to the formal sector Because the agricultural sector was both export oriented and subsistent-safeguarding purchasing power in rural areas without consuming significant imported goods-it played the role that the informal sector usually filled. These results show that it is important to distin- guish the part of the informal sector linked to the formal sector from independ- ent microenterprises and the "survivalist" countercyclical component of the informal sector. The significant share of subcontracting work in the informal sector, particu- larly the female share, makes it an important engine for growth. The World Bank recently assessed the role of homeworkers in the Philippines and Indonesia dur- ing the crisis, and the ILO is doing the same in Thailand Overall, the significance of the informal sector raises the following important questions for global development policies * How and how far can the informal sector be an engine for growth in devel- oping economies? * Can it play an efficient role in providing social insurance for the poor and victims of economic turbulence? * Can social insurance policies directed at workers in the informal sector help promote gender equality in labor relations? * Can public intervention support private household social insurance, and can governments design social security systems to cover families in the informal sector? The World Bank and ILO studies show promise in launching a debate and further analysis to offer answers to these questions. Social Policy and the Informal Sector 1 313 Notes I Clearer international definitions andl harmonized methods for measuring the informal sector would be very helpful 2 Examples include the models of Lewis, Harris, and Todaro, which are based on unemployncnt as a factor of equilibrium on the urban labor market Such models recognized the role played by thc infornmal sector 3 Jacques Charmes, "Le developpement du secteur informel Entre le laisser-faire et la promotion dans la formalisation, quelle voie pour les politiques publiques et les agences d'aid.1" (conimuni- cation at a scirnar of OECD's Centre de D6veloppement et do Comit6 d'Aide au Developpcment on "Ic sectcur inforniel dans les pa)s en developpement," Paris, December 13-14, 1990)) 4 As stylized by Hubert Schmitz, "Truist and Inter-firm Relations in Developing and Transition Economies," _ournial of Developmnent St(ie(les 34 (April 1998) 32-61 5 M AMaruani, B Reynaud, and C Romani, eds, La fle%ibihie' en Italhe Ddbae tsinr l'einploi (Paris Syros Alternatives, 1989), p 317 6 R Salais, N Baverez, and 13 Reynaud, L'tiiveriton du eh6niage HIsloire ei transJor;naltons d'rie cale'orie en France des a,in&es 1890 auix annese 1980 (Paris Presse Universitaires de France, Economie en hberte, 1986), p 267 7 European economic history and policy commonly use the term "MSE sector" rather than "infor- mal sector" to refer to economic activitics that are not officially or statistically registered 8 Jacques Charmes, "Informal Scctor and Social Capital Resilience of African Economies during Crisis and Adjustment" (Programme CODESRIA of the lMacArthur Foundation on Real Economies in Africa, Zanzibar, July 5-10, 1999), p 26 9 Institut National de la Statistique et des Etudes, Donn&es Sociales (Paris Institut National de la Statistique et des Etudes, 1999), p 350 10 International Labor Organization, The Dilemma of the Informal Sector, Repor of the Director General, International Labor Conference, 78th session (Geneva, 1999), p 65 11 Jacques Charnies, "The Contribution of Informal Sector to GDP in Developing Countries Assessment, Estimates, iMethods, Orientations for the Future" (OECD-EUROSTAl:State Statistical Committee of the Russian Federation, Nonobserved Economy \Vorkshop, Sochi, Russia, Octobcr 16-20, 2000), p 14 BOX 42 Conference on Social Policymaking in Europe and East Asia, Roma/Caserta, February 21-23, 2002 This final project seminar, jointly sponsored with the Italian Higher Institute of Public Administration and the Ministry of Finance, focused on how social policy is formulated in terms of issues and institutional arrangements. The seminar covered six topics, with pension reform, par- ticularly Italy's recent experience, a focal point. The seminar first explored social security systems broadly, intro- duced by Meesook Kim's paper on the Republic of Korea's social secu- rity system (see Chapter 15). Discussion focused on the role of active labor policies in Korea compared with some European countries, espe- cially Ireland and the United Kingdom, which share a common experi- ence in their move from passive to active labor policies. The notion of an optimal level of social spending sparked particular interest; partici- pants noted wide variation across Europe in social spending (from 35 percent of gross domestic product in Sweden to 15 percent in Iceland) and in the composition of social policies. Political and institutional fac- tors play a central role in how countries balance social spending and eco- nomic growth, whereas welfare policies reflect different value systems built over long periods. Second the seminar addressed Italy's experience with pension reform, with a presentation by Giuseppe Pennisi and Mauro Mare (see Chapter 16). Discussion ranged widely over common and different pension sys- tem challenges in Europe and East Asia, which include acute financial imbalance in many European public pension schemes, as well as the rel- ative advantages and disadvantages of pay-as-you-go and funded systems. The third topic focused on labor market policies, with presentations by Jonas Gahr Store on the Norwegian experience (see Chapter 7) and Dzung Nguyen Huu (Vietnam) on Vietnam's labor policy reforms. Jonas Store's fascinating case study underscored how a comprehensive labor policy could fight unemployment while avoiding serious macroeconomic continued on next page 316 l Facing Aging and Providing Social Security BOX 4.2, continued imbalances. However, Norway's Solidarity Alternative now faces serious challenges, including the need to maintain macroeconomic stability, bal- ance rights and obligations of social partners, and contend with strong labor unions Growing absenteeism and use of sick leave are symptoms of a system in crisis Mr. Huu noted that active labor policies in Vietnam are still at a nas- cent stage, even though unemployment is a central social concern. Recent measures to curb unemployment include the National Program on Employment (1992), designed to encourage demand for labor through public investment in social infrastructure. Despite such efforts, many vul- nerable groups-especially demobilized military personnel, redundant workers from state enterprises, and female workers-remain outside the social protection system Links between labor policy and pension sys- tems, balancing work and family for working women, and family care for children and the elderly alike are topics of almost universal concern The solidarity formula was of particular interest to the group. The fourth topic focused on health insurance policies, focusing on Thailand (presentation by Juree Vichit-Vadakan) and Italy. Participants noted striking parallels between challenges facing the two health systems, particularly in measurement, equity, and the role of public versus private services Participants discussed the complexities of forging consensus on priorities for health spending and the implications of decentralization (a particularly sensitive issue in both Thailand and Italy), seeing efficiency in spending as a vital barometer. The fifth topic was the role of information in social policymaking, with Giuseppe de Filippi (Mediaset, Italy) in the chair, and a presenta- tion by Colin Moynihan, who emphasized how new information tech- nologies shape policymaking. Pitfalls include a premium on public rela- tions as opposed to policy discussion, and cynicism about statistics and a disdain for projections-which foster short-term approaches and yawn- ing knowledge gaps. Possible solutions include publication of efficiency and effectiveness indicators for social policy. Discussion emphasized the crucial role of education in raising public awareness of social policy issues and thus overcoming short-termism. Conference on Social Policymaking in Europe and East Asia, Roma/Caserta, February 21-23, 2002 1 317 BOX 4.2, continued The final session looked to new ventures in social policy, based on presentations by Jean Wolas (intergovernmental delegation for social innovation, France) and Luc Tholoniat (see Box 6.2). Jean Wolas spoke of "social economy" as a new venture for social policy in France. Globalization has changed relationships between the global and local economy, raising challenges to democratic governance, benefit sharing, and protection of local stakeholders. In contending with those chal- lenges, the French government has rediscovered the value of the third sector-civil society. Luc Tholoniat outlined the open method of coor- dination that has emerged within the European Commission in response to criticism that the European Union lacked coherent social policy. More use of benchmarking is an important new policy direction. PART V Labor and the Welfare State Introduction Part V addresses a critical set of social policy issues: the need for countries to boost employment and maintain an active and viable work force The world of work is changing dramatically in virtually every part of the world, including both Europe and East Asia, owing to globalization and the advent of the knowledge- driven economy. Countries face complex choices in deciding how best to prepare for such challenges In Chapter 19 Stein Kuhnle, Aksel Hatland, and Sven Hort show how the supposedly hidebound Scandinavian countries have transformed universal social security benefits into provisions that promote work-thereby sparking strong economic growth. The authors argue that the vibrant records of Denmark, Finland, and Sweden show that strong welfare states are compatible with eco- nomic growth These states are in fact the most family friendly in Europe, main- taining policies that encourage the participation of both women and men in the labor force Scandinavian nations nonetheless appear to have coped successfully with these new challenges, showing that developed democratic welfare states are quite good at preserving the legitimacy of social systems while pursuing eco- nomic transformation and new vitality. The chapter concludes with a thoughtful discussion of the many noneconomic dimensions of social policy, including its reflection of social and political norms and values and the central goals of com- batting poverty and maintaining social peace. In Box 5.1 Gazier and Herrera trace parallels in the experiences of govern- ments during the Great Depression and during the recent East Asian crisis in tackling large-scale unemployment. Although efforts to enhance employment can prove helpful in the short run, this experience highlights how extensive state involvement in creating jobs can prove counterproductive Today countries can 321 322 1 Labor and the Welfare State use public expenditures as a lever while ensuring that social partners take more responsibility for enhancing the mobility of workers and unemployed people. In Chapter 20 Colin Moynihan and Jacqueline Butler tackle the daunting challenge of youth unemployment, which is twice as high as adult unemploy- ment in both industrial and developing countries. Young people's transition from education to working life is critical, because early unemployment points squarely to joblessness later in life. However, European experience shows that broad labor market programs rarely improve job prospects for the young. The Nordic coun- tries again offer the best model, integrating education, labor market, and welfare policies with local community agencies that provide the individualized services youth need. Most important, strong overall growth in employment is essential to ensuring jobs for youth Box 5 2 focuses on the United Kingdom's efforts to address youth unemployment. The United Kingdom has systematically engi- neered new institutions to improve young people's transition to employment, including the creation of a "New Deal" for young people. In Chapter 21 Jacques Rojot explores the many different forms of worker par- ticipation in play in Europe-all based on the notion that employees perform better when their work is useful, interesting, and meaningful, as well as when they hold well-designed jobs for which they are well qualified. The author argues that, as East Asian countries move to high-quality, high-tech manufacturing, enhancing worker initiative and involvement in firms' decisionmaking will become crucial, and he examines how East Asian countries might encourage such participation. In Chapter 22 Ruud Dorenbos investigates the role of public employment services, which aim to match employees with jobs, promote equal opportunity, and enable labor markets to function more efficiently. The author provides infor- mation on how such services operate in both Europe and Asia. Globalization, technological development, and changing labor markets and state roles are bring- ing profound changes to these services as former bureaucratic organizations move toward customer-oriented services and collaborate with private employ- ment agencies. In Chapter 23 Bernard Gazier and Remy Herrera recount the effects of the 1997-98 crisis on employment and incomes in the Republic of Korea and the country's response in the form of active labor market policies, including extended vocational training and public works jobs Korea also relaxed regula- tions limiting firms' ability to lay off workers and allowed growth among agen- cies that provide temporary workers. As recent European experience shows, today's "transitional labor markets"-the global shift from long-term to short- term jobs and discontinuous careers-can accommodate older workers and Introduction 1 323 enhance gender equality, but they also imply the need for a strong government role in retraining the work force. In Chapter 24 Christine Erhel outlines European experience in monitoring and evaluating the effects of active labor market policies to determine whether the benefits of such programs are worth their cost. Such analysis is especially important because active labor market policies can hurt some groups. Evaluations can range from surveys and interviews with participants to scientifi- cally based studies and macroeconomic analysis European experience sheds light on how best to conduct such complex studies, as well as on which active labor market policies prove most effective (or ineffective). In Chapter 25 Eduardo Gonzalez-Biedma traces the evolution of European thinking on how to tackle structural unemployment, which the region has faced since the oil shocks of the 1970s and 1980s. Governments throughout Europe are moving toward more flexible labor markets, which can generate more jobs than rigid labor regulations in the long run. New approaches to tackling unemploy- ment also include enhancing job seekers' mobility and fostering open relation- ships between management and unions, as well as careful government efforts to spark new industries. Overall, the author argues, crises can provide the crucible in which economies shed their inefficient industries and embrace more vibrant sources of growth and employment. Boxes 5.3 and 5.4 provide highlights of two seminars that concentrated around the challenges for labor markets arising from the East Asia crisis The first was a major conference in Seoul in December 2000, which weighed flexibility versus security and situated labor policy instruments within broader social policies. The Bangkok seminar in October 2000 was a joint venture with the International Labor Organization, which stretched boundaries of cross-discipline reflection Among interesting facets was a discussion of media roles in policy formulation and a probing exploration of social standards. CHAPTER 19 A Work-Friendly Welfare State: Lessons from Europe by Stein Kuhnle, Aksel Hatland, and Sven Hort ABSTRACT. This chapter focuses on issues for the welfare state, partic- ularly in periods of recession and economic recovery It reviews the expe- rience of European welfare states in their multifunctional, challenged, and reforming dimensions It argues that a strong and well-developed welfare state can be a strong asset, as illustrated by recent developments in the Nordic countries. It discusses the interaction between social secu- rity and labor markets, and it suggests some lessons from Scandinavian and European experience The Scandinavian countries have made a remarkable recent comeback as inter- esting cases of social, economic, and political development after many economists and commentators had essentially categorized them as hopelessly overloaded welfare states in which economic growth and dynamism were hampered Scandinavian growth performance, trend growth of gross domestic product (GDP) per employed person, government employment, and social expenditures all suggest that comprehensive social security and welfare indeed need not rep- resent barriers to economic growth, and they may even prove advantageous when countries are hit by economic turbulence A combination of comprehensive wel- fare and social security programs with economic growth is possible and may offer a path that reduces poverty and opens possibilities for development with fewer inequalities of income distribution. In this chapter we contend that European welfare states and social security programs have historically fulfilled, and still fulfill, several functions. Examples of interactions between social security and the labor market are outlined, with 325 326 1 Labor and the Welfare State a special focus on what has happened in Scandinavian countries in recent decades. The Swedish old-age pension reform of the late 1990s is presented as a particularly interesting example of a "work-friendly" social security and wel- fare arrangement. We argue that national unemployment insurance schemes are intrinsically work friendly. Finnish and Swedish developments during the 1990s show how an elaborate social security system functioned as an important shock absorber when the countries fell victim to sudden economic downturns, helping generate rapid recuperation and new economic growth at low social cost In this era of eco- nomic globalization, there may be a still greater objective need for a consolidated national social security system than before. Changing characteristics of labor markets in the global economy may encourage changes in pension systems, like those in Sweden, that make them more transparent and more fair, in the sense that pensions reflect contributions during all working years. The Scandinavian welfare state emphasis on public social services for children, the elderly, and the sick suggests some lessons, perhaps controversial in other cultural contexts, about the evolution of both work-friendly and family-friendly social security and welfare state programs. A comprehensive welfare state is not a necessary prerequisite for a good soci- ety, but neither are the two mcompatible. In our view, the welfare state historically and at present serves multiple functions, and its achievements and success can be measured only against considerations of specific values and political and social goals (for example, fairness, justice, cohesion, stability, material and physical secu- rity, well-being, and economic growth). Economic growth is possible with or with- out an advanced welfare state and a strong government role for welfare responsi- bility. Developments over the last 20 years bear out the empirical possibility of alternative visions of a good society that are based on different value assumptions and political choices. It is perhaps time to recognize this fact more generally-not least when considering lessons to be drawn for other regions of the world whose political histories, traditions, and cultures differ from those that spearheaded industrial modernization and development of social security institutions. Welfare States, Recession, and Economic Recovery Many economists and commentators used to portray the Scandinavian or Nordic states as victims of a "welfare state sclerosis."' These countries, however, have made a remarkable image comeback and are now at the center stage of theory and practice of social and economic development.2 When the 1997-98 crisis hit Asia, A Work-Friendly Welfare State Lessons from Europe I 327 the Nordic countries (with the important exception of Norway) were still, like Japan, struggling to overcome the effects of the recession of the early 1990s. Their situation has changed dramatically since then. Finland, the hardest hit by the Nordic recession,3 ranked number I among nations in the World Economic Forum's recent survey of the global business environment,4 replacing the United States, which was ranked top the previous 2 years. Three other northern European countries made the top 20: Denmark, number 6 (up from number 7 in 1999 and number 8 in 1998); Sweden, number 7 (previously numbers 4 and 7); and Norway, number 20 (previously numbers 18 and 14) Among East and Southeast Asian countries, Singapore, number 9 (previously numbers 12 and 10), Japan, number 14 (previously numbers 14 and 18), and Hong Kong, China, number 16 (previously numbers 21 and 12) were among the top 20. Those results are quite consistent with similar rankings and evaluations for the same period 5 Both investment banker Merrill Lynch and the U.S. Internet Council recently published reports that likewise single out the countries of northern Europe as business-friendly environments, in this case Sweden (number 1), as well as Fin- land (number 5) and Denmark (number 8). In important respects the Nordic countries have defied expert advice from organizations such as the Organization for Economic Cooperation and Development (OECD)6 and the International Monetary Fund (IMF). Witness a recent statement' Sweden should cut taxes and overhaul its unemployment benefit system if it wants its current economic upswing to continue.... The IMF said there was "considerable room" for tax cuts and urged the country not to use its current surplus to increase spending It also supported lower long-term unemployment benefits to encourage recipients to seek jobs."7 A similar IMF statement about Norway included sweeping advice that the public sector must be made more efficient and reduced in size. The experience on the ground, however, suggests that the economy and the public sector (mostly the welfare state part of it) have grown in parallel,8 and today even the IMF indi- rectly admits that a different tune has worked in Sweden, although it still claims the familiar old tune as an eternal favorite. Tables 19.1 and 19.2 show indicators of economic growth since 1970 in the Nordic countries and the United States, as well as averages for the European Union (EU) and OECD countries. Tables 19.3 and 19.4 offer data on relative scope of government employment and social expenditure as a percentage of GDP for four different groups (and types) of European welfare states. 328 1 Labor and the Welfare State Table 19.1 Growth Performance, Selected Countries, 1970-99 (average annual rates of change in GDP) 1970-1980 I%) 1980-1990(%) 1990-1998 (% 1999 (%) Denmark 2 2 19 2 3 16 Finland 3 4 31 1 5 3 5 Norway' 4 2 15 31 0 8 Sweden 19 21 11 3 8 European Union 3 0 2 4 1 7 2 3 OECD' 34 30 23 27 United States 3 2 3 2 3 0 4 2 a Data for Norway are for the mainland only the oil and gas sector is not included b Data for the OECD average excludes the Czech Republic, Hungary, and Poland Source J Elmeskov and S Scarpetta, "New Sources of Economic Growth7" (paper presented at the 28th Economic Confer- ence of osterreichische Nationalbank, Vienna, June 15-16, 2000) Table 19.2 Trend Growth of GOP per Person Employed, 1980-90 (average annual rates of change) 1980-90 (%) 1990-98 (%) Denmark 1 5 24 Finland 24 29 Norway 21 25 Sweden 16 24 European Union 23 1 8 OECD 28 24 United States 11 17 Source J Elmeskov and S Scarpetta, "New Sources of Economic Growth?" (paper presented at the 28th Economic Conference of Osterreichische Nationalbank, Vienna, June 15-16, 2000) Table 19.3 Government Employment as a Percentage of Total Employment 1974-95 in Different Types of European Welfare States (unweighted averages) 1974 (%) 1985 (%) 1995 I%) Continental Europea 147 187 188 Scandinaviab 20 0 26 9 294 Southern Europe' 105 142 155 United Kingdom 196 21 5 144 a Continental Europe includes Austria, Belgium, France, Germany, and the Netherlands b Scandinavia is Denmark, Finland, Norway, and Sweden c Southem Europe includes Greece, Italy, Portugal, and Spain Source S Kuhnle and M Alestalo, "Introduction Growth, Adjustments, and Survival of European Welfare States," in Survival of the European Welfare State, ad S Kuhnle (London and New York Routledge, 2000) A Work-Friendly Welfare State Lessons from Europe 1 329 Table 19.4 Social Expenditure as a Percentage of GDP in Different Types of European Welfare States, 1980-95 (unweighted averages) 1980 (%) 1990 (%) 1995(%) Continental Europea 28 1 29 6 301 Scandinaviab 25 6 28 1 321 Southern Europec 150 180 222 United Kingdom 21 5 24 3 27 7 a Continental Europe includes Austria, Belgium, France, Germany, and the Netherlands b Scandinavia is Denmark, Finland, Norway, and Sweden c Southern Europe includes Greece, Italy, Portugal, and Spain Source S Kuhnle and M Alestalo, "Introduction Growth, Adjustments, and Survival of European Welfare States," in Survival of the European Welfare State, ed S Kuhnle (London and New York Routledge, 2000) These data are not directly comparable They nonetheless illustrate that eco- nomic growth, even if sometimes modest, has on the whole been quite steady;9 that the Nordic countries had higher labor productivity, defined as GDP per per- son employed, an the 1990s than the average of EU countriesto and the United States; that labor productivity rose over the last decade compared to the previ- ous one; that European countries vary significantly in terms of scope of govern- ment employment and size of public sector; that growth tn government employ- ment has been quite compatible with expanding welfare states and economic growth in Scandinavia; that there is thus no clear-cut relationship between scope of the welfare state and economic performance, and that on average the share of GDP spent on social purposes increased all over WVestern Europe until the mid- 1990s. These data clearly do not prove how and to what extent the type and size of the welfare state is conducive to economic growth, but the data indirectly do indicate that, at the macro level, welfare state growth and economic growth have occurred simultaneously durang the last 20 years. Sweden, like other Scandinavian countries, illustrates that a combination of high tax revenues, labor market security, declining unemployment, generous uni- versal welfare services, and economic growth is possible The IMF, interestingly, credits the renewed Swedish economic upswing in part to Sweden's creativity and entrepreneurship, and to the persistent Swedish government emphasis on core welfare state policies such as a broad tax basis, high tax revenues, security in the labor market, the world's most generous paid parental-leave schemes, increased child allowances, and extra resources for universal preschools and public care Finland is an even more elegant examplc of such a happy marriage between a generous universal welfare state and strong economic growth." Finland's 330 I Labor and the Welfare State experience suggests that if low poverty levels and relatively high levels of income equality are the objective, then it pays to have a well-developed, universal wel- fare state in place when a sudden economic shock or crisis hits, as it did Finland, partly as a result of the breakup of the Soviet Union, in the early 1990s. Finland's advanced welfare state not only served those goals but also was conducive to rapid economic recovery thereafter, with steady, solid economic growth since 1994 (see Table 19 5) Though social security and welfare reforms (such as cuts in benefit levels) were undertaken in thel990s, they were handled in a politically consen- sual way that avoided both new poverty and increased income inequality.i2 Finland was the only OECD member country in which the recession of the 1980s or 1990s was deeper than the Great Depression of the 1930s,13 but with the fundamental difference that an advanced welfare state was in place when the later crisis hit. The dramatic social and political repercussions of the 1930s were avoided (repercussions which were moderate in Scandinavia compared with the rest of Europe, thanks to other factors such as crucial social-political pacts between agrarian and industrial interests as well as labor and employers). The Finnish example demonstrates the positive potential of a strong welfare state in a time of crisis, with advantages both during the crisis period and thereafter. Recent and historical European and Scandinavian developments underscore the ways in which the character and direction of social policy and the quality of links between the state and major economic groups in society at critical junctures Table 19.5 Economic Recession and Recovery in Finland Real growth of Real growth of Social expenditures Unemployment Year GDP (%) social expenditures (%) (as a % of GDP) rate /%) 1989 57 29 238 31 1990 00 69 252 32 1991 -59 88 299 66 1992 -32 71 336 117 1993 -06 20 346 163 1994 37 25 339 166 1995 39 07 320 154 1996 41 26 315 146 1997 56 -08 294 127 1998 56 -05 274 11 4 Source Hannu Uusitalo, 'Social Policy in Deep Economic Recession and After The Case of Finland,' Paper presented at ISSA Conference on Social Security, Helsinki, September 25-27, 2000 chttp //www issa int/pdf/helsinki2000/topic4/2uusitalo pdf> accessed October 28, 2002) A Work-Friendly Welfare State Lessons from Europe 1 331 of economic, social, or political crisis have importance for the way out of the crisis The Finnish and Swedish experience of the last decade also illustrates how demo- cratic systems and institutions can adapt to new or unexpected challenges. What institutions are in place matters when a sudden, unexpected crisis in society occurs. Contemporary European welfare states, some argue, face challenges of an entirely new character: "rapid transition to post-industrialism, increasing glob- alization, sweeping changes in demography and social relations, trends towards supranational integration and a new, post-cold war politics."'4 These states nonetheless appear to have coped quite successfully with the significant new challenges in the 1990s.iS This supports the hypothesis that developed demo- cratic welfare states are quite good at making adjustments in public policies in such a way that the legitimacy of the system can be preserved at the same time that new vitality and economic transformation are achieved. European countries have worked to create work-friendly welfare states (that is, welfare states that promote high labor force participation conducive to pro- ductivity). All of the Nordic countries (which are among Europe's most compre- hensive welfare states because they provide income transfers and services on a more universal basis) showed increasing labor productivity in the 1990s com- pared with the previous decade, at levels everywhere higher than for the United States and for the EU average (see Table 19 2). Among European welfare states, the Nordic countries were also the most family friendly, in that they have devel- oped policies conducive to labor force participation of both women and men in families with children or other care responsibilities (this also relates to work friendliness) (see Table 19.6). The Nordic countries have long had the most Table 19.6 Female Labor Force aS a Percentage of Total Female Population Aged 15-64 in Different Types of European Welfare States, 1960-95 (unweighted averages) 1960 I%) 1974 I%) 1995 (%l Continental Europea 421 44 9 59 2 Scandinaviab 48 9 60 9 72 5 Southern Europe' 31 6 37 6 49 0 United Kingdom 46 1 54 3 66 0 a Contiental Europe includes Austria, Belgium, France, Germany, and the Netherlands b Scandinavia is Denmark, Finland, Norway, and Sweden c Southem Europe includes Greece, Italy, Portugal, and Spain Source S Kuhnle and M Alestalo, "Introduction Growth, Adjustments, and Suovival of European Welfare States." is Survival of the European Welfare State, ed S Kuhnle (London and New York Routledge, 2000) 332 1 Labor and the Welfare State extensive provision of local government welfare and care services for children and the elderly among American or European welfare states.16 Work friendliness is also related to issues of a "business-friendly environ- ment " We argue that social security and business are not incompatible phenom- ena like fire and water: whether they are in harmony or conflict depends on the concrete construction of the system of social security and its interplay with the society in which it works, as well as the cultural and political characteristics of that society. European Welfare States: Multifunctional, Challenged, and Reforming The welfare state is a European political invention17 developed, expanded, adjusted, refined, and modified over a period of more than 100 years since Bismarck launched a comprehensive compulsory social insurance program in Germany in the 1880s. That action spurred Europe-wide legislative initiatives on Arbeiterfrage (the social question). European nation-statcs have developed welfare state institutions and programs of varying characteristics, but what- ever the brand of the welfare state, in its post-World War II shapes it must be seen as a significant institution conducive to the consolidation of democratic development. A basic principle for economic transfers forms part of this history and is still crucial to social security; it was first elaborated in the British 1834 Poor Law report. The "principle of less eligibility" means that income of those people who receive benefits should always be lower than the lowest-paid members of the labor force; the underlying assumption-that it will never pay for individual cit- izens or households to prefer social security benefits to work or gainful employ- ment-still applies. This principle is defended partly with economic arguments (incentives) and partly with arguments of justice or fairness There are two important exceptions The first and most important is the basic pension (and other basic benefits), because its main goal is to prevent and relieve poverty (though if income from paid work is lower or similar, these benefits may create a disincentive to work). The second exception is sickness benefits Quite often in Europe, especially in Scandinavia, employees in collective agreements, and some- times in legislated schemes, receive full pay, or close to full pay or income com- pensation, during periods of sickness How different compensation levels of health-related benefits affect labor participation and duration of absence is a major theme in political debates in many European countries. A Work-Friendly Welfare State Lessons from Europe I 333 Interaction between Social Security and the Labor Market In Europe compulsory social security systems have coexisted with modern labor markets since the end of the 19th century. The Bismarckian national legislation in Germany set the stage for the development of social insurance and simultaneously forced the collective actors on the labor market-employer associations as well as trade unions-to become social partners involved in the daily business of provid- ing pecuniary supplements to the working population. Institutions of social insur- ance became tripartite organizations that were jointly supervised and sometimes also administered by the central or local state and the social partners on the labor market As an unintended consequence, labor market participation in due course became partly regulated by agreements and regulations addressing not only wages-the wage-labor contract between the organized actors on the labor mar- ket-but also wage or earnings-related cash and in-kind benefits, so called benefit tide-overs Over the years such benefits have become part of a mixed incentive system of sticks and carrots to get people into gainful employment. Scandinavian experience over the last century illustrates the interactions between high and increasing labor market participation and expansion of social security systems. From the 1960s onward "high and sustainable" full employ- ment (expanded from male to include also female breadwinners) became the public policy norm of Keynesian demand management. Unemployment was no longer viewed as a private matter, but rather as a central public concern. Expansion of social security, meanwhile, involved upgrading, in terms of pecu- niary generosity and population coverage, all or most branches of social insur- ance At the macro level, unemployment insurance is expansionary also, despite some theories that suggest unemployment insurance offers incentives for indi- viduals to withdraw from the labor market (implicitly an opposite effect). Thus, the labor market has developed into a key aspect of the monetized economy, whereas social security systems have become core institutions of the modern wel- fare state Macroeconomic steering became part of societal governance. Much more is at stake In Scandinavia, systemic integration has gone hand in hand with social integration. At the micro level the implication and risk for the individual of being out of work is marginalization and social exclusion At the macro level, the government responsibility to create conditions under which enterprises can flourish has become linked to the general duty of every able- bodied person to work or achieve an education, schooling, or practice in order to enter into gainful employment. In sum, values other than those implicit in the rationality of the pure economic man have long been in force 334 I Labor and the Welfare State Notwithstanding the existence of an elaborate system of social security pro- viding transfer payments of different types, from child benefits to sick pay to old- age pensions, the dominant norm in Western societies, including Scandinavia, has been that young and middle-age adults-in recent decades to an increasing degree also including women-should be active in the labor market to gain an income. Persons with severe disabilities (the "deserving poor") are the exception to this norm. Acceptance of this norm has been sustained and reinforced by the fact that not only money income, but also such social characteristics as power, prestige, and status in the community are affected by whether an individual works and by the type of work he or she performs. Following social control mechanisms in fairly coherent communities such as Scandinavia, where the distinction between state and society to a great extent has been blurred, "disobedience" can lead to a loss of reputation or even exclusion from the dominant social group, even if disobedience is to the pure pecuniary advantage of the individual. This is an important aspect of the moral logic of the modern welfare state. In Scandinavia, labor force participation rates in the "modern sectors" of the economy (thus excluding agriculture and unpaid household work) have increased throughout the 20th century, with the exception of parts of the 1930s and parts of the 1990s (decades of economic downswings or crisis in northern Europe). In the post-World War II period, with the exception of Denmark, men's participation rates increased until the late 1970s, whereas women were still entering the labor market until the early 1990s, and in Norway throughout the 1990s. In Scandinavia, "housewives" have more or less disappeared as a social category. The impact of the development of the education system and the increasmg emphasis on higher com- petence and skills has meant that the definition of working-age adults has been adjusted upward because most of those in the age group 16-24 are now students The age and gender composition of the labor force in Scandinavia has been drasti- cally transformed during the last century. Low mobility, both within and across nations, has been a long-standing and serious problem for Europe's economic development. Workers often find it too risky to leave their job or local community to look for a job elsewhere When workers carry the economic risks connected with loss of jobs, they are likely to fight for strong employment protection, making labor markets less flexible. Nationwide schemes, which cover employees and workers independently of spe- cific employment conditions or characteristics of place and type of firms or employers, may thus be advantageous to labor mobility, structural economic change, and economic growth, because national social security systems can be constructed and administered in ways that encourage higher mobility or at least reduce the risk or cost for the individual worker who wants to look for a better A Work-Friendly Welfare State Lessons from Europe 1 335 job. Unlike under many employers' pension plans, employees do not lose pension rights under national (public) schemes if they move from one company to another There are also advantages to systems that coordinate across unemploy- ment benefits and employment services (such systems date back as early as 1909-11 in England). In sum, an adequate social security system can offer a health-bringing medicine for labor market "sclerosis." The development of a free labor market in the EU and the European Economic Area, with the concomitant coordination of national social security systems so that social rights are to a great extent transportable across countries, represents an additional incentive to labor mobility and labor productivity in Europe and can form part of a response to economic globalization and increased international competition. The fact that intra-European labor mobility has been very limited can partly be explained by Europe-wide problems of unemploy- ment, but perhaps more by the variations in languages and cultural habits. Many European countries have developed alternative strategies of job creation that differ from the liberal market approach that emphasizes deregulation and minimal welfare. Such alternatives have typically involved combined packages of social security reforms-that is, stricter conditions for receiving benefits, better training programs, job-search training, and so forth. The Netherlands is an interesting case, where the "Dutch disease" was apparently (or allegedly) cured and transformed into the "Dutch miracle" through a host of policy reforms dur- ing the 1990s.18 This example and others indicate that European welfare states have the ability to reform their societies toward what is generally considered "better" through piecemeal, democratic political processes and channels, involv- ing civil society or organized actors in the labor market. The changing role of women in the labor force has particular importance; wit- ness the Scandinavian countries. Because they face significant employment prob- lems associated with taxation, social security contributions, and unemployment benefits,19 a distinctive feature of Nordic welfare states since World War II has been the goal of full employment. Until the mid-1960s, this meant primarily full male employment, but women's opportunities for paid employment gradually increased, especially in the public ("welfare") sector. In Finland that entailed pre- dominantly full-time jobs, whereas part-time work has been more common in the other Scandinavian countries. Female employment in Scandinavia has reached the highest level among West European OECD countries (see Table 19.6), though female labor force participation rates have increased in all parts of Western Europe since the 1960s and 1970s (also an element of the Dutch miracle). The increasingly generous maternity and paternity benefits and parental- leave schemes during the 1980s and 1990s have been conducive to high female 336 1 Labor and the Welfare State labor force participation in Scandinavia, as well as gradual development of childcare services (though this is not peculiar to Scandinavia). One factor explaining the development of these generous schemes, making Scandinavian countries at once more work friendly, women friendly, and, in a sense, family friendly, is very likely the rapid political mobilization of women in political par- ties; trade unions; and decision-making arenas like the parliament, government, and public bureaucracy. Welfare State Reforms: Some Scandinavian Examples Since the late 1950s and 1960s, Scandinavian countries have moved to transform universal and flat rate or means-tested social security benefits into more work- or income-related mechanisms. This has happened in pension systems (Denmark being a slightly deviant case with generous universal basic pensions and supple- ments linked to employment period), sickness benefits, and unemployment bene- fits. Social security systems have come to be based more on the loss-of-income principle, or "status-maintenance" principle of the Bismarckian system, thus introducing a stronger work incentive yet at the same time sustaining universal pensions for all citizens-irrespective of labor market participation-in order to prevent poverty.20 Early retirement schemes, made more generous in recent years, could, on the contrary, be looked upon as promoting work disincentives, thus transmitting the notion that exit from work (among those who have worked for a long time) is a welfare gain. The Nordic economic and welfare systems experienced a decisive reorienta- tion in the late 1980s and early 1990s.21 Scandinavian welfare states have in sig- nificant respects been work oriented at least since World War II (some might argue since the Reformation in the 1500s), but increasing levels of unemploy- ment, which have remained relatively high for longer periods in Denmark (since the mid-1970s) and Finland and Sweden (since the early 1990s), gave rise to debates about labor market rigidities, and taxation systems and generosity of social security benefits came under scrutiny as possible work (or employment) disincentive factors Taxation does not appear to explain unemployment prob- lems, because unemployment was low (except in Denmark) in the 1980s at a time when marginal tax rates were high 22 Similarly, it is not clear that the tax reduc- tion reforms in the late 1980s and early 1990s had any major effect on work effort, measured by labor force participation and hours of work.23 Lower taxes may have induced employers to hire new workers and employees, but we have no data to A Work-Friendly Welfare State Lessons from Europe I 337 Table 19.7 Unemployment Rates (Commonly Used Definitions), Selected Years 1988-98 1988 (%) 1990 /%) 1992/%) 1994 /%) 1996 /%) 1998 /%) Denmark 86 96 113 122 88 74 Finland 45 35 131 184 163 137 Norway 32 52 59 59 49 42 Sweden 17 16 53 80 80 75 European Union 9 4 7 9 9 6 116 113 10 8 a Figures for 1998 are projections Compare with the lower figures for Finland 1996-98 in Table 19 5 Source P Kosonen, 'Activation, Incentives, and Workfare in Four Nordic Countries." in MIRE, Comparing Social Welfare Systems in Nordic Europe and France (Nantes, France Maison des sciences de l'Homme Ange-Gu6pin, 1998) corroborate this. The connection between economic work incentives and employment Is far from clear.24 On the other hand, social security benefits, which largely depend on previous earnings, create incentives to be employed (but to be efficient, require that the characteristics of the benefit systems are known to the unemployed, not at all an obvious assumptionl) If there is a lesson to be learned, it must be that direct links do not necessarily exist between taxation levels, unem- ployment and employment, and generosity of social security benefits. Tables 19.7 and 19.8 give an overview of development of unemployment and employment rates tn the Nordic countries. The early 1990s saw a declining trend in employment levels and an increase an unemployment levels, whereas an the latter half of the 1990s, the opposite trend was apparent (at the same time as increasing numbers of young working- age people were engaged in higher education, perhaps affecting unemployment statistics). Although employment rates on average declined in the EU in the 1990s, the rates increased an Ireland and the Netherlands.25 Labor utilization is Table 19.8 Employment Rates (Employed Persons/Persons of Work Age, %), Selected Years 1983-96 1983 (%) 1990 %) 1994 /%) 1996 /%) Denmark 71 7 771 72 8 74 7 Finland 73 2 741 601 62 2 Norway 73 9 73 9 72 7 76 8 Sweden 78 5 80 9 70 2 72 7 Source P Kosonen, "Activation, Incentives, and Workfare is Four Nordic Countries," in MIRE, Companng Social Welfare Systems in Nordic Europe and France (Nantes, France Maison des sciences de l'Homme Ange-Gu1pin, 1998) 338 I Labor and the Welfare State Table 19.9 Average Annual Working Hours for Total Economy, 1980,1990,1998 1980 1990 1998 Denmark - 1,492 1,527 Finland 1,755 1,677 1,674 Norway 1,512 1,432 1,401 Sweden 1,439 1,480 1,551 European Union 1,755 1,659 1,620 United States 1,831 1,819 1,833 Source J Elmeskov and S Scarpetta. 'New Sources of Economic Growth?' (paper presented at 28th Economics Conference of fisterreichische Nationalbank, Vienna, June 15-16, 20001 lower in the EU than in the United States. The Nordic countries have higher employment rates than the United States, but this is offset by lower hours worked and results in lower labor utilization (see Table 19.9). Various evaluations of the effects of active labor market policies (training, education, protected work, and so forth) in the Nordic countries indicate that the effect seems to be positive, seen from the perspective of both the unemployed and the employers. Incentives to be employed have become stronger rather than weaker (given income-related social security benefits and pensions) in Scandinavia in recent years, but policies have also changed to greater use of the stick: more elements of control and discipline are imposed on the unemployed. Pension Reforms and Welfare State Issues The rolc of pension systems in overall social security systems and in labor incen- tives has particular importance. European old-age pension systems are often extremely complicated, developed through long historical processes. Benefit schemes often include elements that were perhaps justifiable when they were introduced but have survived long beyond their initial rationale. Those benefits create their own interest groups that fight for their privileges. A good example is the retirement age for police and army officers. Their low pension age was justi- fied at a time when they had to engage in physical fights and had to be fit until they retired. Today such officers usually have paperwork jobs after age 50, and there is no longer any good reason why they should retire earlier than other groups of white-collar workers. A Work-Friendly Welfare State Lessons from Europe 1 339 European pension systems have faced two important and related problems: first, a complexity that makes the systems difficult to understand at the most basic level, and second, internal contradictions and complex features, arising from their historical development, which may go in opposite directions from both original intent and contemporary policy. Public conceptions, and wide- spread misconceptions, about how the system works cause adaptations that are counterproductive to work-friendly welfare states (for example, a system may offer higher earnings if one retires early from a full-time job than if one gradu- ally reduces hours worked and retires later) A major recent study of social secu- rity, published by the International Labor Organization, comments on aspects of this problem: There is often a lack of transparency in the administration of social secu- rity provisions. This weakness applies to the failure to explain adequately the broad concepts and objectives of the scheme, and to the failure to advise the insured persons how their pension records are progressing or what is happening to the contributions they have paid. Many schemes do not issue regular pension or contribution statements Often the legislation is complex and difficult to understand for insured persons, employers and even administrators. The problem applies across the whole spectrum of social security provisions, but is particularly acute in the case of pensions since, first, the contribution rates are highest and, second, benefits may not be payable for 30 years-on retirement. People inevitably wonder what is happening to their contributions.26 The modern Scandinavian pension system reforms are an interesting case, notably the Swedish pension reform of the late 1990s A central aim was to make the pension system more work friendly by making it more transparent, at the same time retaining the element of universal income guarantee for all elderly persons The lack of clear and understandable links between contributions and bene- fits is a main reason Sweden introduced a new pension system during the lat- ter half of the 1990s. It was launched in 1999 and will be phased in gradually Since 1960 Sweden has had an obligatory two-tier pension system, with a uni- versal flat rate basic pension for all inhabitants and an earnings-related supple- mentary pension. This pension scheme met rising criticism during the 1980s and 1990s, and a political consensus gradually evolved on the system's main weaknesses: expenditures were not linked to economic growth (or change), and 340 1 Labor and the Welfare State supplementary pensions presupposed a growth of approximately 2 percent annually, so if the growth rate was slower (as it had been in the 1975-95 period), contributions had to be increased Longer life expectancy and growing num- bers of pensioners added to the bill. But an important problem was a funda- mental unfairness in the lack of a link between contributions and benefits The national supplementary pension favors those with erratic earning trends or those who work for a short time, while being unfair to those whose income develops consistently and who work for a long time This is due to the benefit formula determining pension rights on the basis of the 15 years of highest income and a minimum qualifying period of 30 years. Two people with the same lifelong earnings may receive very different pensions, although they have both paid in the same amount in contribu- tions It is not fair that those who work all their lives and have a steady earnings profile should receive less from the system than those who enjoy a short career.27 The new scheme is based on the principle of lifelong earnings. Each year employers and employees pay contributions amounting to 18.5 percent of the income of all economic active residents in Sweden. Pensions will be equivalent to the value of all paid-in contributions at the time of retirement. For many years the retirement age in Sweden has been 65, but in the new system there is no fixed retirement age Pensions can be claimed from the age of 61, and there is no upper age limit. Annual benefits are calculated on the basis of the person's remaining life expectancy at the time of retirement The later a pcrson retires, the higher the annual pension. It is also possible to take out reduced pensions and combine these with part-time work. Individuals who have earned little or no pension are guaranteed a minimum pension, financed by general taxes, paid after age 65. Persons who have earned pensions through contributions that are below the guarantee level will always receive a higher pension than the minimum. Income is supplemented (financed by taxes) for parents who care for small children, for national service, and for recipients of certain disability and early retirement pen- sions These additions are financed by taxes. This new Swedish pension system gives a simple and clear message to all resi- dents in Sweden: it always pays to work. All income from work will increase pen- sion rights The system is work friendly and transparent. Everyone can see the principle of justice that it is based on: proportionality between work and pension rights A Work-Friendly Welfare State Lessons from Europe 1 341 Concluding Comments Social security and welfare state arrangements serve many purposes. Social secu- rity policies are about political and social preferences, shaped by culture, tradi- tions, economic and technological developments, social structure, and social rela- tions. We should be particularly cautious about claiming any "correct path" for social policies because values and preferences affect perceptions about fairness, social justice, social cohesion, stability, and equalization of life chances. There is, nonetheless, a core aspect of social security systems, which is how to organize support in ways that also offer incentives to work, an important goal of most gov- ernments. There are many rather unexplored linkages. Unemployment insurance systems can facilitate the process of restructuring in industry and business. Economic security may reduce workers' and employees' resistance to change. Compulsory government schemes, which cover employees and workers inde- pendently of specific employment conditions or characteristics of place and time of firms or employers, may favor labor mobility, structural economic change, and economic growth. National schemes for unemployment insurance are in very important respects work friendly. Recent European experience indicates that there are no obvious links between size of the welfare state, taxation levels, employment rates, labor productivity, and economic growth. The examples of Sweden and, especially, Finland during the 1990s show that comprehensive, democratic welfare states are fully capable of making policy adjustments to stimulate new economic growth when hit by serious economic recession. Social security represents a shock absorber, while democratic institutions are responsive to economic and social challenges We can expect, in this world, more rapid changes in labor markets, more mobility, more flexible work, and more career shifts during the period of labor market participa- tion Those new patterns will demand changes in pension systems to make them more transparent and less dependent on "best years" and "last years" of earn- ings, and thus more fair, in the sense that the pension reflects contributions during all working years The Swedish pension reform of 1999 offers other European as well as East Asian countries a Iesson bccausc it combines thc man- ifest work-friendly aspect with a universal guarantee of income security in old age Status maintenance is combined with poverty prevention. Another aspect of Scandinavian welfare states, their emphasis on public social services for children, the old, and the sick, is less universally accepted Such government schemes may be considered both work friendly and family friendly If families are relieved of some of their burden (itself a contested 342 i Labor and the Welfare State concept) as caregivers (for their young, old, and sick family members), labor market activity and labor mobility can increase, and thus also can economic productivity and growth. Government social policies can provide the basis for flexible solutions for families, for workers and employees, and for firms. Social policies can make it possible, if desired, for both husband and wife to combine family obligations with full-time (or part-time) gainful employment. If glob- alization also has a cultural dimension, such as spreading the idea, demand, and expectations of greater gender equality in all spheres of society, it may well be that welfare states with social policies that are relatively more gender sensitive will be the most successful in terms of both economic development and social peace and harmony. Family friendliness of social policies can also affect fertility rates (lack of such policies in parts of Europe such as Italy and Spain may have contributed to sharp drops in fertility rates, whereas strong policies may have spurred relatively higher rates in the Nordic countries). Whether the welfare state-with its programs for retirement and disability pensions, sickness insurance, occupational injury insurance, maternity insurance and paid parental leave, child and family benefits, unemployment insurance and labor activity programs, health and personal social services, kindergartens, and so forth-is a blessing or not for economic growth and efficiency is a topic for much research and debate 28 At least seen at the macro level and over longer time spans, the welfare state and economic growth have gone hand in hand. Economic growth is possible with a number of welfare state constructions of varying scope and generosity. Economic growth and efficiency are, however, clearly not the only goals of European national welfare policies. Politics and welfare state construction are also about equalization of life chances, social justice, social security, social cohesion, and stability, all in various ways related not only to economic growth and dynam- ics of economic development, investment climate, and so forth, but also to politi- cal preferences, ideologies, interests, and values. Thus, what kinds of welfare state policies are possible is also at all times a question of what is considered desirable by governments and voters, and what is considered desirable-what the state can and ought to do29-is a question of political and cultural context (norms, expecta- tions, value structures) as much as a question of level of economic development and theories and knowledge of prerequisites for economic growth and efficiency. An important lesson from the European experience is a simple one: the wel- fare state does and may serve many functions. Debates on what are proper les- sons to be learned and what are proper welfare policy solutions in other, non- European, contexts can thus be framed in many ways. Social protection and welfare are often discussed in terms of poverty relief and meeting minimum A Work-Friendly Welfare State Lessons from Europe 1 343 needs for income and services. Poverty reduction was historically a central rea- son that governments put social legislation on the agenda. But the establishment of European welfare states is about much more, especially in their Nordic and continental European variants. Originally, considerations of social harmony and regime support were important. Over time, many programs were developed to insure against events and risks that cause income loss, to allow reallocation of income over the life-cycle and redistribution across social groups, and to provide a sense of security for all citizens.30 Europe's welfare states appear to have been a societal stabilizer, preventing serious social rebellion, strong revolutionary move- ments (except for right-wing extremism in the pre-World War II period when the welfare state and parliamentary democracy were still weak and underdevel- oped institutions in many parts of Europe), and extensive poverty. The combina- tion of structures of democratic governance, regulated capitalist market economies, and relatively comprehensive welfare institutions has rather success- fully accommodated changing social needs. Social and political challenges have continuously filled government agendas, but empirically dramatic crisis theories since the mid-1970s have fared rather poorly in the European context.31 Notes I F W. Scharpf, "The Viability of Advanced WVelfare States in the International Economy Vulnerabilities and Options," European Review 8, no 3 (2000) 399-426, Maurizio Ferrera and Martin Rhodes, eds, Recasttig European Welfare States (London Frank Cass Publishers, 2000), and Stein Kuhnle, ed., Survival of the European WelJare State (London and New York Routledge, 2000) See, for example, Assar Lindbeck et al , Turning Swedeni Around (Cambridge, Mass, and London MIT Press, 1994), for a penetrating anablsis not only of how the Swedish welfare state is (or was) perceived to be in a severe cnsis, but also of Sweden's economic and political institutions 2 Kuhnle, op cit 3 Financial Times, August 9, 2000 4 The World Economic Forum's list, compiled in cooperation with the Center for International Development at Harvard University, available at 5 For instance, the 2001 survey of the Intcrnational Institute for Management Development in Lausanne, Switzerland 6 OECD, The Welfare State in Criiv (Paris OECID, 1981) 7 Financial Times, September 2000, pp 9-10 8 Stein Kuhnle and Matti Alestalo, "Introduction Growth, Adjustments, and Survival of European Welfare States," in Survival of the European Welfare State, ed S Kuhnle (London and New York Routledge, 2000) 9 Labor productivity also increased in Germany and Portugal in the 1990s 10 M Heikkila and H Uusitalo, eds, The Cost of the Cuts Studies in Cutbacks in Social Security and Their Effects in the Fin/and of the 1990s (Helsinki STAKES, 1997), H Uusitalo, "Social Policy in 344 I Labor and the Welfare State Deep Economic Recession and After The Case of Finland," paper presented at ISSA Conference on Social Security, Helsinki, September 25-27, 2000 11 P Eitrheim and S Kuhnlc, "Nordic Welfare States in 1990s Institutional Stability, Signs of Divergence," in Survival of the European Welfare State, ed S Kuhnle (London and New York Routledge, 2000) 12 Uusitalo, op cit 13 Ferrera and Rhodes, op cit 14 Ferrera and Rhodes, op cit , p 1 15 J Kohl, "Trends and Problems in Postwar Public Expenditure Developments in Western Europe and North America," in The Developmenit of Welfare States in Europe and America, eds A J. Hcidenheimer and P Flora, (New Brunswick, NJ, and London Transaction Books, 1981), Mikko Kautto, Matti Heikkila, Bjorn Hvmden, Staffen Marklund, and Niels Ploug, eds, Vord:c Social Policy Changtng Welfare States (London and New York Routledge, 1999) 16 Anthony B Atkinson, The Economic Consequences of Rolling Back the Welfare State (Cambridge, England Cambridge University Press, 1999), A Sandmo, "Social Security and Economic Growth," in European Instiute oJ Social Secunty Yearbook 1994 (Leuven, Belgium Acco, 1995) 17 Stein Kuhnle, "European lessons of the 1990s," in Survival of the European Welfare State, ed S Kuhnle (London and New York Routledge, 2000) 18 J Visser and A Hemerijck, A Dutch Miracle Job Growth, Wtelfare Reform, and Corporatism tin the Netherlands (Amsterdam Amsterdam University Press, 1997) 19 This is underlined in many OECD publications, in particular in the OECD jobs Study (Paris OECD, May 1994), as well as in EU reports See also P Kosonen, "Activation, Incentives, and Workfare in Four Nordic Countries," in MIRE, Comparing Social Welfare Systems in Nordic Europe and France (Nantes, France Maison des Sciences de l'Homme Ange-Gu6pin, 1998) 20 The main source of information is The Pension Reform-Final Report (June 1998) 21 According to P Kosonen, op cit 22 Comparing tax reforms and unemployment patterns, Kosonen (op cit ) comes to this conclu- sion 23 N A Klevmarken, I Andersson, P Brose, E Gronqvist, P Olovsson, and M Stotlenberg- Hansen, Labor Supply Responses to Swedish Tax Reforois, 1985-1992 (Stockholm National Institute of Economic Research, Economic Council, 1995), R Aaberge and J Dagsvik, "Labor Supply Responses and Welfare Effects of Tax Reforms," Scandinavian Journal of Economics 97, no 4(1995) 635-59 24 Kosonen, op cit 25 Jorgen Elmeskov and S Scarpetta, "New Sources of Economic Growth" (paper presented at the 28th Economic Conference of Osterrelchische Nationalbank, Vienna, June 15-16, 2000) 26 C Gilion, J Turner, C Bailey, and D Latulippe, cds, Social Security Pensions Development and Reform (Geneva International Labor Organization, 2000), pp 2445 27 The Peiision Reform-Final Report, op cit, p 9 28 Atkinson, op cit , Sandmo, op cit 29 Kees Van Kersbergen, "The Declining Resistance of Welfare States to Change" in Survival of the European Welfare State, ed S Kuhnlc (London and New York Routledgc, 2000) 30 Atkinson, op cit, 5-6 31 Van Kersbergen, op cit BOX 5.1 Social Protection and Experience from the Great Depression by Bernard Gazier and Remy Herrera The Great Depression was a laboratorv for social protection measures, and the successes and failures of the wide range of programs introduced in its wake have relevance today. Three countries took quite different approaches in responding to social needs during the crisis and saw dif- ferent results in both the short and long term. France's' situation was dire from 1929 to 1939, as it confronted 10 percent losses in gross domestic product during the first 4 years, followed by lingering stagnation. What is striking, however, is that the unemploy- ment rate reached only 6 percent. The explanation lies not in the French economy's strength nor in any "active" labor market policy but in two self-reinforcing factors. First, official figures counted only people who lost jobs, excluding newcomers to the labor market. Second, France was still largely a rural country with close family ties, so many people returned to the countryside to participate in harvesting and other rural activities when they lost jobs. They found shelter and relied on family solidarity, both more or less sufficient. Thus, although many people faced sharply curtailed incomes, alternatives worked reasonably well. The New Deal initiatives that the Roosevelt administration pursued in the United States are still impressive today. The Civil Works Adminis- tration (CWA) hired 4 million unemployed in January 1934, and the Fed- eral Emergency Relief Administration (FERA) reported 20 million bene- ficiaries during winter 1934. The Social Security Act, passed in 1935, created retirement, health protection, and family assistance programs. Public works programs were perhaps the most controversial and com- plex of those initiatives. In the face of criticism that it paid too-high wages and disrupted labor markets, CWA was abolished in 1934. It was quickly replaced by the WAorks Progress Administration (WPA), which paid wages that varied with local conditions. The WPA's coverage steadily became more restrictive: it funded 2.9 million jobs in 1936, 1.5 million in 1937, continued on next page 346 I Labor and the Welfare State BOX 5.1, continued and 3 million for a very short period during the 1938 slump, compared with a maximum of 11 million to 12 million unemployed in the 1930s. FERA programs also excluded many "real" unemployed, referring them to employment agencies After providing emergency help in 1934, the U.S. government essentially chose to rely on private employment to assist most unemployed, despite very low economic activity. Social distress in the 1930s was relieved only by World War II, which reemployed most of the unemployed, and the peace that followed, which provided jobs for both veterans and civilians. For its part, the United Kingdom developed a deliberately counter- cyclical policy in tandem with important social developments after 1935, and the country recorded strong growth from 1935 to 1940. The United Kingdom created social protections progressively, introducing unemployment benefits and pensions without spectacular breaks or reforms. Laws and other incentives encouraged construction of individ- ual and standardized dwellings, prompting a housing boom: newly built houses doubled in the 1930s compared with the 1920s. After 1937 paid holidays stimulated mass leisure Together those factors boosted popu- lar consumption. The policies that the Republic of Korea and other East Asian coun- tries pursued in the wake of the 1997-98 economic crisis present paral- lels with the French, U.K., and U.S experience. The East Asian approaches included traditional solidarity, mass emergency public works and attendant problems when they halted, and countercyclical macro- economic policies as the backdrop. Family solidarity worked quite well in East Asia during the crisis. Even in urbanized Korea, rural and traditional solidarity still operates. Strong saving behavior indicates a community that can act as a buffer and accumulate private income protection and assets without relying on state help. Some East Asian initiatives resemble those of the United Kingdom, with macroeconomic policies complementing public works programs, enhancing the economy's ability to absorb laid-off workers and helping reconcile mobility and solidarity. Some elements of the CWA-WPA process in the United States high- light the limits and potential conflicts of public policies designed to Social Protection and Experience from the Great Depression 1 347 BOX 5.1, continued address crisis conditions. State-managed mobility is always difficult in times of tight budget constraints. It begins with volunteers and can end in authoritarian decisions that either suppress programs or, more likely, reorient them in restrictive ways. This applies especially to relatively "small" or "lean" welfare states, which experience both implementation and legitimacy problems when they expand their social protection pro- grams rapidly. In response, governments often develop less attractive forms of "workfare" and direct-quickly and bluntly-those with poor reemployment prospects to private labor markets. This process can aggravate segmentation of the labor market. Can countries escape this vicious circle? Maintaining extensive state involvement in jobs and labor market mobility over an extended period is neither possible nor desirable. Transitional labor markets encourage more decentralized control of labor market policies and expenditures by municipalities and nonprofit organizations. The French experience dur- ing the 1930s resembles an informal transitional labor market, strongly unequal and heterogeneous. Today governments must actively respond to social needs while mak- ing their social protection policies fairer and more efficient. To meet that challenge, they must fulfill two closely linked conditions. First, they need to develop alternatives to the stark dichotomy between public and pri- vate management. Second, they must ensure that social partners assume some responsibility for enhancing the mobility of workers and unem- ployed. Overall, the state should not bear the entire social protection bur- den. Public expenditures can and should be used as a lever, together with active mobilization of social networks and activities in the third sector. Some networks and activities can be informal (through associations, cooperatives, and mutual benefit societies), supported by the traditional solidarity links that remain particularly strong in parts of East Asia. 1. Bernard Gazier, La cnse de 1929 (Paris: Presses Universitaires de France, 1989). CHAPTER 20 Youth Unemployment: Challenges and Ideas by Colin Moynihan with Jacqueline Butler ABSTRACT. This chapter sets youth unemployment in the context of globalization trends, which threatcn to worsen social exclusion among young people The chapter highlights experience with programs designed to foster youth employment in Europe and some of the risks and pitfalls demonstrated in recent experience. It highlights the importance of a sharp focus on the transition period from education to employment and the vicious circle that can trap young people who do not make the transi- tion smoothly The most successful programs have a sharply targeted and tailored character and are part of sound national strategies and effective public-private partnerships New industries, cross-border trade, and a diverse global labor market offer the possibility of providing decent work for people in all societies. But we also risk a more divisive scenario a recent report by the U.K. Industrial Society, Wlherever Next? Work in a Mobile World, contrasted a "supermobile elite" with an "immo- bile poor" and spelled out the potentially damaging consequences. Many contemporary economic changes, including technological innovation and liberalization of international trade and finance, have far-ranging repercus- sions for those who lack access to transport and face local exclusion from work opportunities, shops, banking, and other services. Many new jobs require skills in information and communication technologies and the ability to operate in new forms of work. Yet only 1 in 250 Africans has access to a telephone. The digital divide is more acute in developing countries but is by no means restricted to them More than 100 million people in industrial societies live below the poverty line, defined as half the median income, and 37 million are jobless.2 349 350 1 Labor and the Welfare State In Europe higher levels of growth offer great potential for reducing social exclu- sion, but the very globalizing trends and communications revolution that con- tribute to Europe's healthy balance sheet have adversely affected the employment prospects of those in the weakest position in the labor market. In seeking solutions to these problems, policymakers must focus on youth employment. Most young people are better educated, are wealthier, are physically healthier, and live longer than their elders, but changes in the labor market, fam- ily relations, and social structures present them with new sets of risks and chal- lenges. Opportunities for those without skills are fewer, traditional sources of employment are disappearing, and people are far less likely to hold a job for life. Young people today face the prospect of periodic successive job change and the need to acquire new skills throughout their working lives to remain employable. This chapter addresses European efforts to help ensure that all young people acquire such skills and can move successfully from learning to work. Social Exclusion, Social Capital, and Youth Unemployment Social exclusion has become a catchall phrase for some of society's most entrenched problems: sharp income inequality, unemployment, poor access to services, child poverty, homelessness, and drug and alcohol abuse. Solutions to social exclusion and youth unemployment are inextricably linked because social exclusion is both a cause and effect of youth unemployment. Social exclusion results from a paucity of social capital and also tends to erode it. Social capital is an asset embedded in relationships-among individuals, com- munities, networks, and societies. Social networks have always been an important resource for recruitmg people into a range of economic activities. Those networks work less well in areas with entrenched long-term unemployment. In Europe a lack of access to support networks that help people land an edu- cation, homes, jobs, services, and benefits is contributing to what the UK think- tank Demos calls "network poverty."3 Those who lack appropriate skills, are in difficult social or economic situations, have disabilities, face personal and insti- tutional prejudice, or lacking learning opportunities, may find themselves caught in a vicious cycle Poor and unemployed people are less likely to live in commu- nities where others might be able to put them in touch with a job, and role mod- els are limited or negative. This pattern has a particularly detrimental impact on young people because risk factors for social exclusion cross generations In the United Kingdom, 34 Youth Unemployment Challenges and Ideas 1 351 percent of sons and 37 percent of daughters of men who fall in the lowest quar- tile of earnings also end up in the lowest quartile. Only 13 percent of such sons and 12 percent of such daughters end up in the top quartile The daughter of a teenage mother is 1.5 times more likely to become one herself than is the daughter of an older mother And during the 1980s, young men age 23-33 were twice as likely to be unemployed for at least 1 year if their father had been un- employed at age 16.4 Young people are most vulnerable to social exclusion when leaving home, leaving care, and leaving education, times at which wider changes have weakened or removed some of the support systems on which people, particularly adoles- cents, have traditionally depended. For many young people, family ties have become fluid and more fragile. More young people have grown up in single- parent households, which are disproportionately likely to be low income, and they may be less well equipped to cope with other difficulties. Weaknesses in government policies can compound the resulting problems. Homelessness, teenage pregnancy, and unemployment cross boundaries between services and departments, which can lead to "orphan" issues, underenrollment in services, and duplication of cffort. Policy coherence, among government departments and between local and national levels, is key to addressing both social exclusion and youth unemployment Increasingly, the issue of social exclusion is assuming center stage. The European Union states that "the fight against social exclusion is of the utmost importance for the Union,"5 reflecting the view that social exclusion carries high costs in the form of reduced social cohesion, higher crime and fear of crime, a less-skilled work force, and a drain on public finances. If social exclusion is reduced, then society as a whole will benefit. Businesses are also realizing that corporate success depends on both social and economic outcomes, and that investing in young people is a business imperative This is reflected in corporate-backed ventures that aim to improve young people's lives and their workplace experiences and opportunities. One such initiative Is the International Youth Foundation's6 Global Alliance for Workers and Communities, whose members include Nike, Gap, the World Bank, the John D. and Catherine T MacArthur Foundation, Penn State University, and St John's University. Kevin Quigley, executive director of the Global Alliance, observed recently that because the average age of employees in factories in developing countries is 23, and 80 percent to 85 percent are female, "we see the workplace as a venue for human development and an opportunity to help people develop skills that will enable them to build better lives for themselves and their families."7 352 l Labor and the Welfare State The Dimensions of Youth Unemployment The best defense against social exclusion is a job, and the best way to get a job is to have a good education with the right training and experience. However, the International Labor Organization (ILO) reports that unemployment rates are twice as high among the 15-24 and 25+ age groups than among adults across both industrial and developing countries. ILO research also shows that youth unemployment rates were significantly higher in the 1990s than in the 1980s in a number of countries. The young do not choose to be unemployed: as ILO notes, "the army of the unemployed is a conscript rather than volunteer army,"8 and it considers the large and growing number of unemployed youth "one of the most daunting problems faced by developed and developing countries alike " Early unemployment can permanently impair people's future productive capacity; the longer an unemployment spell lasts, the more difficult it is to find work. Unemployment can prevent young people from making the passage from adolescence to adulthood that entails establishing a household and a family. Evidence also suggests that unemployed young people suffer more health prob- lems than those who are employed, including lower rates of general health, more anxiety and depression, high rates of smoking, and higher suicide rates. Young people are clearly at high risk of social exclusion if they fall into a downward spi- ral of worsening health and unemployment. Young people's transitions have always been difficult, but recent research underscores that the transition to adulthood is more complicated today. Distinctions between childhood and adulthood have blurred, with many young people remaining dependent on their parents into their twenties and beyond, even while they experience more autonomy earlier in their teenage years. At the extreme end of the spectrum, high levels of youth unemployment can lead to alienation from society and from the political process, which may in turn give rise to social unrest This is the soil in which the seeds of terrorism can find root. Poverty breeds resentment, resentment breeds hatred, and hatred breeds ter- rorism. It is no accident that terrorism has found adherents in the Palestinian ter- ritories, where standards of living have fallen by some 40 percent since the signing of the Oslo accords, and unemployment rates range from 20 percent to 30 percent In seeking solutions to youth unemployment, we must clearly identify its causes. Social exclusion is both a cause and effect of unemployment, but social exclusion alone does not explain the high and persistent patterns of youth unem- ployment. Indeed, a superficial examination would suggest that the economic sit- uation of youths should have improved, thanks to declining cohort size,1O longer education, and higher skills So why has the economic position of young people Youth Unemployment Challenges and Ideas 1 353 deteriorated? The ILO has ruled out a number of causes, including high youth wages (they have becn falling) and minimum wages (they are low). The evidence does not suggest that youths are being priced out of the market in any major way Indeed, the relative pay of youths has dropped throughout the countries of the Organization for Economic Cooperation and Development (OECD)." Changes in aggregate demand, the growing demand for skilled workers, and the rising participation of women who compete with the young for jobs appear to explain in large measure the rise in youth unemployment.12 Another explanatory factor in OECD countries is the role of higher home ownership, which impairs people's mobility-critical in economies that depend on flexibility and adaptabil- ity. If the biggest cause of youth unemployment is the level of aggregate demand, solutions to youth unemployment are inextricably linked to reducing overall unemployment, even if the youth share of the population continues to fall. Unfortunately, unemployment in general, and youth unemploymcnt in particu- lar, fluctuate in large irregular cycles, and we are a long way from understanding why aggregate employment has trended upward over the last couple of decades. Knowledge about how to enhance employment-intensive economic growth is also rather scant Despite conventional wisdom, high unemployment does not primarily seem to result from job protection, labor taxes, the power of trade unions, or wage inflexibility Recent research suggests a number of alternative explanations Above all, changes in commodity prices, especially oil prices, seem to predict cyclical movements in unemployment in Europe and the United States reasonably well 13 Those findings suggest that the most important factor in unemployment is the macroeconomic environment. Economic stability and growth drive employment; a strong economy generates jobs. But although a strong economy may create employment, it will not create employability. Thus our educational system, our processes for enabling youth to make the transition from education to employment, and our safety nets for catch- ing those who slip through the cracks assume central importance European Solutions to Youth Unemployment Skills and education are at the very heart of the problem. Firms no longer sim- ply recruit from the pool of available unemployed if those in that pool do not pos- sess the necessary skills The search for talent is global, and enlarging the pool of talent is critical. As requirements for knowledge, qualifications, and skills rise and populations age, young people cannot afford to enter the labor force unequipped for changing career patterns. 35 l Labor and the Welfare State The good news is that young people in OECD countries seem to be recogniz- ing the new realities of the labor market and are responding by postponing their entry and undertaking more education.'4 In virtually all OECD countries, enroll- ments in higher education rose from the 1980s to the 1990s. Research in the United Kingdom has shown that higher education brings earnings that are 20 percent higher on average, and a chance of unemployment that is 50 percent lower.'5 A few decades ago in the United Kingdom, only a minority stayed in education until age 18 or 21. This new trend toward higher education is partly a response to poor labor market conditions and partly a response to the demand for higher skills. An extraordinary meeting of the European Council in November 1997 pro- duced the European Employment Strategy, whose full employment goal16 entails disseminating best practices to reflect the interdependence among national labor markets. The strategy recommended that member countries adopt early interven- tion programs to combat youth unemployment typified by the Nordic approach, given that youth unemployment is lowest in Denmark, Norway, and Sweden 17 The United Kingdom has responded with the New Deal for Young People, a welfare-to-work program that links income assistance to a requirement that recipients participate in work programs or education and training (see Box 5.1). France has established Emplois Jeunes (Young Jobs), backed by an employment counseling service, and Germany has created the Immediate Action Program for the young unemployed. In general, the burden of joblessness among the young falls on the least edu- cated and the least skilled. For example, some 40 percent of young people partici- pating in the United Kingdom's New Deal cannot read "the basic instructions on a medicine bottle."'8 Education systems across Europe are facing new challenges that affect all participants. For generations, teachers have been the gatekeepers of knowledge, but technology is challenging that role More information is now avail- able to the connected student than any individual teacher could know, and educa- tion needs to draw on a new and wider range of resources. But the United Kingdom is now demanding higher standards and more ambitious targets from an increasingly outdated infrastructure-the schools and universities retain basic structures that are centuries old. The Lisbon European summit in March 2000 set a goal to halve the number of 18- to 24-year-olds with only lower-secondary-level education by 2010. Toward that end the summit called for developing schools and training centers into mul- tipurpose local learning centers, linked to the Internet and accessible to all, and for establishing learning partnerships among schools, training centers, firms, and research facilities. Youth Unemployment Challenges and Ideas 1 355 Advances in secondary schooling in France, the Netherlands, and Sweden appear to have more impact than labor market programs and work experience and training for unemployed youth 19 High-quality early childhood intervention also appears to exert lasting effects. The Perry Pre-School Program in the United States is a well-known example disadvantaged, low-lQchildren given intcnsive support from age 4 to 5 report higher adult earnings and lower levels of criminal behavior. In fact, research on the program suggests that for every US$1 invested in a preschool child, the taxpayer saves US$7 over time The benefits of more education cross generational boundaries Research from the United Kingdom's Social Exclusion Unit shows that by age 22 months, children whose parents stayed in school beyond the minimum leaving age develop significantly faster than those whose parents left school at the minimum age.20 The challenge for Europe is to develop coherence among education, employ- ment, and income-support policies, because the transition from initial education to working life for young people is so critical and prone to pitfalls and traps The United Kingdom's Social Exclusion Unit has shown that a lack of participation in education, training, or employment for 6 months or more between ages 16 and 18 is the single most powerful predictor of later unemployment. This pattern is also associated with high levels of depression and poor health as well as with teenage parenthood (for women) and a criminal record serious enough to be a barrier to employment (for men).21 The goals are therefore to keep the numbers of young people who are not in education, employment, or training to a mini- mum and to ensure that the small numbers who do fall through the cracks rap- idly reenter education and training. Some countries are making more progress than others at finding pathways into work for the young, as demonstrated by OECD studies Although debates con- tinue about education versus training, it is well understood that young people need solid qualifications and expertise in broad occupational fields as well as reasoning, problem-solving, and communication skills. Countries that empha- size apprenticeship and quasi-apprenticeship have more effective transition out- comes,22 but a broad mix of apprenticeship, school-based vocational training, and general education is what is most needed. What is called for are coherent national packages that draw from a definable number of key ingredients: * A healthy economy and labor market * Well-organized pathways from initial education to work and further study * Opportunities to combine study and workplace experience 356 I Labor and the Welfare State * Safety nets for those at risk * Effective information and guidance systems * Policy processes that involve governments and other stakeholders Countries can achieve good transition outcomes with different combinations of the key pieces, but sadly the general pattern among OECD countries is a lack of policy coherence and insufficient commitment and resources in most countries Several OECD countries have set up labor market programs designed to help youths in the job market. On the supply side are programs that link schooling to work and second-chance programs that boost the skills of youths who encounter trouble in the job market. On the demand side are programs that raise youth wages through a minimum wage and that target specific employment opportuni- ties at youths However, large untargeted youth employment and training programs gener- ally do not improve either employment prospects or earnings for the young, espe- cially the disadvantaged.23 Indeed, such programs are often a sign of a country's failure to develop more effective pathways for young people. One study found that large-scale programs designed to move young people from unemployment to work, such as the United Kingdom's Youth Opportunities Program and Youth Training Scheme in the 1990s, actually lowered the probability of employment 24 Narrowly targeted and carefully evaluated programs can assist specific youth categories, but far more important are local partnerships between educational institutions, employers, and communities in easing the school-to-work transi- tion. This explains the success of apprenticeships such as those in Germany and Japan, whose Jisseki-Kankei system achieves good transition outcomes for young people.25 If active labor market programs do not help unemployed youth into jobs, then what approach does work when transitions from education to work have failed' The OECD was particularly impressed by the Nordic countries, which devel- oped a "youth guarantee" during the 1980s, after many programs developed ear- lier yielded disappointing results.26 The new programs integrated education, labor market, and wclfarc policies and local delivery mechanisms. Most impor- tant, this safety net is not standardized it deals with each young person on a case-by-case basis. Local providers track early school leavers and work closely with a wide range of community agencies to provide the services youth need. Youth Unemployment Challenges and Ideas 1 357 What Might European Experience Offer East Asia? Some 60 percent of the world's youth live in the developing countries of Asia, and the East Asian economic crisis in 1997-98 showed that here also younger people are more vulnerable to external shocks because they are the first to be retrenched. Although any policy designed to address such challenges must take into account national labor market conditions, educational institutions, and tra- ditions, European experience can offer both positive and negative lessons in this effort: * The right macroeconomic policies-sustainable job-enhancing economic growth-remain an indispensable component of any strategy to redress youth unemployment * Education and training systems must be geared toward the demands of the labor market to expand the pool of talent * Prevention is always better than cure * Investment in better, earlier, and longer education helps prevent social exclusion, but it must develop the attitudes, competencies, and skills required for the world of work. * Countries should consider preschool programs that intervene early in the lives of children and their parents * Effective transition from education to work is vital, and safety nets and universal access to high-quality career information are two key features of effective transition systems. * Labor market programs rarely improve overall job prospects for the young, but narrowly targeted and carefully monitored and evaluated programs can be effective for some young people. Because of its vital social and political importance, the search for remedies to youth unemployment is akin to a worldwide quest for the Holy Grail Solving the enigma is a matter of finding and assembling the right elements. No country has yet cracked the code and found a total solution, although some have made better progress than others. At the heart of all solutions is the need to establish a clear transition and strong connections between general education and continu- ing study at higher levels, and between education and work. 358 1 Labor and the Welfare State Notes I Judith Doyle and Max Nathan, "Wherever Next? Work in a Mobile World," Industrial Society Futures, April 26, 2001 2 Demos, The Wealth and Poverty of Networks Tackling Social Exclusion, Demos collection 12, (London Demos, 1997/1998) 3 Ibid 4. Social Exclusion Unit, UK Government, PreventingSocial Acluson (London March 2001) 5 Presidency conclusions, Stockholm European Council, March 23-24, 2001 6 The International Youth Foundation was set up 12 years ago to improve young people's prospects and now operates in more than 60 countries See Web site The Global Alliance for Workers and Communities Web site is . 7 United States-Indonesia Society (USINDO), "Economic Briefing," available at 8 ILO, Employing Youth Promotng Employment-Intensive Growth (report for the ILO Interregional Symposium on Strategies to Combat Youth Unemployment and Marginalization, December 13-14, 1999, published in 2000) 9 Ibid 10. The percentage of the world's population represented by youths age 15-24 is declining From 1980 to 1995 it dropped slightly, from 19 percent to 18 percent This decline occurred in all regions except Africa, where the number of youths as a percentage of total population continues to rise I David G Blanchflower and Richard B. Freeman, "Cohort Crowding and Youth Labor Markets A Cross-National Analysis," in Youth Unemployment and Joblessness in Advanced Countries, ed David G Blanchflower and Richard B Freeman (Chicago University of Chicago Press and INBER, 1999) 12 ILO, op. cit 13 Ibid 14 Ibid 15 Ambitionsfor Britain. the Labour Party's manifesto, 2001 16 "The European Unit and the member states are fully committed to the goal of full employment and see it as an important way of meeting the challenge of an aging population " Presidency con- clusions, Stockholm European Council, March 23-24, 2001 17 OECD, From Initial Education to Working Ltfe Makitg Transitions Work (Paris OECD, 2000) 18 Tessa Jowell, minister for employment, welfare-to-work, and equal opportunities, Minutes of Evidence to House of Commons Education and Employment Committee, May 17, 2000 19 OCED, op cit 20 Social Exclusion Unit, U K Government (2001), op cit 21 Social Exclusion Unit, U K Government, Bridging the Cap New Opportunitiesfor 16-18- Year- Olds Not in Education, Employment, or Training, report for the Cabinet Office, Deputy Prime Minister, London, July 1999 22 Greece, Hungary, Italy, Portugal, and the United Kingdom have the least successful outcomes, according to the OECD, op cit. 23 Ibid 24 ILO, op cit 25 Jisseki-Kankei is a system of semiformal contact that exists within the strong institutional link- age between schools and employers in Japan 26 OECD, op cit BOX5:2 How the United Kingdom Has Addressed Youth Unemployment by Colin Moynihan Most young people in the United Kingdom enjoy a fairly smooth tran- sition from school to work. Full-time study leads to higher education or to reasonably skilled and secure employment at around age 18, which can include an apprenticeship at age 16. However, a large minority lacks guidance and a clear path to a good job and career opportunity. At any one time, 9 percent of the 16-18 age group does not participate in education, training, or work for long periods after leaving school at age 16. The United Kingdom has systematically engineered new institutions to improve young people's transition to employment. Local Training and Enterprise Councils act as brokers between training providers, employ- ers, and young people, and as facilitators of large public programs, including the New Deal for Young People (NDYP) and Modern Appren- ticeships i Numerous other local, regional, and national agencies also offer education and training Critics, however, question the cost and effectiveness of so many intermediaries and of reforms designed by cen- tral administrations and implemented by bodies that owe their existence to the program. The government is thus trying to reform its approach by providing clear incentives for young people to stay in education through age 18, by providing education maintenance allowances, and through youth cards that would offer access to leisure, sports, and transport. The United Kingdom's NDYP targets youths age 18 to 24 who have been unemployed for at least 6 months. The mandatory program includes a 4-month "gateway" period of advice and support followed by one of four options subsidized employment, full-time education and training, voluntary work, and environmental work. Participants receive an allowance that is to cover exceptional expenses. Each participant is assigned a personal adviser, who provides assis- tance with job search, career advice and guidance, and preparation for contnued on next page 360 I Labor and the Welfare State BOX 5.2, continued the four options, which are designed to help enrollees progress toward finding and remaining in employment: * The employment option provides a subsidized, waged job with an employer * The environmental task force option provides a job with a wage or, more often, a "benefits-plus" package. * The voluntary sector option provides work with a nonprofit organi- zation on either a benefits-plus or a wage basis * The full-time education and training option can last up to a year. If a participant leaves an option early without good reason, he or she may lose 2 weeks of benefits, with subsequent violations resulting in a loss of benefits for 4 weeks and then 6 months However, some 60 per- cent of those entering NDYP find employment before they reach the end of the gateway period. The program provides further assistance if par- ticipants return to the unemployment register. Tables have turned since the privatizations of the 1980s under the Conservative Party The Labour government now favors some private sector involvement in solving both youth unemployment and education challenges, though the issue remains controversial. Community involvement by the private sector can ensure that young people, especially those living in areas of high unemployment, can make contact with the world of work. Such contact can occur through men- toring, work experience, school-business partnerships, or direct business involvement in community projects Partnerships between local govern- ments and employers especially can improve the transition between school and work.2 The United Kingdom has created Employment Zones where public- private partnerships pursue efforts similar to the NDYP for the long- term unemployed (these are not aimed specifically at youth) These initiatives are appreciably more flexible than the New Deal. Personal advisers may combine benefit and training money into packages that enable people to obtain the help they need to return to work. Working Links, the company that runs most Employment Zones, receives most How the United Kingdom Has Addressed Youth Unemployment 1 361 BOX 5.2, continued of its payment only when a client has remained employed for 13 weeks. This approach has not only moved people into employment but also has helped create small businesses, which in turn have hired more par- ticipants in the program. Not only do Working Links advisers find jobs for about half of enrollees-about twice the number for most welfare- to-work programs-but also most participants keep their jobs longer. 1. The numbers in apprenticeship programs declined steadily in the United Kingdom between 1960 and 1990. The number of apprentices in manufacturing fell from 240,000 in 1964 to 54,000 in 1990. The Modern Apprenticeships programs was launched as a national initiative in 1995 to provide young people age 16-25 with training leading to NVQ/SVQ (National Vocaticnal Oualification/Scottish Vocational Qualification) skills at Level 3 or more. In February 1998 there were 117,000 modern apprentices. They are regulated by training frameworks designed for each sector by employers in conjunction with the relevant National Training Organisation and the Department for Education and Skills. For more information see on National Occupational Standards. 2. Organization for Economic Cooperation and Development, From Initial Education to Working Life: Making Transitions Work (Paris: Organization for Economic Cooperation and Development, 2000). CHAPTER 21 Worker Participation in Europe and East Asia by Jacques Rojot ABSTRACT: Worker participation encompasses a diverse array of tech- niques designed to enhance employees' stake in the success of their firms and ultimately in their national economies This chapter outlines different strate- gies in use in Europe, an experience that could prove useful as East Asian countries move from low-wage industrialization to high-tech manufactur- ing, which requires a high degree of worker efficiency and initiative. This chapter focuses on participatory schemes at the enterprise level-elements of social policy that indirectly affect countries' economic and social outcomes. The chapter's guiding framework is management theory rather than economics or political science. Both in theory and in practice, worker participation is difficult to define. The dictionary lists several meanings for "participation," including to take part in something, to share something, to partake of some quality, and to belong to a larger entity All those meanings apply to worker participation We can distin- guish two further aspects of participation: cooperative participation, in which all join together to work toward the same goal, and deliberative participation, in which debate and possible disagreement occur before a decision is made Followers of many different ideologies have advocated worker participation under various guises. For anarchists it might mean free individuals operating in perpetually changing relationships, for revolutionary socialists it might mean that every worker has an equal voice in managing an undertaking. For social democrats participation might take the shape of nationalization, through which ownership of the main means of production and exchange passes to the people; under Christian social doctrine participation implies collaboration between labor 363 364 1 Labor and the Welfare State and capital toward a greater goal. Finally, managers might see participation as a tool to foster a more harmonious working relationship with workers to yield a more efficient and effective enterprise. As might be expected, this chapter focuses on the final point of view Even then another misunderstanding occurs in that worker participation unavoidably addresses power but is often confused with full and unquestioning cooperation by employees with the goals of management As with all industrial relations, par- tictpation involves a mix of conflict and cooperation and enables parties to man- age temporarily an opposition of interests Forms of Worker Participation in Europe Worker participation includes at least six possible strategies for a firm: * Job and organizational redesign and improvements in the quality of life at work * Involvement of employees in managerial decisionmaking * Employee sharing in profits or cost reductions * Employee ownership * Employee identification with the company * Promotion of corporate citizenship Job and Organizational Redesign and Improvement of Quality of Life at Work The working hypothesis behind this strategy is dual. First, employees will per- form better when their work is useful, interesting, and meaningful, and when they hold well-designed jobs for which they are well qualified. Further, when they understand the meaning and usefulness of their work, workers will develop self-esteem and a feeling of achievement and thus will fulfill their potential in performing their work The organization of work should allow workers to apply all their competence to the tasks at hand, and thus to achieve a degree of self- realization Examples include new ways of organizing work and developing organizations, job enlargement and enrichment, ergonomics and health and safety, semi-autonomous work teams, and quality circles (when they include efforts to improve working conditions). Quality circles are an interesting example of a double cultural twist. Designed initially by a U.S. academic expert (Deming) for managing production and later Worker Participation in Europe and East Asia 1 365 extended to the whole organization by another expert Uuran), quality circles were used extensively in post-World War II Japan and arc widely credited with Japan's rapid and dramatic improvement in production and quality. The conccpt then recrossed the Pacific and was implemented in the United States, and it sub- sequently crossed the Atlantic to Western Europe The social and lifestyle condi- tions prevalent in Japanese enterprises favored the growth of quality circles, but they required specific training and a change in scope in U.S. and European set- tings, where they have proved less successful 1 The conditions under which quality circles were implemented made the dif- ference. Work groups were already common in Japanese enterprise, and Japanese managers also allowed rank-and-file employees to express new ideas more easily. The absence of sharp distinctions between negotiation, consultation, and coop- eration in Japan further facilitated employee involvement.2 The basic techniques of a quality circle are simple It consists of 3 to 15 employees, usually ones from the same work group or ones holding similar jobs, who meet at least monthly to share work experiences and identify and solve prob- lems. The team is composed of volunteers who sometimes meet on their own time (formerly in Japan) or on company time (in the United States and often also in Japan nowadays) A supervisor (most often in Japan) or an emerging leader within the group (often in the United States) guides the group, and mectings occur outside the hierarchical context Members learn basic statistical methods and receive all quality control information. The team focuses on continuously improving the management of the work site to boost quality and achieve individual and mutual self-realization. Objectives are thus not limited to corporate goals but also include creating a workplace worthy of members of the enterprise, one that respects its human resources and contributes to limitless progress in the competence of each mem- ber. At each meeting, the group selects issues that may cause problems down the line, whose cause cannot be attributed to other work groups and whose solution can better the working environment and make work easier Facilitators may train members, guide the initial meetings, solve problems within the group, and serve as a liaison between the group and the staff mem- bers controlling the resources to which the group needs access. Expected results include not only better quality, more efficient and practical procedures, and reductions in waste, but also better teamwork and interpersonal relationships as well as higher wages for members Examples of quality circles include health and safety committees, employees' referendums, German-style Mabestimmung, information and consultation schemes, and works councils (at the plant, company group, and European levels). 366 1 Labor and the Welfare State The quality circle concept may also encompass employee membership on the board of directors, user involvement, and, in the narrowest definition, groups exclusively geared toward improving the quality of a firm's services or products. Employee Involvement in Managerial Decisionmaking The hypothesis here is that employees are motivated to perform their best if they have significant input into decisions concerning their work and their fate They can better implement decisions they help make Moreover, nobody knows jobs better than the people who hold them, and thus they have the potential-equal to if not vastly superior to management's-to make the enterprise more efficient. Employee involvement in managerial decisionmaking can entail variations in * Scope: from routine work performance to overall strategy * Extent: from basic information to an equal voice in decisionmaking * Nature: direct or indirect, through representatives elected or appointed, union or not * Type: from informal to highly formalized * Level: from job-site workers to boards of directors * Time span: from execution of decisions to their design * Domain from the social to the financial operations of a firm The German system of codetermination (Mitbestimmung) is one significant example of this type of participation. The German system complements rather than substitutes for collective bargaining. Collective agreements establish mini- mum wages and conditions at a regional level, and two worker-participation channels complement this at the enterprise or plant level. Boards of Directors Worker representation on boards of directors is the best known but probably not the most important form of German codetermination. In Germany, unlike in the United States and most of Europe, company boards are split into two different bodies: the managing board ( Vorstand) and the supervisory board (Aufsichsrat), which appoints and controls the managing board and can request detailed infor- mation but does not perform any direct management function. Three forms of worker representation on the board of directors coexist. The first is limited to enterprises with more than 1,000 employees in the mining, coal, and steel indus- tries. Here the supervisory board generally includes 11 members (in some cases 15 or 21). Five members represent shareholders, and five represent employees; Worker Participation in Europe and East Asia 1 367 together they elect one neutral member Two of the employee representatives must belong to the enterprise, but the remaining three can, and often do, come from outside bodies such as union federations. In this specific model alone, workers also have a representative on the managing board. The second model applies to certain types of enterprises such as joint stock companies that employ at least 500 employees One-third of the members of the supervisory board, whose number depends on company size, must be workers' representatives. If worker representation excceds two, the additional members can be outsiders chosen by the workers. The third model applies to all enterprises with more than 2,000 employees, whatever legal form the enterprise takes. In this case the supervisory board con- sists of an equal number of shareholders' and workers' representatives, with the board's size depending on the number of employees working at the enterprise. Since this board has no neutral chair, the shareholders have the final word in case of deadlock because a shareholder representative chairs the board. Works Councils The most important feature of the German codetermination system, works councils, are compulsory in every plant with more than five employees. By law these councils represent all workers in a plant, whether or not they are union members Howvever, in practice the councils and the unions maintain close ties because a large majority of council members are union members. Unlike in many other countries, where works councils include management representatives, in Germany the councils are composed exclusively of workers' representatives and act as counterparts to managers. Workers elect council members for 4 years by secret ballot and may reelect the members. The size of the council depends on the number of employees. When a council exceeds 100 members, it may appoint a committee dealing with economic affairs. Multiplant (or multioffice) enter- prises establish a general works council. The purview of works councils ranges from a mere right to information and consultation to a veto over decisionmaking, to the most important: codetermina- tion of enterprise decisions. Codetermination means that management cannot legally make any decision without the consent of the works council. Codetermi- nation also gives both sides the right to take initiatives and to involve a conflict- resolving body in case of disagreement Works councils apply their participation rights in three arenas: personnel, eco- nomic, and social. Personnel matters include planning for staffing and vocational training as well as hiring, transferring, and dismissing employees. Economic mat- ters include all aspects of the economic management of the enterprise, including 368 I Labor and the Welfare State investment, production, and marketing. Social matters focus on the social conse- quences of economic decisions Full codetermination applies to an impressive list of social matters, including work rules, work time, pay systems, and health and safety. Works councils maintain veto power in personnel matters, but in eco- nomic matters they hold only the right to information and consultation, except where decisions might cause substantial disadvantages to the work force, such as partial or total closure of a work site or a merger of the company. In those cases management must reach a compromise with the works council. This wide range of participative powers makes the the works council central to both daily life and exceptional circumstances at a work sitc. Employee Sharing in Profits and Cost Reductions The working hypothesis is that employees will be motivated to perform their best if they feel they will benefit financially from their efforts. Examples of this form of participation include piece rates and wage individualization, suggestion boxes; profit-sharing formulas (based on profits, income, or other indexes, calculated for individual employees or for teams); and gain sharing. France is probably the birthplace of profit sharing. The country officially recorded the first example in 1842, and it is the only European nation to require profit sharing by law A complex web of procedures governs this multipronged system, which can include employee ownership. Gain Sharing This wide-ranging process applies to enterprises ranging from private sector firms to public services with an industrial or commercial purpose, without restriction on enterprise size or profit making. Gain sharing can be applied at the level of individual enterprises, components of an enterprise, or groups of enter- prises By law gain sharing must have a collective nature, but it does not have to apply to every employee. Gain sharing must be established by one of four methods of agreement between management and employees. It may be part of a traditional collective agreement with a union, a specific agreement with representatives of a plant union, a special agreement between an employer and a works council (which may or may not be union), and a proposition by the employer ratified by two-thirds of employees (provided that at least one union agrees, if any unions exist at the plant) Specific criteria used to measure gains and compute bonuses are selected from three broad categorics-profit or performance under any definition, an increase in productivity, and any mode of collective payment allowing an effective Worker Participation in Europe and East Asia 1 369 association of the employees to the enterprise-or from any mix of the three. Within that broad definition, the only real mandatory conditions are that gain sharing must be linked to a collective effort to improve the performance of the enterprise, and it must not substitute for any part of an existing wage. Gain shar- ing may not exceed 20 percent of an enterprise's wage bill and an individual maxi- mum. Important tax advantages accrue to both enterprises and employees on their gain shares, particularly if employees freeze their bonuses for 5 years in an enter- prise savings plan (described briefly later). Profit Sharing Profit sharing is compulsory by law but much morc restricted in scope, applying to enterprises earning a taxable profit and employing more than 100 workers. The amount and computation of profit sharing follows a complex legal formula that now includes a fallback version. Profit sharing applies to all employees, and important tax advantages accrue to the company and employees, whose profit share is frozen for at least 5 years. A firm must introduce profit sharing through modes similar to those of gain sharing, or fiscal penalties apply. A profit-sharing firm decides howv to manage the frozen bonuses. Shares may accrue in the form of company shares to employees, insured loans by employees to the enterprise for investment, investments outside the enterprise, and investments in enterprise savings plans Enterprise Savings Plans All enterprises may establish enterprise savings plans, which manage voluntary contributions from employees and possible matching contributions from employers within legally established ceilings, as well as employees' profit-sharing and gain-sharing bonuses. Firms may establish enterprise savings plans through the same methods as gain-sharing plans or may do so outside any agreement. Payments into the plan are frozen for 5 years and invested in either company shares or in the stock market. Tax advantages apply to matching contributions from the enterprise and to profits reinvested for employees Impact of Gain Sharing and Profit Sharing Profit sharing, which is mandatory, is more cumbersome and has had a limited impact except in large enterprises, where it is prevalent. Such schemes cover some 5 million wage earners, of whom 60 percent received a bonus recently. Interestingly, gain sharing, which is voluntary, has growvn annually since a new form was established in 1986, and it is prevalent in small and medium-size busi- nesses Some 3 million wage earners received a voluntary bonus in recent years 370 I Labor and the Welfare State Employee Ownership The working hypothesis is that employees will give their best performance if they feel they are investing in businesses of which they own a part. Examples include individual shareholding plans (employee stock option plan type, generally used in the United Kingdom), matching savings plans, collective shareholding plans, and deferred stock options (the last two usually only for managers or very senior managers) Scandinavian employee ownership also includes wage funds, which are countrywide or voted on by unions. Employee Identification with the Company The working hypothesis is that employees will identify with and remain commit- ted to their company if they feel it has their interests and welfare at heart, or if they feel they belong to a working community that shares common values. Examples of types of participation that foster employee identification include cultural engineering, survey feedback, organizational development, enterprise projects, and perhaps total quality management, although the latter borrows fea- tures from several other types of participation. Promotion of Corporate Citizenship The working hypothesis is that labor is not a commodity like any other. If treated as a responsible citizen of the organization, an employee will behave like a citi- zen. Here the employee is considered as more than a seller of skills in a labor market or an individual in a hierarchy; he or she is also and foremost a partici- pant in a civic order. Thus superiors must treat the employee with dignity, he or she must have a say in his or her life at work, and he or she must benefit from guarantees in the exercise of hierarchical power. Some ways to promote corpo- rate citizenship include implementing industrial democracy in the Scandinavian meaning, establishing nonunion grievance procedures and open-door policies, appointing ombudsmen or employee delegates, and allowing appraisals of man- agement by subordinates. Success of European Worker Participation Strategies None of these working hypotheses have been proven, but the abundant literature tends to indicate that such programs are often effective when sensible and applied by believers. Evidence also underlines the relative efficiency of well- implemented programs. However, the results are difficult to generalize across organizations because careful implementation is crucial. Critical success factors Worker Participation in Europe and East Asia l 371 seem to include full organizational commitment, reliable implementation, and a clear and fair strategy Worker Participation in East Asia Indonesia, the Republic of Korea, Malaysia, the Philippines, and Thailand have each adopted a strategy for industrial development that holds significance for worker participation: they have emphasized, to different degrees, export-oriented industrialization as opposed to import substitution.3 The foremost need was for- merly to structure industrial relations to dampen conflict and encourage economic development, or to attract foreign direct investment to low-wage production areas. Government-the main actor in industrial relations in these countries-therefore restricted the formation of unions, prohibited strikes in essential or export sec- tors, and regulated overtime. Strategic industrialization, however, has brought these countries within the orbit of globalization. The result is growing pressure to remain competitive in a liberalized product market. Instead of focusing on low-cost production of light manufacturing goods for export, these countries are moving to higher-value prod- ucts and innovation in the manufacturing process. At this stage these countries can- not remain competitive simply by controlling labor costs: they also must now rely, to an ever-larger extent, on employees able to operate complex high-tech machin- ery These countries must thus pursue a strategy aimed at both boosting skills and creating a more flexible labor market.4 The strong emphasis on education and the high level of literacy5 in East Asia makes such a strategy possible. Flexibility, however, entails complex tradeoffs Flexibility can be categorized under five main headings.6 * External quantitativeflexibility, which includes temporary work, short-term contracts, part-time work, call contracts, zero-hour contracts, long trial periods before definite hiring, job sharing, and massive use of government- sponsored schemes for integrating young people into the labor force * Externalization, which includes buying instead of making components, using self-employed labor, contracting out work, and using independent contractors or employees "on loan" from other firms * Internal nunier:calflexibihty, which includes the use of variable and flexible working hours, night work, overtime, weekend shifts, and annualized duration of work to better modulate the volume of labor according to demand while keeping labor costs constant 372 j Labor and the Welfare State * Functionalflexibihlty, which includes encouraging employees to perform multiple tasks, hiring consultants, abolishing craft barriers, providing on- the-job and formal training, planning for human resource management, and improving employees' adaptability to change * Wage flexibility, which links the rate of pay to a firm's financial constraints, either indirectly through mechanisms from the simple (piecework rates) to the complex (dual wage scales) or directly through unpaid overtime or pay cuts in bad times Firms need the flexibility to manage the size and wages of their labor force, which implies the use of traditional authoritarian, Taylorist tools Yet efficiency and high-quality production require individual initiative and a positive attitude on the part of employees because many jobs in technology-intensive environ- ments are too complex for managers to clearly outline in advance. Management cannot simply demand or even simply pay for cooperation and initiative-they must be voluntary.7 Firms will have to find ways to manage this contradiction, and some worker participation options may hclp. Social protection in the five East Asian countries will also affect the outcome. Unemployment insurance exists and to a limited extent, only in Korea.8 Despite an aging population and relatively early retirement age, many countries lack effective pension plans, and postcrisis plans to stimulate self-employment and create small and medium-size enterprises face significant challenges. Here again some participative tools might be useful. Three of the six models of participation seem ill adapted to transfer to East Asia, although some features might be of interest. Promoting corporate citizen- ship is linked to industrial development in Western Europe and the high levels of individualism that prevail there Fostering employee identification with the company similarly aims to introduce features already seen as more prevalent in Asia, such as widespread community spirit. Employee oxvnership assumes a well- developed stock market, so the approach will be less relevant in countries with numerous small enterprises and a significant informal sector. However, several forms of participation could prove particularly useful. Participation in Decisionmaking Worker participation can improve the quality of decisionmaking because it brings together different strands of information and enables participants to anticipate consequences. Employees also implement joint decisions faster and better because they understand them. Participatory techniques may be especially helpful when Worker Participation in Europe and East Asia 1 373 used to establish sensible criteria for unavoidable layoffs and priorities for rehir- ing, shortcr work hours, transfers, and early retirement. The benefits of partici- patory methods are morc obvious still in gaining meaningful and informed coop- eration from employees facing complex tasks. The involvement of differing points of view also satisfies procedural justice, because those who may stand to suffer will have been heard Wide-ranging possibilities for such participation encompass the level (from job-site councils to national tripartite economic and social councils), extent (from simple information to codetermination), field (from personnel matters to strate- gic financial choices), and channels (from union to elected to appointed repre- sentatives). The German modcl is often credited with fostering smoother and more efficient industrial relations. Its features are closely linked to German his- tory and social features and need considerable adaptation elsewhere, but the ideas can apply beyond Germany Job and Organization Redesign and Improvement of Quality of Life at Work Employees react in a variety of negative ways to a working life that is tedious and set within an oppressive authoritarian hierarchy Those reactions range from uncooperative attitudes to low-quality work even when the labor market prevents more active responses such as absenteeism and turnover This is particularly true if employees' education and training create a gap between their expectations and the realities of the job. Individuals have many and varied talents; the managerial system and organizational structurc can encourage them to use those talents to benefit the enterprise. Employees at the bottom of the hierarchy also have inti- mate knowledgc of daily work unequaled anywhere else in the organization At the company level, options include restructuring work methods, adapting technology to the needs of employees, and redesigning the organization's social structure. Work restructuring can range from simply introducing buffer zones on assembly lines to replacing assembly lines with moving platforms, culminat- ing in decentralized autonomous work teams. A much cited example of work redesign is the creation in Japan and the spread in Europe and North America of the "lean production model" in the automotive industry Employee Sharing in Profits and Cost Reductions A clear system of profit sharing reinforces cooperation between employees and management, and appears to boost productivity.9 Such systems do not 374 1 Labor and the Welfare State necessarily exclude a union role, which can be instrumental in creating them and ensuring their fairness and clarity. However, profit sharing also builds expectations and puts pressure on management to improve efficiency Of particular interest is a combination of profit sharing with tax exemptions. Employers can deduct profit- and gain-sharing contributions from taxable income and wage taxes, so those mechanisms stimulate employment. Employees, too, can usually exclude the benefits from income tax if several years elapse before they receive them in cash During that time the funds are invested and accrue interest, so they act as forced savings that can help mitigate the lack of retirement pensions and unemployment compensation and work to stimulate entrepreneurship. A further possibility is to contribute frozen profit shares to enterprise savings plans. Held in the name of employees, the shares remain at the disposal of the enterprise for investment and development. Such a system, designed to build complementary pensions, has created a powerful tool for expanding enterprise capital in West Germany and could help develop young enterprises in East Asia. The contributions need to go into a general fund that protects the shares of employees if an enterprise fails Conclusion: Introduce Elements of Participation Carefully It is beyond the scope of this discussion to propose a model for implementing participation schemes, but several factors can prove crucial. First, a national reg- ulatory framework is essential. European experience shows that the more free- dom to experiment these frameworks allow, the better. Cultural features need careful consideration. References to "Confucian tradi- tions" and other East Asian features are often too sweeping to be useful. Still, participation programs should take careful account of national characteristics, especially when they are favorable. For instance, Korea has already partially implemented labor-management councils at the enterprise level,iO as well as a tripartite consultation procedure at the national level.ii Malaysia could tie profit- sharing schemes to its Provident Fund Each country must determine the degree to which unions should be involved in any system of participation. Data on the five East Asian crisis countries indi- cate a low rate of unionization on average, generally around 10 percent, although this information varies with the source and reference year.12 Plural unionism tends to be the norm, and fragmentation into multiple small bodies, as in Malaysia,13 may make coordinated activities difficult. A large informal sector fur- ther complicates the involvement of organized labor, as does the participation of Worker Participation in Europe and East Asia 1 375 some labor movements in separatist and class struggles. Still, unions are an irre- placeable tool for collective bargaining, which in turn is essential to participation. Above all, any participation strategy needs to be multipronged, because employee participation in decisionmaking, for example, is likely to create expec- tations for profit sharing, organizational redesign, and a higher quality of life at work, and vice versa Notes I R Wood, F Hull, and K Azumi, "Evaluating Quality Circles The American Application," Calfornia Management Review 26, no 1(1983) 37-53 2 T Hanami, Managing Japanese Workers (Tokyo. Japan Institute of Labor, 199 1) 3 S Kuruvilla, "Linkages between Industrialization Strategies and Industrial Relations/Human Resource Policies Singapore, Malaysia, the Philippines, and India," Industrial and Labor Relations Review 49, no 4 (1996). 4 S Kuruvilla and C Erickson, "The Impact of Globalization on Industrial Relations in Asia A Comparative Review and Analysis" (paper presented to the 12th World Congress of the International Industnal Relations Association, Tokyo, May 29-June 2, 2000) 5 I Gough, "Welfare Regimes in East Asia and Europe" (paper presented at the parallel Asia- Europe Meeting (ASEM) session to the Annual World Bank Conference on Development Economics Europe 2000, Paris, June 27, 2000) (See Chapter I ) 6 J. Rojot, "National Experiences in Labour Market Flexibility," in Labour Market Flexibiity, Trends in Enterprises, ed Organization for Economic Cooperation and Development (OECD) (Paris OECD, 1989) 7 Ibid 8 G Bechterman, "Labor Market Impacts of the East Asian Economic Crisis" (paper presented to the 52nd Annual Conference of the Industrial Relations Research Association, January 6-10, 2000, Boston, Joint Seminar with the North American Economics and Finance Association on International Perspectives on Unemployment); Gough, op cit 9 A Le Roux, "L'Interessement des salarins contribution i l'identification des conditions de suc- ces," (doctoral thesis, Unuversite de Paris 1, 1999) 10 Chi Sun Kim, "South Korea," monograph in the International Encyclopediafor Labor Law and Industrial Relations (Deventer, Netherlands, and Boston, Mass Kluwer, 1995) 11 D Campbell, "Recovery from the Crisis The Prospects for Social Dialogue in East Asia" (paper presented to the 12th World Congress of the International Industrial Relations Association, Tokyo, May 29-June 2, 2000 For example, compare G Bechterman, op cit, with P Arudsothy, "Malaysia," monograph in the International Encyclopedia for Labor Law and Industrial Relations (Deventer, Netherlands and Boston, Mass. Kluwer, 1995) 12 Compare G Bechterman, op cit, with S Kuruvilla and C S Venkataratnam, "Economic Development and Industrial Relations The Case of South and South East Asia," Industrial Relations journal 27, no I (March 1996), for instance 13 P Arudsothy, op cit CHAPTER 22 Public Employment Services in Europe and Asia by Ruud Dorenbos ABSTRACT In Europe, public employment services play a key role in fighting unemployment. This chapter explores programs and institutions across the continent that help job seekers find work. The chapter also examines public employment services in East Asia and highlights future needs and options in both regions. This chapter reviews European experience with public employment services (PES), which help people prepare for and find decent work PES activities cover a wide variety of tasks, including identifying job openings, helping job seekers assess employment opportunities, and matching job seekers with employers. All Western European countries pursue PES activities, and they are becoming more important in transition and developing countries, including East Asia. Rationale for Public Employment Services Unemployment and unfilled labor demand often occur side by side, even in rela- tively homogeneous segments of the labor market such as submarkets for a spe- cific profession in a specific region. Thus, unemployment-in this situation called "equilibrium" or "frictional unemployment"-prevails. Surplus labor supply or high excess demand may also persist, with many countries facing persistently high unemployment known as "structural unemployment " Employment services are a key instrument for addressing such structural problems in the labor market. 377 378 I Labor and the Welfare State Government does not always have to provide such services; in many countries private placement and training agencies step in. However, PES constitute a large share of the employment services market in many countries, and most analysts recognize that PES can solve problems that the market cannot 2 * Private agencies cannot provide the collective good of improving trans- parency in the labor market because they view information on job seekers and vacancies as a proprietary asset. * Private agencies serve only some employers and the few job seekers who can afford them. * Long-term unemployed need sheltered jobs3 and government aid. o Placement services may benefit from economies of scale that only the gov- ernment can provide. o Private services may exploit weak and inexperienced participants in the labor market. * In European Union (EU) countries, many job seekers receive unemploy- ment benefits, and the government must review eligibility criteria such as availability for work and job search intensity. These arguments suggest that governments play an important role in the employment services market, though the important contributions of private employment agencies to the functioning of labor markets have gained growing recognition. Scope of PES Responsibilities Public employment services usually focus on two main objectives: to make the labor market function more efficiently, and to promote equal opportunities for different groups of job seekers. PES originally linked job seekers and vacancies by providing information and brokering matches. However, with the rise in unemployment in Europe, fewer vacancies are available to serve the unemployed. When full employment is unattainable, countries can at least try to avoid a situa- tion where unemployment hurts some groups more than others, so PES now promote equal opportunities and try to prevent long spells of unemployment. PES typically include the following:4 * Providing information on both the current and future labor market * Offering information on occupational and educational choices Public Employment Services in Europe and Asia 1 379 * Brokering jobs * Implementing active labor market policies, which can include training, placement, and job subsidies * Managing the selection of participants in such services and administering unemployment benefits PES may also play a central role in promoting coherence among labor market policies, though the Ministry of Labor often performs this task. In some coun- tries PES also include responsibilities in immigration and job protection. Each European country handles these tasks differently. For example, in Greece PES rather than employers nominate half the candidates for subsidized hirings. In Ireland PES post job descriptions, job seekers ask for more details, and job offi- cers select suitable candidates-an approach that may better meet the needs of both job seekers and employers. In Portugal PES use computer programs, self- service techniques, and the Internet to quickly match job seekers with jobs.5 Institutional Structure In most European countries PES is a department of the central government, usually the Ministry of Labor (see Table 22.1).6 Only in Australia has a private corporation replaced PES, which contracts with private, community, and public sector agencies to place unemployed people in jobs 7 Public services may oper- ate as a fully integrated part of the Ministry of Labor or as a separate executive agency. The PES head is normally accountable to the minister regarding objectives, resources, and performance. The minister maintains direct control over PES if they are fully integrated into the department.8 However, services for job seekers and employers have become more important, and government organizations are rarely known for being client friendly. PES may also suffer from arbitrary politi- cal intervention, leading to unsound business decisions. Structuring PES as an executive agency within a government department solves most of those prob- lems, but success depends on a good working relationship between the minister and the head of PES. In most EU countries, the government and social partners such as unions and employers jointly administer PES. This allows PES to gain the partners' trust and inclines them to include target job seekers in their collective agreements. However, this presupposes a well-developed system of industrial relations, and even then individual employers may not offer jobs to targeted groups. The suc- cess of this popular model depends on mutual understanding among ministers, Table 22.1 Employment Services in Europe Private Administrative Concentrationl Centralization! Unemployment employment structure deintegration decentralization Financing benefits services Belgium (ONEM)a ONEM implements Three regions are A central office Financing is partly Under federal law, Private employment measures drafted by responsible for placing regulates and public and partly from ONEM is responsible services are allowed three directorates of workers in three monitors regional and unemployment for distributing (state monopoly was the Employment subdivisions of ONEM community offices insurance funds unemployment abolished in the Administration and forvocational benefits 1990s) Like in the training The federal Netherlands, government is temporary work responsible for social agencies are the most security important private organizations in the recruiting process Denmark (AF)b Ministry of Labor Labor market policies Councils at regional Direct expenses are Ministry of Labor PES monopoly was manages PES and services are level have financed by general administers abolished in 1990 No indirectly through usually separated, but considerable taxation through the unemployment restrictions exist for National Labor Market regional offices can influence Some Ministry of Labor Part benefits, with tight private employment Authority Tripartite decide on their own conflicts with the of the costs of supervision by PES services commissions structure national goals of the services can be and ministry of participate in Ministry of Labor and charged to employers benefits and sanctions management at National Labor Market Unemployment for refusing suitable national and regional Authority have insurance is financed work PES implement level emerged through general "activation" taxation and obligations after compulsory employee people have received contributions insurance benefits for 2 years Finland (VTML)' Ministry of Labor Concentration of Decisions concerning State finances the flat A different ministry Private agencies may manages PES at services is high PES placement into rate unemployment pays insurance not charge fees to regional and local aim to provide subsidized jobs and allowance benefits, but PES workers for services levels Tripartite employers and lob other selective Unemployment offices retain the aimed at employment council for labor seekers with one employment measures insurance benefits are power to determine affairs advises the contact person or have been delegated cofinanced by the recipients' availability ministry division to regional and local state, employers, and for work and thus offices employees, and the eligibility for benefits state finances active PES have little labor market policies influence over municipalities' decisions concerning social assistance benefits Germany (BA)d Ministry of Labor and PES are responsible Local PES offices are Financing is partly Unemployment Private employment Social Affairs provides for job brokering, tightly supervised by public and partly from insurance is managed services are allowed legal supervision vocational guidance, headquarters but also unemployment within the PES but only employers Tripartite system unemployment have growing insurance funds administrations pay fees directly influences the compensation, and responsibility The Budgets are prepared Recipients must BA's purpose and the active labor market self-governing mainly by the register at PES and be way it operates policies, but these principle is agencies and available for work divisions are rather systematically applied approved by the Social assistance is separated at all administrative federal government administered by levels different Lander continued on next page Table 22.1, continued Private Administrative Concentrationl Centralization! Unemployment employment structure deintegration decentralization Flnancing benefits services Greece (OAED)6 An independent Despite formal Regional offices have Largest part comes Unemployment PES enjoys a quasi- agency manages PES integration, PES administrative from employer and benefits are monopoly Since 1931 The impact of social functions are largely autonomy, and employee administered by PES, private employment partners is relatively separated decentralization is contributions, with but people entitled to services operating on limited owing to increasing substantial support benefits can maintain a commercial basis conflicts between from ESF Some their claims through (not nonprofit employers and trade financial support the local benefit office organizations) have unions comes from the and sign on with no been banned general taxation obligation to visit through the Ministry employment offices of Labor regularly Ireland (LES)i PES is an independent PES focus mainly on LES is decentralized Unemployment Unemployment Temporary agency The impact of vocational training, by design, but basic benefits are financed benefits are employment agencies social partners is but different reporting through a global administered by a and private placement relatively high owing managers oversee requirements are social insurance separate office agencies are allowed to consensus between each PES function centralized Within the fund (employers, network Registration to operate under employers and trade Local offices have a main PES agencies, employees, and with FAS or LES is not license FAS and unions higher concentration decisionmaking self-employment a condition for benefit private agencies of functions authority at the lower contributions) The receipt cooperate by level is limited state may add extra advertising vacancies funds Active labor in local offices market policies are mainly financed by FAS Unemployment assistance is financed by the general taxation Italy (SCI)9 Ministry of Labor PES services are Legislation Social security INPS directly handles Private agencies are manages PES offices rather separated establishes the duties payments are handled payment of special not allowed in Italy, at all levels Tripartite of regional and local by INPS Employers benefits, such as in with the exception of commissions play an offices, but each level pay most case of effective some headhunting important role at all has some autonomy unemployment dismissal, but SCIs offices levels benefits Employment handle ordinary subsidy and grant unemployment programs are financed benefits through the general budget Training is financed by the regional government and European Social Fund (ESF) The Netherlands Arbeidsvoorziening is Although divisions Although divisions Financing is partly Unemployment Private employment (Arbeidsvoorziening) supervised by the perform different perform different public and partly from benefits are regulated services are allowed, Ministry of tasks, they cooperate tasks, they cooperate unemployment by several separate and some divisions of Employment and closely and often closely and often insurance funds divisions that will Arbeidsvoorziening Social Affairs reside in the same reside in the same become one-stop are also being Some parts of building building centers privatized Arbeidsvoorziening are independent continued on next page Table 22.1, continued Private Administrative Concentratonl Centralization! Unemployment employment structure deintegration decentralization Financing benefits services Portugal (IEFP)h Ministry of Labor and PES deconcentrated IEFP has been given Labor market policy is Unemployment Temporary work Solidarity has overall into 6 divisions, national targets for financed out of the benefits are agencies and private responsibility for PES although they overlap direct job creation global social administered by a placement agencies A permanent council The targets are insurance fund separate office operate under license for social cooperation distributed across contributions network with local and supervision and that includes the regions and offices from employers, offices Integration of must notify PES of government, trade employees, and the benefit and placement vacancies filled on a unions, and employer self-employed A functions is extensive biannual basis associations has an small percentage of advisory role value added tax also goes to the fund IEFP receives 4 7% of the fund's annual income ESF and the European Regional Development Fund are also principal sources of income Sweden (Af)' National Labor Market Sweden has a one- AMS lays down Financing is partly Unemployment Private employment Board (AMS), which stop entry service, general guidelines for public and partly insurance funds or services for assisting falls under the which means that all employment services, from unemployment trade unions and hiring out labor Ministry of Labor, is measures are but employment insurance funds ESF administer are prohibited from responsible for available through the services decide how pays a small part of unemployment requiring payment employment services same organization to implement them active labor market benefits from employed or policies unemployed lob seekers a ONEM = Office National de l Emploi b AF = Arbejdsformidlingen c VTML = Valtion Tyomarkkinalaitos d BA = Bundesanstalt fur Arbeit e OAED = Labor Force Employment Organization fLES = Local Employment Service g SCI = Sezione Circoscrizionale per l'Impiego h IEFP = Instituto do Empreso e Formacao Profissional IAf = Arbetsformedlingen Sources OECD, The Public Employment Service-Oenmark, Finland, Italy(Paris 1996), and OECD, The Public Employment Service-Greece, Ireland, Portugal(Paris 19981 c U' 386 1 Labor and the Welfare State EXHIBIT 22.1 Tripartism in Italy and Ireland A high degree of tripartism, such as in Italy, can contribute to relative neglect of the weakest unemployed In Italy tripartite influences have worked to focus both unemployment benefits and active placement efforts on victims of collective layoffs, rather than on people from small firms who have lost work or those who have been unable to find a first job In Ireland area-based partnerships in disadvantaged regions address economic disintegration and offer a new framework for creating jobs and fighting persistent unemployment A partnership run by an 18-member board includes public employment services, social partners, and community groups social partners, and PES leadership. In Greece, for example, the impact of social partners is limited by conflicts between employers and trade unions. In Ireland, in contrast, partners' impact is strong because of a high degree of consensus between employers and trade unions (see Exhibit 22.1). Countries have often had to adjust PES organizations to accommodate new employment and labor market policies, decentralized government activities, inte- gration of social services, and competition in delivering them. Integrating Public Employment Services Integrating job counseling, unemployment benefits, and active labor market programs is desirable,9 but the same institution need not deliver them (see Exhibit 22.2 and Table 22.1). Sweden, for example, combines job counseling, training, and labor market measures but leaves unemployment insurance to trade unions.1O In Germany and the Netherlands municipalities play a key role in implementing job-creation schemes for the long-term unemployed, because such individuals often rely on social assistance handled by municipalities. However, since municipalities lack the expertise to place unemployed persons in unsubsi- dized jobs, cooperation between municipalities and PES is essential.ii Two new approaches help in organizing service integration:'2 customer- focused integration, also known as one-stop centers, and tiered service delivery. One-stop centers can integrate aid and quickly resolve problems for unem- ployed people. The U.S. government has been a pioneer of this approach, and France, the Netherlands, Sweden, and the United Kingdom, have recently set Public Employment Services in Europe and Asia 1 387 EXHIBIT 22.2 Integration of Public Employment Services in Greece and Ireland In Greece public employment services (PES) formally integrate placement, benefit administration, and referrals into active labor market programs However, people can sign up for and maintain their benefits through the local benefit office with no obligation to visit employment offices regu- larly So despite formal integration, PES functions retain a degree of separation In Ireland, the social benefit network administers some types of subsidies for hiring and the employment office network administers others Differences in culture and attitude between the two administrations have complicated strategic planning and made it difficult for staff to provide clients with information and advice on the full range of programs up one-stop centers. The tiered delivery approach tailors PES services to the needs of job seekers and employers,13 which means that those in greatest need can receive the most assistance. Centralized versus Decentralized Structure Most European countries have decentralized PES, often by giving regional PES boards autonomy (see Table 22.1). Decentralization makes services more accessi- ble and targets them to specific regional needs, but central management provides less detailed oversight. Sound policy guidelines and trust in local judgment are therefore essential prerequisites. Too much decentralization can lead to uncer- tainty among job seekers and employers about what they can expect from PES, and practical problems may occur when regional PES organizations use differ- ent information systems. A highly decentralized structure introduced in the Netherlands in 1991 faced such problems. As a result, the government reduced the autonomy of the regional PES boards after 4 years, though they still have considerable freedom in choosing the mix of policies and determining how to implement them (see Exhibit 22.3). Managing Unemployment Benefits Administering unemployment benefits through PES, as in Austria, Germany, Japan, and Norway, can enable the system to identify highly motivated or unmo- tivated job searchers and to adjust regulations to motivate people to accept jobs 388 1 Labor and the Welfare State EXHIBIT 22.3 Decentralization in Finland, Denmark, and the United Kingdom In Finland and Denmark decentralization combined with monitoring of procedures and rewards for performance has encouraged regional managers to follow national strategies and have per- haps motivated staff Most important, decentralization has helped managers adapt labor market strategies to local needs In the United Kingdom locally administered job-search programs have allowed staff to main- tain close contact with both clients and employers This cooperation helps in tracking labor mar- ket trends So for example, as part-time jobs have grown, they have been pushed as an attractive option for young people and employers to hire them. Such a structure may also encourage administrative efficiency and improve monitoring. However, disadvantaged groups normally rely on social assistance rather than on unemployment benefits, which munici- palities usually provide, so PES may pay less attention to people who need help the most (see Exhibit 22.4 and Table 22 1). Some analysts favor close administrative relationships between unemploy- ment benefits, PES, and local one-stop functions.i4 In the Netherlands, one-stop offices register newly unemployed persons and use the information to determine enrollees' benefits and their "distance" to the labor market, which determines the assistance they receive for reintegrating into the market. Integrating unemployment benefits with PES is particularly important in countries facing mass unemployment. However, such integration can produce unexpected and dramatic results. For example, the initial version of Poland's unemployment compensation system caused an "added-worker effect": all regis- tered unemployed people were eligible for unemployment benefits, they need not have been previously employed, and the entitlement period was open ended This encouraged many "secondary workers"-household members not included in the labor force under the centrally planned government-to enter the labor market. The number of unemployed people registering in the labor offices was thus greater than the number who were laid off. To avoid paying unemployment benefits to people who were actually outside the labor market, Poland quickly modified the eligibility criteria Public Employment Services in Europe and Asia 1 389 EXHIBIT 22.4 Combining Unemployment Benefits with Public Employment Services in Portugal, Ireland, and Finland The Organization for Economic Cooperation and Development reports that in countries where people must register with public employment services (PES) to receive unemployment benefits, more unemployed participate in PES In Portugal, this has helped raise the proportion of the unemployed who register with PES to high levels In Ireland, where such an obligation does not exist, only about half of the unemployed register for PES, making it difficult for job counselors to quickly match job seekers with vacancies In Finland unemployed workers often refuse to enter training or accept short-term jobs because doing so resets their unemployment duration at zero Because PES staff members allo- cate subsidized jobs according to how long people have been unemployed, this greatly reduces their chances of receiving a subsidized job This interaction between benefit administration and the allocation of subsidized jobs does not promote efficiency in labor market services Financing Most EU countries finance active labor market policies and unemployment insurance benefits through both general revenues and employer and employee contributions. The European Union also finances a significant portion of active labor market policies (Table 22.1). In Germany, employees' and employers' contributions largely finance the PES, Bundesanstalt fur Arbeit (BA). The BA has its own budget, which requires the approval of the federal government The BA transfers any budget surplus to the reserve fund and draws on the fund to cover budget deficits. The federal gov- ernment provides the liquidity needed to balance the accounts in the form of an interest-free loan The BA uses its funds mainly to provide unemployment bene- fits and active labor market services such as vocational training, vocational reha- bilitation of persons with disabilities, and promotion of job creation. In Greece and Portugal, employer and employee social security contributions similarly finance unemployment benefits and a large portion of active labor mar- ket policies, although the European Social Fund and the European Regional Development Fund (Portugal) also finance employment subsidies and vocational training. The PES agency in Greece (OAED) operates its own budget, though 390 I Labor and the Welfare State ministerial decrees specify spending targets for active labor market policies in consultation with the OAED board of directors. EU funds are channeled to OAED through the labor ministry. PES in Portugal (Instituto do Empreso e FormaSao Profissional-IEFP) pre- pare the annual budget based on projected social security contributions, the min- istry's spending targets, and budget submissions by regional directorates. After assessments from the governing board and the supervisory commission, the cen- tral office decides how much will go to regional directorates in detailed program categories. Each region then manages its own budget (see Exhibit 22.5). In Ireland, unemployment insurance benefits are financed through the social insurance fund, which collects contributions from employers, employees, and the self-employed. The state may add a small subsidy from the general budget. The proportion of the fund allocated to paying unemployment insurance varies according to need. Assistance for job seekers entitled to no or reduced insurance benefits accounts for over three-quarters of spending on unemployment bene- fits, entirely financed-like other social assistance programs-through general taxation In Denmark, Finland, and Italy, insurance contributions partly fund employ- ment services but do not cover labor market policies. General taxation covers direct expenses such as staff wages through the Ministry of Labor. In the Scandinavian countries PES now can ask for fees from employers. Measuring Market Shares and Output Placing job seekers is usually seen as the core business of PES. However, employ- ers advertize and fill vacancies, and job seekers look for and find jobs, through avenues other than PES, so the role of PES in this placement market is of inter- est. PES may execute the brokering function through various means. i Active brokering entails a large role for the job officer, who when notified of an opening contacts suitable candidates and selects some for an interview. This method quickly connects job seekers with jobs, but it requires heavy staff input. Under semi-open advertising, job seekers can ask for more details regarding an advertised vacancy and job description at PES, and a job officer selects suit- able candidates. This method may be more successful in meeting the preferences of both job seekers and employers. Under fully open advertising, PES make public all details regarding vacan- cies, but job seekers must take all further action. This can occur through self- service job shops where job seekers can find information through billboards and computers. Job seekers can contact a PES officer for more information or address Public Employment Services in Europe and Asia 1 391 EXHIBIT 22.5 Budget Allocations to Regional and Local Offices in Portugal In Portugal indicators such as working-age population and number of registered and long-term unemployed are used to determine regional budgets for public employment services Each region then manages its own budget The national budget is usually revised twice a year In the course of the year, requests for adjustment usually come from the regional directorates after feedback from their local offices The headquarters agency may authorize transfers of funds between regional budgets, and regional directorates themselves can shift funds However, any funding shifts between training centers and employment services must be decided at the national level This process of budgetary adjustment in consultation with various bureaucratic layers allows a flexible response to local needs and program take-up the company directly. A disadvantage is that it is difficult for PES to determine the status of vacancies and whether information provided by PES helped a com- pany fill a vacancy or a job seeker find a job. Data from the Organization for Economic Cooperation and Development (OECD) show that PES market shares vary from less than 5 percent of all hires in the United States and Switzerland to around 30 percent in the United Kingdom and Italy. The average market share is 16.4 percent, which means that almost 85 percent of placements occur through other search and recruitment channels. Those channels include newspaper advertisements and informal chan- nels as well as private placement agencies. Measuring Impact Countries commonly use market shares, registration rates, and filling rates to measure the impact of PES, though not all are meaningful. Job matching is only one type of PES activity and may thus not convey the actual impact of services. PES are most important as a search channel for individual job seekers, especially the unemployed. Indicators of the net output or net impact of PES-the degree to which PES activities make a difference in the functioning of labor markets- are still needed. The answers to two questions can shed light on net impact: * Do PES activities increase the total number of filled vacancies? * Do PES activities lead to more equality in employment opportunities? 392 1 Labor and the Welfare State Recent analysis indicates that PES activities exert a limited impact on the functioning of the labor market.16 The evidence suggests that they help improve the labor market situation of disadvantaged groups such as the long-term unem- ployed, but deadweight, substitution, and displacement effects are high even among those groups. To maximize net output, PES often give high priority to people who have dif- ficulty finding a job. Placement activities are often insufficient, so PES usually offer training and placement subsidies PES may outsource some of these activi- ties to external agencies. People trained with a PES subsidy or who receive other PES services may find jobs through search channels other than PES. However, those cases do not count as PES placements, so the placement concept may not prove useful. PES should not concentrate entirely on disadvantaged groups for two reasons. Some labor market shortcomings such as lack of information and underinvest- ment apply to most participants. And concentrating entirely on disadvantaged groups might discourage employers from registering their vacancies with PES. Evaluating Public Employment Services Evaluations of public employment services are indispensable in justifying and improving them. The three evaluation phases include ex-ante evaluation, evalua- tion during implementation (performance evaluation), and ex-post evaluation (impact evaluation).'7 Most studies have focused on the impact of active labor market policies on the employment and earnings of unemployed job seekers. Studies of PES in Canada, Hungary, Poland, the United Kingdom, and the United States can shed light on the effectiveness of these services. 18 In Hungary and Poland an evaluation of the impact of specialized employ- ment services (ES) on employment and earnings suggested that the ES did not exert a significant impact.'9 However, if the economy is improving, some sub- groups who chose to use ES assistance-particularly women-fared significantly better than those who did not. The results underscored that ES consisted of more than job referral. In Hungary ES acted as a one-stop shopping center for all forms of reemployment services as well as income support for the unem- ployed; administrative costs per person were relatively small. In the United Kingdom an evaluation of state employment services indicated that most job seekers used more than one method of job search21 and that use of PES is countercyclical, with greater reliance on PES during recessions. It indi- cated that job centers exert the most beneficial impact on groups who tend to Public Employment Services in Europe and Asia 1 393 make more use of them-the less skilled and the long-term unemployed. The evaluation further showed that the use of job centers raised the probability that people would move from unemployment to employment. Another study showed that job seekers who turn to PES over other job-search methods had shorter unemployment spells than those who did not.22 An examination of 18 evaluations of job-search assistance/employment serv- ices concluded that job-search assistance is one of the most successful active labor market programs 23 In general, such assistance costs little. Not surprisingly, however, the results depended significantly on whether the economy was grow- ing and on the availability of public funds 24 All studies indicate that PES activities lead to private payoffs-that is, PES help integrate the unemployed back into work, especially the long-term unem- ployed. However, the studies did not examine whether PES are meeting their goal in the most efficient way and, therefore, whether they should be the main actors in employment services. Nor did evaluations assess program design and implementation, staffing, and intensity and quality of services-all of which help determine whether employment services will succeed Most Western European countries have reached social consensus on the desirability of employment services: their right to exist does not depend on their effects, and measurement of success or failure is designed to improve their effi- ciency However, the questions surrounding PES can be broader in transition and developing countries because immature institutions can produce unsuc- cessful outcomes Nonetheless, overall, the administrative costs of PES per unemployed person are relatively small, whereas the social value of PES serv- ices is large Role of Private Employment Services Until the 1990s regulations were so tight in most European countries that few private job placement agencies existed because they were restricted to specific segments of the labor market. However, Australia, Switzerland, the United Kingdom, and the United States, have a long history of private job placement services, while other countries such as the Netherlands have allowed commercial temporary work agencies.25 In the 1990s many European countries abolished the state monopoly in job brokerage In most cases PES continued to exist as a government-funded organization, but job-placement activities and efforts to reintegrate target groups were privatized (see Table 22.1). 394 1 Labor and the Welfare State This partial privatization of PES activities stemmed from several factors. Dissatisfaction with PES pushed employers toward other channels of place- ment, constraints on public spending hurt PES, and people became convinced that private companies can produce higher-quality products and services more efficiently. The "carrier-wave theory" holds that PES help the disadvantaged only when the services retain a high market share because only then will employers regis- ter their vacancies with PES. If PES concentrate their activities on the disad- vantaged, companies will be reluctant to list their vacancies. If this theory holds, it would not make sense to allow PES to compete with private placement serv- ices. However, private employment services do not necessarily reduce the mar- ket share of PES,26 because many unemployed persons have to register with PES to receive unemployment benefits. This gives PES a clear advantage over pri- vate competitors. Furthermore, PES can offer services for free, whereas employers have to pay for private employment services. PES may therefore con- tinue to play their traditional role even when the state monopoly in job broker- age is abolished. The roles of public and private employment services may become more complementary, with PES concentrating more on the disadvan- taged. This is a third model between state monopoly and maximum private sec- tor involvement (see Exhibit 22.6). Data on private placement services are scarce, particularly in transition and developing countries. However, the expectation is that the role of private job placement agencies will grow. Transition and developing countries should devote extra attention to the legal and regulatory framework governing private employ- ment agencies. Although the market for job placement on a purely commercial basis may be limited in many countries, governments can enhance the role of the private sec- tor. In the extreme case, a job network of private, community, and governmental organizations can replace PES, as occurred in Australia. Because agencies receive a fee for each unemployed person they place in a job and win extra fees for plac- ing long-term unemployed, the system is strongly driven by outcomes. The Australian government took considerable risks in introducing this far-reaching reform in a short time span,27 and employment ministers all over the world will follow the pioneering step closely.28 The Netherlands has pursued a more cautious approach, with reintegration activities still publicly financed but implementation outsourced to private com- panies. In this model the government still plays an important role because it finances employment services, prioritizes labor market reintegration measures, chooses among competing bids, and regulates the employment services sector. Public Employment Services in Europe and Asia 1 395 EXHIBIT 22.6 Collaboration between Private and Public Employment Services in Sweden, Switzerland, and the Netherlands In Sweden public employment services (PES) accept advertisements for their vacancy lists from private agencies if the advertisements include information on employers or if PES are familiar with the listing firms This allows PES to offer services to a broader public, because private agen- cies have established themselves in areas that PES have been unable to supply or that would be inappropriate for PES intervention In Switzerland private employment agencies predate PES, and the law expressly calls for collaboration between public and private placement agencies Facing growing unemployment, the government encouraged regional placement offices and private employment agencies to exchange information on vacancies and job seekers and provided common training to both partners The regional offices also have the option of contracting with private agencies to provide services Representatives of both public and private placement agencies initially expressed fears and reservations Local governments were especially concerned that private agencies would confine their efforts to the most easily placed workers, leaving the regional placement offices to contend with the most difficult-to-place job seekers and discouraging employers from contacting them The private placement agencies, for their part, feared that the regional placement offices repre- sented unfair competition, since they were subsidized by the state Thanks to a working group that brings the parties together regularly to discuss problems and determine how best to collabo- rate, those reservations have practically disappeared In the Netherlands a public-private partnership begun in 1995 resulted in a joint venture between PES and two private organizations to deliver services to employers in the area of tem- porary work In all these cases cooperation has broadened the expertise of both public and private employ- ment agencies and enhanced the ability of job seekers to obtain the information they seek on vacancies and training With the rapid growth of private employment agencies, public and private operators have had to cooperate. Such cooperation can take many forms, includ- ing detailed contractual arrangements, more general contracts with close moni- toring of financial results, informal arrangements organized at the local level, or some combination of the three. Complementary arrangements between the pub- lic and private sectors are most appropriate for specific labor market segments and services. 396 1 Labor and the Welfare State Future Directions Globalization, enhanced competition, technological development, and changing state roles are bringing profound changes to PES as former bureaucratic organi- zations move toward customer-oriented services. Changing labor markets are also forcing PES to rethink their institutional role, working methods, and goals. Setting Priorities for Concerted Action The new EU employment policy focuses on PES, and the 1998 employment guidelines define four major pillars (see Exhibit 22.7).29 EXHIBIT 22.7 EU Employment Guidelines 1998 Public employment services (PES) are a key institutional component of the 1998 European employ- ment guidelines and are based on four pillars * Improving employability The 1998 guidelines assign a high priority to improving the employa- bility of the unemployed PES are directly and indirectly responsible for assisting job seekers by providing counseling in job-search techniques, establishing work and training incentives, and improving access to training and work experience * Developing entrepreneurship The employment guidelines call for encouraging self-employ- ment and job creation in social services and other activities that serve needs that the market does not yet recognize * Encouraging the adaptability of businesses and their employees Aside from helping the unem- ployed, PES need to manage structural change within enterprises and economic sectors by help- ing workers and employers minimize the risks of unemployment and find alternative jobs Cooperation with social partners can help relieve potential tensions * Strengthening equal opportunities between women and men PES should promote female par- ticipation in the labor market and more equal gender representation in occupational sectors with a traditional gender bias PES can play important roles in improving the employment prospects of handicapped workers and fighting unequal access to jobs among racial and eth- nic minorities, as well as among older and younger workers Public Employment Services in Europe and Asia 1 397 Promoting Access to Vacancies PES are expected to provide ever more help to employers seeking new employ- ees. PES in various EU countries have already introduced technological innova- tions and developed new ways of bringing employers and job seekers together. The Internet will potentially have the most profound implications. An interest- ing example is the Internet-based self-service system in Sweden, which allows applicants to search vacancies by occupation or regional municipality as well as find career information In the Netherlands the kansmeter assesses the "dis- tance" of job seekers from the labor market to help tailor employment services to their needs. Ensuring Systematic Case Management The European employment strategy encourages PES to focus on the individual needs of problem groups, such as youth and the long-term unemployed, through careful diagnosis Interviews and tailor-made action plans have proven effective in preventing long-term unemployment.30 Coordinating Delivery of Services to Job Seekers PES can offer a full range of labor market services such as income replace- ment, information, counseling, brokerage, and training. PES themselves need not perform all these tasks, but they should coordinate them. Exploiting Synergies between PES and Other Actors PES must not only coordinate different employment services but also interact with other participants in the labor market, such as regional and local authori- ties, trade unions, educational institutions, and private employment agencies. Using PES to Facilitate International Labor Mobility PES heads in the European Union and European Economic Area (EU/EEA) have agreed to strengthen their cooperation, with the principal aim of facilitat- ing international labor mobility within the EU/EEA. 398 1 Labor and the Welfare State Employment Services in East Asia Indonesia, the Republic of Korea, Malaysia, the Philippines, and Thailand all maintain public employment services, though they are at quite different stages. In Korea PES are responsible for job-search assistance, vacancy tracking, and placement. They also administer unemployment benefits and offer counseling and career information. In 1998 the Korean government introduced comprehensive measures such as job retention, job creation, vocational training, job placement, and social protection to address growing unemployment. It has expanded the number of government-managed PES agencies from 52 in February 1997 to 134 in March 1999 (Table 22.2); local governments run 281 employment services. Korea has relaxed regulations on private placement agencies, and that meas- ure plus rising Internet use in Korea has boosted the number of private agencies. PES, too, have made use of the Internet by adapting the Canadian WorklnfoNet, an electronic labor exchange system known in Korea as Work-Net. The system aims to provide information on vacancies, training, career information, and employment policies, though in practice it gives limited information compared with the Canadian system. The Korean government is now developing databases of registered unemployed and has introduced a worker-profiling system that can be used to identify long-term unemployed early Despite these changes, PES agencies are still considered less than efficient; counselors are inexperienced and they provide limited information. Malaysia PES handle registration and placement, counseling services, occu- pational guidance, promotion of occupational mobility, and regulation of private employment agencies. Public employment agencies focus only on domestic labor, whereas private agencies include foreign labor. Private employment agencies must register under the Private Employment Agencies Licensing Unit and report regularly to PES (before the crisis quarterly, now monthly) on the num- ber of registrations and placements. The crisis increased the need for clear labor Table 222 Number of Public and Private Employment Agencies in Korea Public, managed by Public, managed by Date Private central government local govemment February 1997 1,432 52 285 March 1999 1,756 134 281 Source Korea Labor institute, 'Policy Options for Income Support and Active Labor Market Programs A Synthesls of the Korean Experience.' paper prepared for the conference 'Labor Market Policy Its Implications for East and Southeast Asta,' Manila, March 1-2, 2001 Public Employment Services in Europe and Asia 1 399 market information, as retrenchment of workers was common. PES reacted by registering those workers and making special efforts to help them find jobs. The system successfully placed many workers in 1998. In the Philippines the economic crisis hastened the establishment of employ- ment services. The main PES tasks are to register the unemployed, establish a national registry of labor skills, provide job placement information, and mediate employment programs and services. Under the Public Employment Service Offices Act 1998, the government aims to establish a PES office in every province, key city, and highly urbanized municipality. Those local offices are sup- posed to detect firms that may lay off workers and try to take measures to help both workers and firms. The government improved the labor market information system and created the Phil-Jobnet, an Internet-accessible job-matching system, in November 1998. Those who do not have direct Internet access can use public workstations in 43 regional offices. The number of PES agencies and private job agencies has grown. The latter offer paid services to job seekers and employers, locally and abroad, monitored and regulated by the government. A radio station broadcasts job openings and is collaborating with public employment services to provide other job-related information. Thai PES offer assistance to job seekers and employers through 85 branches, 9 located in Bangkok The branches register and place unemployed, provide job counseling and career guidance, and organize job fairs to enable employers and job seekers to meet and to provide job orientation and career guidance. PES offers a free skills test to help match job seekers to vacancies. The public employment office encourages Thais to work abroad by providing information on foreign jobs. The government supervises private agencies to protect workers from dishonest agencies. Private agencies are quite popular, especially among unskilled workers. PES in Indonesia are responsible for registering unemployed people and for providing information to job seekers and matching them with employers. PES provide services for domestic placement only, whereas private agencies also pro- vide placements abroad The government strictly regulates private agencies. Cross-Regional Experience Much European PES experience has relevance for East Asia, though not in any mechanistic way, because institutional structures, labor market characteristics, and economic restructuring differ markedly even within countries. 400 1 Labor and the Welfare State (a) Improving the labor-nmarket information system Information technology can make labor market demand and supply more transparent and enhance the impact of PES. In the 1980s the International Labor Organization (ILO) iden- tified four major obstacles to building labor market information capacity in developing countries: inadequate relationships between customers and produc- ers, underuse of existing resources, an information gap in the informal sector, and inadequate understanding of methodology.31 European PES are develop- ing a full-fledged self-service system called EURES that will offer first-line counseling and job-brokerage services. The system, which uses the Internet and other media, makes services accessible anytime and anyplace and covers the entire European labor market. The system will be linked to easily accessible personal employment services East Asian countries, with numerous foreign workers, might benefit from a regional approach to labor market information systems. A job database would enable employers to advertise vacancies outside their national borders, greatly enlarging the labor pool and helping PES fill vacancies. (b) Training personnel: Labor offices in Eastern Europe have been unable to provide high-quality employment services,32 and similar problems have arisen in East Asian countries when the number of registered unemployed rises sharply and quickly. Besides a bigger workload, labor office employees confront a relatively new problem: placing job seekers in a shrinking labor market. Counselors need more information and training. European PES prefer to hire people with backgrounds in public administration, but even they need addi- tional training.33 Developing and using a regional database would also require training. Determining how best to allocate funding for training staff members is critical. On-site training as well as study visits to other regions could prove beneficial. (c) Improving services and information for job seekers and employers Europe has substantial experience in this area, both positive and negative. Employers often consider PES clients low skilled, not very productive, and lacking the right atti- tude toward work. The very existence of PES can raise expectations: jobless people who otherwise might have left the market remain registered because information and counseling give them new hope of finding a job. Results, how- ever, are the key objective, and that means producing a high job-placement rate because both job seekers and employers must see PES as responsive. Public Employment Services in Europe and Asia 1 401 (d) Involving social partners (tripartism)n- PES systems in Europe that are jointly administered by the government and social partners are the most popular and successful. This structure depends on mutual understanding among ministers, social partners, and PES leadership. Such an approach, especially if it includes employers' organizations, makes PES more effective: it is difficult to provide high-quality Job training without fine-tuning curricula to employers' needs. The specific role of the social partners depends on the national context, but their involvement seems especially advisable in countries with a well-developed indus- trial relations system. (e) Establishing active labor market prograis Although active labor market pro- grams appear to have limited direct impact on workers' employability and earn- ings, they serve important social goals Europe has won clear consensus on the need to pursue such policies, and PES are likely to play an important role in fur- ther developing such policies in East Asia, and in targeting vulnerable groups. Programs that promote self-employment in rural and informal sectors seem especially important in East Asia, and efforts to foster entrepreneurship and micro, small, and medium-size enterprises deserve priority (f) Private and public employment services: Strengthening the PES system- particularly setting up an efficient labor market information system-rather than further deregulating the placement systcm appears to be a sensible first priority in East Asia. Governments can develop and run such information systems. The aim should be to diffuse information widely, although the government may decide to charge private placement agencies for the information they use. Government regulation of private placement services deserves careful design. European experience shows that collaboration between PES and private employ- ment agencies can enable job seekers to obtain the information they need on vacancies and training. (g) Flexibihlty. Flexibility and responsiveness in PES organizations are key, given the changing labor market (h) Assessing training needs: PES can play several important roles in designing and implementing training and education policies. They can monitor labor mar- kets and observe where matching problems occur, and they can make labor 402 1 Labor and the Welfare State market forecasts to detect and prevent future skill shortages. PES can also pro- vide information on the labor market to parents, children, the training and edu- cation sector, and the private sector. PES can fund occupational education and training to address mismatches in the labor market. Conclusion Overall, an approach aimed at counteracting unemployment and boosting employment must rest on three interlocking subsystems: * A macroeconomic policy conducive to creating new jobs * An active labor market policy that includes efforts to improve labor mar- ket efficiency and flexibility * Effective coordination of programs to combat unemployment Combining a range of policies into a cohesive program requires rigorous adaptation of the institutional framework and the willingness of policymakers from different arenas to build a strong coalition. Notes I The chapter focuses on the five East Asian countries most affected by the 1997-98 crisis Indonesia, the Republic of Korea, the Philippines, Malaysia, and Thailand. 2 H. Mosley and S Speckesser, "Market Share and Market Segment of Public Employment Services" (discussion paper, WZB, Berlin, 1997) 3 Sheltered jobs allow the unemployed to obtain some work experience and thus improve their position in the labor market 4 The range of PES activities varies from country to country 5 OECD, The Public Employment Office in Greece, Ireland, and Portugal (Paris OEDC, 1998) 6 P Thuy, E Hansen, and D. Price, The Public Employment Service in a Changing Labour Market (Geneva International Labor Office, 2001) 7 Ibid 8 Ibid 9 The OECD jobs Study. Evidence and Explanations (Paris OECD, 1994) 10 Thuy et al., op cit 11 Alternatively, municspalities can hire private compames to place their clients in jobs 12 Thuy et al, op cit Public Employment Services in Europe and Asia 1 403 13 Ibid 14 Ibid 15 OECD, The Pubic Employment Office in Denniark, Finland and Italy (Paris OECD, 1996) 16. J De Koning, "Aggregate Models for Aggregate Impact Analysis," in Labour Market Policy and Unemployment Impact and Process Evaluations in Selected European Countries, ed J De Koning and H Mosley (Cheltenham, England Edward Elgar, 2001) 17 J De Koning, "Evaluation of Employment Policies The Dutch Experience" (paper presented at the Danish Presidency Conference on the effect and Measuring of Effects in Labor Market Policy Initiatives, Kolding, Denmark, May 24-26, 1993). 18. These evaluation summaries are based on H W Risher and C Fay, eds, New Strategiesfor Public Pay Rethinking Government Compensation Programs (San Francisco Jossey-Bass, 1997) 19 Christopher T O'Leary, "Evaluating the Effectiveness of Active Labor Programs in Poland," Upjohn Institute Technical Report No 98-012 (1998), and Christopher T O'Leary, "Evaluating the Effectiveness of Active Labor Programs in Hungary," Upjohn Institute Technical Report No 98-012 (1998) 20 O'Leary 1998, op cit, p 97 21 P Gregg and J Wadsworth, "How Effective Are State Employment Agencies? Job Center Use and Job Matching in Britain," Oxford Bulletin of Economics and Statistics 58, no 3 (1996) 22 J M Thomas, "Public Employment Agencies and Unemployment Spells Reconciling the Experimental and Non-Experimental Evidence," Industrial and Labor Relations Review 50, no 4 (July 1997) 23 Amit Dar and Zafiris Tzannatos, Active Labor Market Programs A Review of the Evidencefrom Evaluations (Washington, DC.. Social Protection Department, Human Development Network, World Bank, 1999) 24 Ibid 25 Firms use temporary work not only to cope with fluctuations in output but also as a recruitment device When labor demand is rising, firms are often uncertain whether their increased need for labor is permanent Temporary labor allows them to buy time Firms may also be uncertain about the productivity of newly recruited workers. Firms can monitor temporary workers for some time and then decide to offer a permanent contract Thus, even with an official state monopoly in job brokerage, private agencies can fill a considerable share of vacancies In the Netherlands, the share of temporary work agencies was almost as high as that of the PES In most countries, however, the market share of the private agencies was much smaller 26. F Buttler and U Walwei, "Different Situational Arrangements for Job Placement," in Institutional Frameworks and Labor Market Performance, Comparative Views on the US and German Economies, ed F Buttler, W Franz, R Schettkat, and D Soskice (London and New York Routledge, 1995) 27 Thuy, et al ,op cit., p 129 28 Thuy, et al, op cit 29 European Commission, The Modernization of Public Employment Services in Europe-Three Key Documents (Brussels Employment and Social Affairs, Employment and European Social Fund, European Commission, 1999) 30 Ibid 31 L Richter, Upgrading Labour Market Information in Developing Countries Problems, Progress, and Prospects (Geneva ILO, 1989) 404 1 Labor and the Welfare State 32 R J Dorenbos, "Labour Market Adjustments in Hungary and Poland " (Thesis, University of Groningen, Netherlands, 1999) 33 In several countries PES may themselves hire long-term unemployed as part of a work experi- ence project However, this system is not always very effective Witness Finland, where subsi- dized hires lasted for only 6 months, imposing a heavy burden on experienced staff members See OECD (1996), op cit CHAPTER 23 Active Labor Market Policies in the Republic of Korea and Europe by Bernard Gazzer and Remy Herrera ABSTRACT This chapter examines the effects of the 1997-98 crisis on employment in the Republic of Korea, and the safety net and active labor market policies the government pursued in response. The chapter explores "transitional labor market" policies-schemes tailored to respond to non- traditional working situations-which have recently emerged in Europe as a possible model for countries facing economic transformation The Republic of Korea is often cited as a capitalist success story for its part in the "Asian miracle."' Before the economic crisis, Korea experienced the fastest real growth in gross domestic product (an average of 8.7 percent a year from 1985 to 1995) and the lowest unemployment rate (2.0 percent in 1996) among all Organization for Economic Cooperation and Development (OECD) countries 2 But the picture changed dramatically during the 1997-98 economic crisis, as Korea's per capita gross national product dropped from US$10,550 in 1997 to US$7,970 in 1998,3 while its rank in purchasing power parity per capita fell from 24 to 55. The government responded by introducing or expanding a wide range of labor market interventions and social protection programs For example, the gov- ernment broadened unemployment insurance and vocational training, wage sub- sidies, and job placement efforts It also launched a public "workfare" scheme, expanded a temporary noncontributory livelihood protection program, and cre- ated a social pension scheme. The wage-subsidy program was massive, paying up to two-thirds of a worker's wage and covering 800,000 workers in 1998. But despite these extraordinary efforts, significant gaps remained. 405 406 1 Labor and the Welfare State Despite its recent high degree of government activism, Korea continues to face tight economic constraints because it belongs to a weakly integrated region, it is highly dependent on business cycles, and it has limited stabilization tools. Korea also faces classic "European problems" such as a rapidly aging work force, concerns about work incentives, and demands for flexible labor markets.4 Europe's beginning embrace of the new concept of "transitional labor market" policies to fight unemployment may offer new possibilities for Korea and other East Asian nations. The Impact of the Crisis on Korea's Labor Market Open unemployment more than quadrupled in Korea between October 1997 and February 1999. This dramatic tightening of the labor market resulted from mas- sive bankruptcies,5 large-scale corporate restructuring, and dwindling employ- ment in all sectors In February 1999 unemployment peaked at 8.7 percent of the work force, with more than 1.8 million workers officially registered as unem- ployed.6 Rising unemployment was accompanied by a significant fall in labor force participation rates,7 a shift from formal jobs to informal activities, and an increase in agricultural and unpaid family employment (implying growing underemployment). Joblessness affected mostly the young and the less educated, with production workers in the industrial sector accounting for 53 percent of the newly unemployed. The crisis also brought marked change to the composition of employment, as many workers shifted from higher-wage, indeterminate-term jobs to lower-wage, determinate-term jobs. The OECD estimated that permanent workers declined by 0.9 million people by 2000, a phenomenon "unprecedented in Korea, where legislation had limited the ability of firms to dismiss workers prior to early 1998."8 By mid-1998, 6 5 million workers-half of the 13 million-person work force-were part-time and contingent employees. Also, real wages fell in Korea by 12 5 percent from mid-1997 to the end of 1998, as the share of salaries in national aggregate income declined sharply from 55.0 percent in 1997 to 50.8 percent in 1998, whereas profits rose from 9.2 per- cent to 9.4 percent. Despite unions' relative weakness (the number of unionists fell from 1.667 to 1.402 million between 1993 and 1998), this trend led to grow- ing labor disputes throughout the country.9 Active Labor Market Policies in the Republic of Korea and Europe 1 407 The Social Impact of the Crisis Massive unemployment and sharply declining incomes caused a rapid increase in inequality. The income of the four lower quintiles of urban households declined in the first half of 1998, whereas the income of the top quintile rose steadily, thanks to higher real interest rates and favorable tax systems. iO Both the International Labor Organization and the United Nations Development Programme reported a rise in national poverty to more than 12 percent, corresponding to 5.5 million persons. The number of ultrapoor- households with per capita consumption less than 80 percent of the poverty line-skyrocketed from 1.1 million in 1997 to 3.0 million in 1998, and the num- ber of marginal poor (per capita consumption of 80 percent to 100 percent of the poverty line) rose from 1.7 million in 1997 to 3.3 million in 1998.11 The number of poor (around US$8 per capita per day) also rose from 2.8 million to 6.2 mil- lion, and the number of near poor (per capita consumption at 100 percent to 120 percent of the poverty line) expanded from 2.7 million to 4.1 million. Korea experienced a sharp increase in the proportion of urban poor,'2 from 7.5 percent in the first quarter of 1997 to 22.9 percent in the third quarter of 1998. This rate dropped in the last quarter of 1998 but was still more than twice the rate of the precrisis period Homelessness was widespread in Korea's largest cities, a phenomenon almost unknown before the crisis.13 Because public schools charge fees, at least 10 percent of unemployed low- income households withdrew their children from school from 1997 to 1998,14 and dropouts were significant even among the upper classes.15 These changes went in tandem with a decline in public spending on education16 and are partic- ularly worrying given the significant role of education and human capital in long- run economic growth. Family transfers of funds to the poorest attenuated some of these effects; some 50 percent of unemployed relied on family support. However, the crisis damaged the traditional family structure itself, as indicated by marked increases in domes- tic violence, divorce, child abandonment, and abuse. Suicides climbed from a monthly average of 620 in 1996 to more than 900 in 1998. The crisis exerted other socially destabilizing impacts: crime rose by 17 2 percent and violent offenses by 14.5 between 1997 and 1998, and property crimes committed by youth under age 18 rose by 17.5 percent. Those indicators show the dire plight of many Koreans and widespread social unrest. 408 I Labor and the Welfare State Korea's Social Assistance and Labor Market Policies The Korean government pursued an austerity program that included slowing inflation and controlling the money supply (by tightening credit, raising interest rates, increasing savings, and removing indexation); reducing budget deficits (by raising taxes and cutting expenditures and subsidies for the public sector); and balancing foreign payments (by devaluing the exchange rate). Encouraged by international financial institutions, Korea integrated this austerity program into efforts to liberalize financial, public, and labor markets through privatization, deregulation, and openness to foreign capital. Despite continuing doubts about the necessity of a welfare state in Korea, the government also decided to implement a series of programs to assist low-income unemployed households with few assets. One such program was Support for Living Costs, Temporary Livelihood Protection, which included loans for hous- ing and new businesses, as well as assistance with tuition and school lunches. The Livelihood Protection Program, designed to assist people "unable to work, such as the handicapped, the elderly, and children," strictly excludes unemployed people and those whose income exceeds about 20 percent of the average nonfarm wage or who have family members capable of helping them. Although this program cost 1.45 billion won in 1999-compared with 1.50 bil- lion won devoted to unemployment benefits-the average allowance remained well below the national poverty line In 1999, for example, benefits ranged from US$44 to US$124 per person per month.17 Many analysts stress this program's inefficiency: "the incremental budget of the Korean Livelihood Protection Program only reached 7 percent of the nev poor while total coverage (such as for the former and new poor) dropped from 32 percent prior to the crisis to just over 17 percent in 1998."18 Even though Korea already had a high degree of wage flexibility, the govern- ment's core labor market goal was to further expand flexibility by allowing firms to freely lay off workers and offer determinate-term jobs. The Basic Employment Act, adopted in February 1998, gave firms the right to lay off workers in case of "urgent managerial needs," including deteriorating business conditions that could force the firm into bankruptcy, reorganizations undertaken to improve labor productivity, and the introduction of new technology Court decisions also recognized business transfers, mergers, and acquisitions as other "managerial reasons" for laying off workers. Firms that want to use layoffs to reduce costs must give 60-day advance notice to worker representatives on steps to avoid layoffs and Active Labor Market Policies in the Rep6bIic of Korea and Europe 1 409 standards for dismissing workers and must sometimes notify the ministry of labor 19 Abuse of the newly legalized layoff system was widespread: firms regis- tered only 10 percent of the 120,000 employees reportedly laid off in 1998 To help create new jobs, the government subsidized the wage costs of firms that hired workers dismissed for "adjustment need," paying twvo-thirds of such wages for small workplaces and one-third for large workplaces or firms The gov- ernment also awarded wage subsidies to cmployers who tried to retain redundant employees by reducing wages and bonus payments, suspending recruitment, cut- ting working and extra hours, closing temporarily, dispatching workers to affili- ated firms, and promoting early retirement. In 1998 almost 4,200 firms received such subsidies, involving more than 780,000 workers The government tem- porarily extended these subsidies from 6 to 8 months in the first half of 1999, funding the program through a payroll tax of 0.3 percent on employers. Although the Public Employment Service tried to improve job placernient, the placement ratio remained below 10 percent of job seekers and lower still for low-wage, "3-D" (dirty, dangerous, and difficult) workers in 1998. This poor efficiency reflects acute staff shortages; public agencies retained only one staff member for every 8,000 unemployed persons. Meanwhile the government pro- gressively eliminated curbs on the job categories private employment agencies could address, and in 1998 more than 1,650 fee-charging offices had established files on 1,435,000 job seekers. The Manpower Leasing Act of February 1998 also relaxed the rules govern- ing private agencies that dispatch temporary workers, whose numbers reached 42,000 by the end of 1998, employed by some 800 such agencies. Unlike in most other OECD countries, in Korea regulations do not limit the employment peri- ods of temporary workers. The government strongly encouraged early retirement, and more than 260,000 workers retired "at the recommendation of the employer" in 1998, although 15 percent of all firms gave children of people who retired early prefer- ence in hiring The result was that the 50-59-year-old age group saw a large drop in employment in from 1998 to 1999 To respond to rapid economic restructuring, the government systematically expanded vocational training. Funded by both a tax on employers (0.1 percent to 0.7 percent of payroll, depending on the size of the workplace) and general tax revenues, this program absorbed about 15 percent of the budget for labor market policies in 1998, and more than 10 percent of the 68.4 percent rise in labor market spending in 1999 People covered by unemployment insurance are 410 1 Labor and the Welfare State eligible for a three-stage reemployment training program lasting from 1 month to 1 year.20 Stipends paid to participants not receiving unemployment benefits vary from 60 percent to 90 percent of the minimum wage, and the stipends fall during training. Nevertheless, only about a quarter of all unemployed received vocational training in 1998. And just 134,000 of 340,000 persons who participated in the program finished their courses, while fewer than 27,000 succeeded in finding a new job. The government developed a voucher program (in 1998) and a tailor- made training system (in 1999) to raise the proportion of hired trainees, fulfill the needs of business, and develop competition among training centers, which included colleges and universities as well as traditional educational institutions. In 1998 central and local government also created more than 437,000 public works jobs lasting 4 to 5 months, whose holders repaired public buildings and improved the environment, among other tasks. Wages ranged from 50 percent of the minimum wage for unskilled labor (former industrial daily workers) to 60 percent of average nonfarm earnings for skilled labor (unemployed university graduates). This program was designed to give priority to low-income unem- ployed people with dependents and with few assets. Despite criticisms that public works jobs went to ineligible candidates, the government decided to intensify employment creation in early 1999, when the problem of joblessness was most acute. That policy produced positive outcomes, even though the jobs created were only short term, and reflects the preference of Korean authorities for creating jobs rather than assisting unemployed workers. For example, the state-owned Korea Telecom and Korea Electric Power Corp. used massive investments in new infrastructure projects to create additional short-term jobs.2' Improving Unemployment Insurance Until very recently, Korea had no social security system or social safety net for the unemployed, testifying to the government's laissez-faire approach to social issues. The only significant welfare scheme available to Korean employees was Industrial Accidents Insurance. 22 The first priority in the government's welfare policy was consistently job creation,23 and numerous firm-level welfare provi- sions also enhanced the key role of employment in welfare.24 Thus Korea ranked among the lowest in the world in social spending as a share of central govern- ment expenditure (8.1 percent in 1985) in the 1980s.25 Active Labor Market Policies in the Republic of Korea and Europe i 411 The expansion after March 1998 of employment insurance and other counter- measures against unemployment was closely tied to unions' acceptance of the relaxation of job security. The latter measure responded partly to pressure from Korean employers' organizations since the end of 1996 for a flexible work-hour system (introduced in 1997) and free layoff procedures Government action to establish and to some extent improve the social insurance system was also designed to moderate social unrest, which had been rising since the late 1980s.26 Korea's unemployment benefit scheme, launched in July 1995 under the Employment Insurance Act, was the direct result of a wave of strikes and labor protests from 1987 onward.27 When the financial crisis struck in 1997, Korea was thus one of four Southeast and Pacific Asian countries (with China, Japan, and Mongolia) with an unemployment benefit program However, the explosion of unemployment and its duration completely overtook that scheme because only 7 percent of the unemployed received an allowance. The worsening crisis stimulated a rapid expansion of the scheme in March 1998 under the Wage Bond Guarantee Fund Act and the Labor Welfare Fund for Small and Medium Enterprises Act The state-managed compensation fund is financed by contributions shared equally between employers and employees. Contribution rates vary with com- pany size, amounting to 0.9 percent of payroll for small firms, 1 1 percent for medium-size firms, and 1.3 percent for large firms. The unemployment bene- fit (job search allowance) is half of a worker's salary during the month before the dismissal. In March 1998 the minimum amount was raised from 50 per- cent to 70 percent of the minimum wage. Because the minimum wage is about 25 percent of national average manufacturing earnings, the minimum unem- ployment benefit is extremely low The maximum is 80 percent of the average manufacturing wage. Average benefit levels are thus lower than in other OECD countries.28 The duration of unemployment benefits remains relatively short, reflecting the view that the scheme should be an incentive to employment rather than long-term income support 29 Benefits last from 3 to 6 months for workers age 30 to 50, and from 4 to 7 months for workers over age 50. Unemployed workers whose benefits expired between July 1998 and June 1999 were eligible for a special 2-month extension, prolonged until the end of 1999.30 In June 1999 the minimum contri- bution period required to qualify for unemployment benefits was shortened from 1 year to 6 months. To encourage job search, the state pays those who find a newv job within less than half the period for which they are entitled to benefits-a lump sum equal to one-third of the remaining benefit. 412 1 Labor and the Welfare State Unemployment insurance was initially limited to workers in companies with 30 or more employees, but the scheme was expanded in January 1998 to firms with at least 10 workers, in March 1998 (after legislative changes to enhance labor flexibility) to firms with 5 or more workers, and in October 1998 to enterprises with fewer than 5 workers. Beginning in July 1999, the scheme covered tempo- rary workers (who were employed at least I month a year), part-time workers (who work more than 18 hours a week), and daily employees (who work less than 30 8 hours a week). The system excludes public officials (who enjoy exceptional employment security), teachers at private schools, postal workers, new workers over age 60, and workers over age 65 When unemployment benefits expire, the unemployed are eligible for public works jobs and vocational training. Thus they can potentially qualify for assistance for a total of 18 months, though in practice few employees receive such extended help.31 Despite the program's rapid expansion, just under 15 percent of the unem- ployed (up from 1 9 percent initially) received benefits in January 1999, the worst period of the crisis.32 Unemployment allowances totaled US$1 billion from June 1996 to June 1999 for some 700,000 persons, an average of US$1,500 per benefi- ciary, underscoring the extreme modesty of the system. Recent Trends: Encouraging or Worrying? Korea's unemployment rates declined sharply and rapidly, and wages grew from early 1999, as economic growth recovered. However, the government cannot afford to be complacent, and further large-scale job losses may result with restructuring of Chiaebol conglomerates.33 Youth unemployment remains high, standing at 12 percent in February 2000 in the 15- to 24-year age group, com- pared with around 7 percent before the crisis.34 The government is offering funds to encourage large companies to offer internship programs to young people, but the program is still quite limited. A serious unresolved problem is the scarcity of secure long-term jobs; the Korean Labor Institute (KLI) forecasted that firms would create some 823,000 new jobs in 2000, most of them short-term. Long-term unemployment is becoming a structural problem The KLI indi- cates that people out of work for more than 6 months accounted for almost 20 percent of total unemployment in November 1999-considerably more than before the crisis According to the OECD data the over-6-month unemployment rate rose from 16.8 percent to 21.3 percent for men and from 10 3 percent to 13.1 Active Labor Market Policies in the Republic of Korea and Europe 1 413 percent for women between 1998 and 1999 The OECD projected unemploy- ment rates of 4.5 percent for 2000 and 4.1 percent for 2001. This jobless total is far higher than Korea has been used to, and the country's days of unemployment in the 2 percent or 3 percent range are over. Thus the need for an enhanced social security system and labor market polucies remains undiminished The Emerging Concept of Transitional Labor Markets In Europe a single authority now wields tools such as determining interest rate and currency devaluations, and member states can use budgetary weapons only within the strong constraints of the Maastricht criteria. This means that if a demand or supply shock hits a sector, a country essentially cannot react with clas- sic macroeconomic tools. This reality is not yet fully apparent because the euro is new and the European zone has not faced major shocks so far, but the ques- tion will inevitably arise of how to respond to such events. Local and regional efforts to encourage worker mobility, offer wage adjustments, and address con- tinued changes in the labor market will therefore assume growing importance As in other regions, the center of gravity in European labor markets is shift- ing from large firms offering long-term jobs to networks offering shorter employment contracts but richer reemployment prospects. This new career profile links to the expansion of the service economy, the feminization of the work force, and the need for periodic retraining Other trends include greater worker autonomy and responsibility, the growth of a results-sensitive compo- nent of wages, reduction in hierarchical levels, and intensified competition between firms and shareholder pressures. WVorker preferences are also shifting toward more discontinuous careers that include personal or family leaves. Such departures from classic full-time employment and the negotiated processes that create and allocate new arrangements are known as "transitional labor markets" (TLMs). The German economist Gunther Schmid articulated this perspective and inspired a host of studies during the 1990s, many financed by the European Community.35 According to this analysis, two main solutions are possible for man- aging "transitions." In the luxury version, a highly qualified new entrant to the work force first travels around the world to gain some knowledge of a firm's mul- tiple activities and then takes a relatively safe first job. The employee then accepts a position with more responsibility and tries to diversify his or her experience 414 1 Labor and the Welfare State before moving on to higher-level challenges. That pattern can accommodate some personal leaves to acquire missing skills or raise children, and firms can adapt com- pensation to include premiums, stock options, and pension funds to each situation. The problem is that this luxury version is relevant for only a few large firms and high-level employees. Most workers can choose among only a few ill-paid and even stigmatizing transitions. TLMs offer a second solution. Transitions can occur in five areas: within a job (from full-time to part-time, from salaried to self-employment), between training institutions and employment (either initial transition, such as appren- ticeship, or retraining); between unpaid socially useful activities and employment (childcare, but also voluntary or political activist work); between unemployment and employment (job search); and between retirement and employment. The idea is to "make transitions pay" by periodically reexamining each worker's situ- ation TLM measures fall between two controversial social policy tools: provid- ing a universal basic income that delinks work and income, and offering employ- ment subsidies that tie full-time work and income closely.36 The midway position may foster more equilibrium between paid work and useful but unpaid social activities. The TLM approach can open alternatives to older workers. For example, early retirement is a costly arrangement often involving no flexibility. Fragmentary French evidence shows that early retirement can be a positive deal for low-skilled and vulnerable older workers, because they receive state payments for a social "resting" policy. Some highly qualified workers can also benefit from retiring early, even if they would prefer to continue working, because they have higher purchasing power and the freedom to look for pleasant unpaid activities such as traveling. However, traditional skilled workers appear to suffer negative outcomes, because they are often strongly attached to their professional life and their enter- prise and can be condemned to inactivity and social reclusion. The overall impact appears to be neutral in the short term and negative in the long term. The TLM approach can enable older workers to opt for part-time early retire- ment while they pursue complementary part-time work with, for example, non- profit institutions Although such arrangements are more complicated than full- time early retirement, the participation of municipalities and nonprofit institutions can lower the costs to the state, and higher productivity and better social integration can compensate firms for their extra costs. With its more systematic use of negotiated mobility, TLM could also help rebalance opportunities between men and women in both the workplace and the home. In Denmark, for example, women take most leaves, which almost always Active Labor Market Policies in the Republic of Korea and Europe 1 415 widen gaps between male and female career prospects. Throughout their careers women face fewer job choices, practical constraints (complex schedules and lack of promotions), and unfair competition with men. Danish labor market policies now include job rotation, which helps ensure that no one stays unemployed without obtaining training or work experience. Workers may take voluntary, paid, 6-month, or 1-year leaves for training and childcare, in which they are replaced by an unemployed person. After the leave period, the previously employed person returns to work, while the person who replaced him or her has gained 6 months to a year of work experience and an employment network, and thus can more easily find work. This rather costly and sophisticated device can help some stigmatized and discouraged persons. The concept of employability encompasses some of these provisions Employability-enhancing policies are interventions on the supply side of the labor market (the worker) that are designed to improve recruitment and ease employment decisions.37 They are based on tailor-made and future-oriented interventions, including personal evaluations of competencies and retraining tra- jectories performed by specialized staff of employment agencies. The beneficiary agrees to these elements. In its 1998 employment guidelines, the European Commission promoted both an "activating" and an "employability-enhancing" approach The guide- lines require each member state to propose a personalized route back to employ- ment for every adult unemployed for more than I year and youth unemployed for more than 6 months. The goal was to engage a minimum of 20 percent of the unemployed in training or retraining programs. Each member state submits an annual report on its labor market policies showing how it has reached the goals or giving reasons for noncompliance. This approach depends on decentralization because local governments must manage such transitions. Germany illustrates the initiative power of local Lander, whereas France has explored global budgets for local agencies. The main risk is inequality among regions, with richer ones fostering effective transitions while poorer regions offer more limited menus. Transfers of some public funds can help adjust for local variations in unemployment and poverty levels.38 Privatization could emerge as a mean of fostering competition and innovation in a field often criticized for bureaucratic bias, but it can also foster competition among local agencies, and experience underlines the need to avoid "skimming," in which agencies place the easiest to place and train the easiest to train. Though elaborated in nations with fully developed social protection systems, TLM approaches may also be relevant for Korea and other East Asian countries, 416 1 Labor and the Welfare State as they, like European nations, face constraints in defining and implementing macroeconomic policies. TLM options enable communities and regions to com- bine state intervention with for-profit and nonprofit initiatives in both formal and informal sectors. The new challenge is for communities to negotiate novel arrangements that limit the costs and rigidities in labor markets while also addressing rising inequality and social exclusion. Notes I World Bank, The East Asian Miracle Economic Growth and Public Policy (New York Oxford University Press, 1993). 2 The Republic of Korea entered the Organization for Economic Cooperation and Development (OECD) as a permanent member in December 1996 3 The data are taken from I'amar Manuelyan Atinc, "From Economic Crisis to Social Crisis," in East Asia Recoviery and Beyond (Washington, D C World Bank, 2000), pp 123-51 4 Ibid, p 144 5 Because Korean unemployment figures do not include self-employed people whose businesses have gone bankrupt, unemployment rates would have been 0 3 or 0 4 percentage points higher if measured by OECD standards Many small, healthy enterprises closed down because of the "credit crunch " 6 Numbers from the National Statistical Office of the Republic of Korea 7 The participation rate (the labor force as a percentage of the total working-age population) declined from 62 2 percent to 60 7 percent between 1997 and 1998, whereas the inactive popula- tion increased by 1 2 million people 8. OECD, Republic of Korea, Economic Studies (Paris OECD, 2000) 9 In 1998 there were 129 disputes involving 146,065 workers and 1,452,096 lost working days, com- pared with 78 disputes involving 43,991 workers and 444,720 lost working days in 1997 See National Statistical Office, Korea Statistical Yearbook 1999 (Seoul National Statistical Office, 2000), pp 178-79. 10 Ibid. 11 Nicholas Prescott and Nanak Kakwani (1999) "Impact of Economic Crisis on Poverty and Inequality in Korea " Processed World Bank, WVashington, DC 12 This poverty incidence is derived from national poverty lines, on the basis of consumption expenditures, and is calculated for urban households only 13 For example, Francois Godement reports that the number of homeless living at the Seoul central railway station was counted at 3,000 in August 1998 See Francois Godement, "Models and Politics for Asian Social Policies," paper prepared for the Asia-Europe Meeting (ASEM) Social Lessons Prolect, May 2000, p 10, available at 14 Students enrolled in middlc school declined by 6 percent, from 2,012,000 in 1997 to 1,897,000 in 1998 15. This rate was as high as 36 percent m some sections of the upper school 16 The Korean Statistical Yearbook 1999 (National Statistical Office, op cit ) shows a declining share of education in government expenditure from 1996 to 1998 Active Labor Market Policies in the Republic of Korea and Europe I 417 17 According to personal communications w ith the Ministry of Health and Welfare 18 Atinc, op cit 19 Firms must notify the ministry if layoffs exceed 10 employees in firms with fewer than 99 employees, 10 percent of employees in firms with between 100 and 999 employees, and 100 employees in firms with more than 1,000 employees 20 These benefits are halved during the secoid pcriod of courses and canceled during the third 21 Intelligence Economic Unit, South Korea, second quarter 1999, p 22 22 Industrial Accidents Insurance revealed difficult working conditions in Korea 23 J Dreze and A Sen, Hunger and Public Action (Oxford Clarendon Press, 1989) 24 Active promotion of firm-centered welfare schemes includes provision of secondary education on company premises, dormitories, and medical facilities. 25 You Jong-il, "Labor Institutions and Economic Development in the Republic of Korea," in Workers, Institutions and Econominc Growth in Asia, ed G Rodgers (Geneva International Institute for Labor Studies, 1994) 26 A national pension scheme was established in 1988, but few Koreans were initially covered The scope of Industrial Accidents Insurance (created in 1964) and the patchy Medical Care System (implemented in 1977) were also extended 27 This replaced a former small-scale scheme, the Severance Allowance System, which served as both a pension fund and an unemployment allowance designed to help young people enter the labor niarket 28 As the International Labor Organization (ILO) notes, this scheme allows for some vertical redis- tribution of income by having a maximum benefit but no maximum contribution (see ILO, "List of ILO Activities in the Asian Region," annex in Decent WVork in Asia ILO Activities in the Region, Thirteenth Asian Regional Meeting, Bangkok [August 2001]) 29 The minimum benefit period was doubled to 60 days in 1998 Before the crisis, it was between 30 and 210 days 30 For example, a person 30-50 years old who had been in the Employment Insurancc Scheme from its inception in July 1995 would receive benefits for 4 months, plus the 2-month extension, for a total of 6 months 31 OECD, op cit, p 162 32 Atinc, op cit 33 Proposed foreign ownership (potential bidders being General Motors, Ford, Daimler-Chrysler, and Fiat) aroused strong union opposition urging nationalization To back this demand, workers at all Korea's main auto plants went on strike for a week in April 2000 34 Personal communication with the Ministry of Labor 35 For initial formulations, see Gunther Schmid, "Is Full Employment Still Possible7 Transitional Labor Market as a New Strategy of Labour Market Policy," Economic and Industrial Democracy 16 (August 1995) 429-56 Also see G Schmid and B Gazier, eds, The Dynamics of Full Emnployment Social Integration by Transitional Labor Markets (Cheltenham Edward Elgar, 2000) 36 P Van Parijs, Real Freedom for All (Oxford Oxford University Press, 1995), and E S Phelps, Rewarding Work How to Restore Participation and Self-Support to Free Enterprise (Cambridge, Mass Harvard University Press, 1997) 37 For a comprehensive analysis, see B Gazier, ed , Employability Concepts and Polices (Berlin Institte for Applied Socio-Economics, 1999), p 329 38 This opens the thorny issue of relevant indicators, discussed in B Gazier, "L'articulation justice locale/justice globale Le cas des "marches transitionnels du travail," Revue Econoniique (Mav 200)0) CHAPTER 24 Evaluation and Monitoring of Active Labor Market Policies by Christine Erhel ABSTRACT: The cost and effectiveness of programs designed to help people find jobs vary widely, so monitoring and evaluating such programs is essential. Europe's experience in pursuing different types of evaluation sheds light on their strengths and weaknesses, and lessons from this expe- rience could save time and effort elsewhere. Active labor market policies are a key clement of employment strategies in Europe. Such policies fall into three broad types: * Measures to reduce mismatch and enhance job search (public employment as well as placement and counseling activities) * Measures to improve the skills of job seekers (training) * Measures to subsidize employment opportunities for target groups Each category encompasses many different types of programs, which can range from short-term to 5-year efforts (such as Swedish relief work and French youth employment contracts). Target groups also vary, although in Europe the main targets are youth and the long-term unemployed. Because governments can choose among different measures to reach a given policy goal, they need to assess the relative efficiency of such measures to under- stand what works best.i Public finance is another argument for monitoring and evaluation. Average expenditures on active labor market policy are quite low-in 1997 the typical Organization for Economic Cooperation and Development country spent about 0.8 percent of gross domestic product (GDP)2-(ranging 419 420 I Labor and the Welfare State from 0.1 percent of GDP in Japan to 2 percent in Sweden However, participants in and expenditures on active labor policies have grown over the last 20 years in most European countries, and some programs are very expensive in unit costs. Governments have clear incentives to obtain information about the impact of programs to optimize their budgetary choices From a macroeconomic perspective, the effects of active labor market policies are unclear: some countries treat these programs as a tool for moderating wage increases, whereas others use them to boost wages. The impact on unemploy- ment depends crucially on this wage effect. This complexity calls for systematic evaluation. It also explains why methodological debates are important in this field: good methodologies are a prerequisite to reliable results. Countries need to tailor their policies to their national labor markets, another important argument for effective evaluation as a basis for policymaking. Countries differ in their implementation and funding of active labor market poli- cies. Some public employment services are directly responsible for programs, whereas others receive only referrals. Some countries administer their programs at the national level, whereas others rely on regional or local levels Funding may be governed centrally or regionally These differences lead evaluators to focus particular attention on the institutional context of active labor market policies. Most European countries have made monitoring and evaluation schemes an integral part of their active labor market policies. Europe's heterogeneity of experi- ence and relatively low expenditures on evaluation have favored cheaper and practi- cal methodologies, with comparative research assuming growing importance. The chapter is organized around the main approaches to evaluating labor mar- ket policies These include monitoring, a prerequisite to basic knowledge about labor market policies; microeconomic studies, which have spawned classic debates on the use of experimental and nonexperimental methods and cost-benefit analy- sis; and aggregate impact and macroeconomic evaluation. Each section highlights good practice. Monitoring Active Labor Market Policies Efforts to monitor active labor market policies should ideally include the following: * Goals specified by political and administrative authorities at the national, regional, and local levels * Defined indicators that allow goals to be measured Evaluation and Monitoring of Active Labor Market Policies I 421 * The actual monitoring process, based on goals and definitions plus the information-gathering framework * Feedback loops to ensure that policymakers take observed patterns into account3 A successful monitoring system depends on planning at the early stages of policy formation, simplicity of indicators, and links to reliable labor market sta- tistics. Monitoring begins at the policy definition stage. the choice and design of programs should include definitions of clear goals and corresponding indicators. Practically, this means that laws and regulations creating new policies should explicitly include monitoring.4 Basic indicators for a monitoring process are financial data (budget allocated versus budget used), information on participants (total numbers and target groups or sectors), and costs per unit (people served per hour expended). Monthly or at least quarterly data are crucial The overall system should be simple enough to apply to all administrative levels-local, regional, and national. Agencies responsible for administering labor market policy (and political authorities) need reliable and comprehensive information about numbers of par- ticipants, program costs, their targeting, and all types of implementation prob- lems. Monitoring facilitates evaluation, since the availability of detailed firsthand data can relieve the need to gather information later, saving time and money. A good monitoring system should rely on a two-level information system that pro- vides both simple and synthetic indicators to help the implementation and decl- sionmaking process Monitoring also encompasses two "feedback loops." First is feedback between monitoring and implementation any indicator turning "red" should lead to reme- dial action. Second, the results of monitoring should inform decisions on the design of new programs. But in Europe, especially, other incentives also have emerged for active labor market policy monitoring, such as the following: Fiscal Constraints and Public Sector Reform. All European countries have faced tight fiscal constraints over the last 10 years owing to the Maastricht criteria, the economic downturn of the early 1990s, and high levels of public debt, just as they sought to reform public services. Modernization of public services has taken several forms, including decentralization, introduction of private sector management methods (for instance, man- agement by objectives), and market-oriented goals In France, Germany, and Sweden, decentralization entails giving local employment offices full 422 I Labor and the Welfare State responsibility for an overall budget that they allocate among programs for the unemployed. This process promotes the introduction of monitoring systems, because local providers, central agencies, and local branches of central agencies all need information on the programs. Role for the European Commission in Activating Labor Market Policies and Sharing Information among Countries. Since the 1994 Essen summit, the European Council has consistently stressed monitoring of employment trends and policies of member states, and the commission has worked to establish employment indicators.' The November 1997 Luxembourg sum- mit underscored the need for member states to exchange information and good practice, and European labor market initiatives include clear incen- tives to develop monitoring systems. For example, the European Union (EU) employment policy guidelines include a series of indicators of results, and the European Social Fund, which cofinances many active labor market programs, requires effective monitoring. European Experience with Monitoring Of course, the reality of monitoring in Europe differs from the ideal. Most coun- tries are now setting up monitoring schemes, but, with the exception of Sweden, they do not systematically pursue quantitative indicators. Identifying the diffi- culties that have hindered development of monitoring despite significant incen- tives is useful, and several effective efforts to develop comparative monitoring are also illuminating. Problems that have hindered monitoring include the following: * Lack of data. * Lack of financial and human resources. Budgets and means devoted to administering labor market policies are scarce in most countries, and moni- toring rarely brings extra resources. * Coordination problems Institutions matter-if there are split responsibili- ties for implementing active labor market policies, monitoring becomes dif- ficult. This is the case in France, where monitoring of active labor market policy falls to two different directorates in the Ministry of Labor, the center for adult vocational training (Association Nationale pour la Formation Professionelle des Adultes), and, in some cases, UNEDIC (unemployment insurance, run by social partners). All these institutions also have regional, departmental, and local agencies. Some progress toward better coordination, Evaluation and Monitoring of Active Labor Market Policies 1 423 especially in collecting data and making it accessible, has occurred, but the complex institutional structure is an obstacle to a comprehensive monitor- ing system. Absence of clear program goals. Weak monitoring systems and a lack of relevant indicators for a given employment measure impede policymaking. Although there are no ready solutions for fragmented institutional structures and weak overall labor market statistics, policymakers can at least identify gaps and problems as they develop monitoring systems. The Swedish Experience: Toward a Comprehensive Monitoring System Information and statistics on labor markets and programs have been part of the Swedish model since the early 1950s.6 The sharp rise in unemployment in the early 1990s reinforced information needs and prompted debates on labor market policy. Sweden has thus developed a comprehensive monitoring system that includes the following elements: * Budget and expense tracking. For each labor market office, data are pro- jected and collected on planned and actual expenditures and planned and actual participants for all labor market programs * Ongoing follow-up and feedback At the national level, continuous budget checks, monthly follow-up of results, annual surveys of firms and partici- pants, and staff surveys every 2 years are all part of the information system (AIS) run by the Labor Market Administration. * An auditing system. The regional office spends 2 days a year auditing each local office, and the central office spends 3 days every 2 years overseeing the activities of each regional branch. All labor market and AIS data are available online. This comprehensive monitoring system not only provides a good basis for evaluation but also is closely related to the decentralization that occurred during the 1990s. The monitoring system allows each placement officer to check budgets and calculate expenditures for a given unemployed person, and thus helps agen- cies allocate budgets and make decisions. Local offices, in turn, provide regular information and feedback to central offices, thereby improving the reactivity of labor market policies and budgets 424 I Labor and the Welfare State Comparative Monitoring at the European Level Comparative monitoring of labor market policies has developed following the recommendations of the European Commission. The goals are to identify Europe's best performers, overall as well as on a given employment goal, and facilitate information exchange on successful experience. The EU labor market information system, known as the European Employment Observatory,7 dates from 1982. This system began as an informal network of national correspondents in national ministries, but it has evolved toward more formal activities. Since 1996 the observatory has consisted of two networks (MISEP and SYSDEM) and an advisory group (RESEARCH). MISEP (Mutual Information System on Employment Policies) collects, summa- rizes, translates, and publishes information on labor market measures to support employment services and policymaking. A group of national correspondents, nominated by national ministries, and a secretariat meet twice a year SYSDEM (System of Documentation, Evaluation, and Monitoring of Employment Policies) was initially designed to provide information on labor markets and employment policies in member states, but it now reports on labor mark-et topics from a comparative perspective and provides information to the European Commission It consists of a network of correspondents from leading indepen- dent research institutions. All information gathered through the European Employment Observatory will soon be available on the Internet. This system provides good information on country performance and allows some comparative evaluations.8 However, it is still far from ideal because it offers little feedback on policy decisions and design and is far from the goal, advanced at the Luxembourg summit, of coordinating national employment policies. Nevertheless, the system enhances member states' awareness of the need for good follow-up of labor market policies. Microeconomic Evaluations The broadest goal of evaluation in labor market policy is to determine what is working and what is not. Micro evaluation addresses this question at the agency or individual level. Such evaluation requires two key pieces of information: * What effects does a given program have on thefifture prospects of program par- ticzpants? Variables include earnings and job status (employment versus unemployment) and type of job (short-term or regular contract, full time or part time). Evaluation and Monitoring of Active Labor Market Policies 1 425 Does the program affect nonlparticipants? Some programs may have adverse side effects For instance, if a policy subsidizes firms that hire long-term unemployed, some firms will reduce their hiring of short-term unemployed. A program is efficient if it has a positive impact on participants' situations and if potential negative effects on nonparticipants (and more generally on the labor market) are minimal. Some negative effects may be acceptable. in the example above, labor market authorities may prefer to help the long-term unemployed obtain jobs even if doing so implies more short-term unemployment Efficiency indicators thus need to link to a policy's goals. Governments developed microeconomic evaluations to measure efficiency and thus to choose among different programs (such as temporary public jobs versus subsidies to private sector jobs) and different schemes (for example, between classroom and on-the-job training) Follow-up indicators cannot pro- vide a good measure of efficiency, but microevaluations such as cost-benefit analyses can link outcomes to costs, facilitating comparison of various programs. Microeconomic evaluations have been the subject of continuing debates regarding experimental procedures. Understanding these debates is crucial to governments' choices despite their technical character because only evaluations grounded in reliable methodology can produce robust results. Methodologies used to evaluate active labor market policies fall into four main categories * Interviews Because microeconomic evaluation focuses on the impact of labor market policy on program participants or the firms hiring them, interviews can provide useful information. However, answers may be biased if employers or participants believe that results will be used to make policy decisions * Postprogram Surveys. Surveys follow the labor market status of participants after a program concludes at various points in time to evaluate differences in individuals' situations before and after participation. Evaluators cannot measure program impact on a simple before-and-after comparison basis: differences might result from factors that have nothing to do with the labor market policy, such as changes in the general labor market situation. Data on nonparticipants with comparable characteristics are therefore needed to measure program impact. * Random Assignment Experiments ("Experimental" Evaluation). These tech- niques resemble those used to evaluate a new medical treatment or drug Individuals who would like to participate in a program (and wvho fulfill the 426 I Labor and the Welfare State eligibility criteria) are randomly allocated to either a treatment/experi- mental group or to a control group. The individuals in the treatment/ experimental group benefit from the program, whereas those in the con- trol group do not. Evaluators collect outcome variables for the two groups at various points in time, with the difference in outcomes interpreted as the program's impact. Theoretically, the estimated impact is free of selec- tion bias because selection occurred on a purely random basis. Quast-experiments. Rather than relying on random assignment, here evalu- ators use general labor market information to construct a control group, matching its characteristics as closely as possible to those of program par- ticipants. The evaluator then gathers data on participants (such as from postprogram surveys) and nonparticipants. The program's estimated impact is the difference between the outcomes for participants and mem- bers of the control group. Evaluation frameworks using a constructed con- trol group are termed "quasi-experimental." Analysts have traditionally considered random assignment experiments the most reliable in evaluating the impact of labor market policy programs because experimental designs are typically free of selection bias and provide easy-to- understand results. This type of evaluation has been implemented mainly in the United States, which has performed some well-known studies of active labor market policies such as the Job Training Partnership Act.9 Quasi-experimental approaches were developed as an alternative to random assignment experiments. The main problem lies in correcting for selection bias because this involves econometric analysis, which is based on implicit assumptions about the bias. iO Cost-Benefit Analysis: Estimating Effects on Employment and Nonparticipants To develop a comprehensive picture of the effects of active labor market policies, governments must know more than simply the effect on participants. If policy- makers target policies at specific groups such as young people and long-term unemployed, they will exert negative effects on the labor market. The most important of these is termed "substitution displacement." For schemes that entail transfers of public funds to private firms (such as job subsidies), if job cre- ation might have occurred in any event, public funds do not contribute to any labor market goal: there is no employment effect. Windfall and substitution could explain 90 percent of the apparent employment effect for some wage Evaluation and Monitoring of Active Labor Market Policies 1 427 subsidies. 11 When active labor market policies are implemented on a large scale, the potential effects on nonparticipants cannot be neglected, though standard impact evaluation cannot identify those effects'2 because microeconomic stud- ies usually focus on a small sample. Cost-benefit analysis uses both microeconomic evaluation and financing data to relate the outcomes of employment programs to their costs. Studies show very different costs for the same efficiency level: for instance, in France, among pub- lic temporary employment programs targeted at young people, the cost of a pub- lic temporary employment program was half the cost of a comprehensive employment program, despite comparable effects on employment rates at the end of the 1980s. This type of information is very helpful to policymakers, although they must also take social considerations into account The most basic cost-benefit evaluations relate microeconomic results (effects on employment and wages) to unit costs. The main difficulty lies in the concept of unit cost, which can be understood as either the cost of the program over its whole duration or its average annual cost. On the one hand, the latter allows evaluators to compare different labor market policy measures directly and is also consistent with annual financial data. On the other hand, average annual costs do not account for heterogeneity of expenditures within a given program At a more rigorous level, cost-benefit analysis requires evaluators to estimate many different gains and losses An ideal cost-benefit analysis should estimate monetary gains and losses for participants (compared with nonparticipants), gains in relation to wages, losses in relation to forgone social benefits, gains in relation to reduced social expenditures, and higher tax revenues versus the costs of the program. To be comprehensive, such an analysis should also include direct and indirect effects on nonparticipants, because some individuals might have lost jobs because of the program, representing a cost to society. These effects are very difficult to estimate and integrate into the cost-benefit framework: most studies, therefore, remain partial and must be interpreted with caution. Given these difficulties European countries have produced very few exam- ples of cost-benefit analysis. This does not mean that knowledge of relative cost- benefit ratios is lacking: labor market agencies calculate annual unit costs, which they can relate to outcome indicators or the results of microevaluations. But the absence of standard cost-benefit ratios is more problematic in making interna- tional comparisons and identifying best practices.i3 The methodological challenges can be summarized as follows: To quantify the impact of an active labor market program on participants, evaluators need a dif- ferential measure (comparing participants' outcomes with those of a control 428 l Labor and the Welfare State group, either experimental or quasi-experimental) The simplest way to perform such a comparative evaluation is to use random assignment experiments, but evaluators need to consider experimental design and ethical questions carefully. Quasi-experimental evaluations have become reliable. Since technical aspects are crucial to the robustness of results, specialists should conduct such evaluations. Evaluations of impacts on participants should be complemented by studies esti- mating the impact on nonparticipants, as well as windfall effects Econometric studies provide the most robust estimates, but interviews can also provide inter- esting information and are easier to conduct Good microeconomic evaluation requires explicit attention to context as much as technical debate among experts European Experience with Microeconomic Evaluations Most microeconomic evaluations in Europe do not meet the foregoing standards Ethical considerations exclude most random assignment experiments,'4 and proper quasi-experiments are not widely used, although their number is grow- ing. Many studies compare the situation of participants before and after the pro- gram, or compare participants and nonparticipants with similar characteristics, but without constructing a control group.' 5 Several factors have curtailed the development of microeconomic evaluation. These include * A lack of monitoring rcsults in a lack of data. * An evaluation culture is absent in political and academic fields. * In some countries (especially the Nordic countries and Germany), active labor market policies were part of a political and social consensus, so dis- cussion of their efficiency and costs was limited until financial constraints and rising unemployment launched new debates e Compared with those in the United States, evaluation programs are smaller, although they are numerous. * In some countries, such as France, employment schemes have varied widely, making it difficult to conduct evaluations. In a 1998 survey I found four examples of random assignment experiments. Three evaluated the effects of job-search assistance for the unemployed, and one evaluated the impact of a training program. I summarize two here.i6 The first exemplifies good acceptance by program administrators, whereas the second illustrates hostility toward random assignment. In both cases, biases affected the experimental outcomes: Evaluation and Monitoring of Active Labor Market Policies 1 429 * Job-search assistance in Elskistuna, Sweden. 17 The evaluated program was a new experimental scheme offering intensified job-search assistance to the unemployed. The sample was composed of 400 participants unemployed for a minimum of 3 months and enrolled in a local employment service Individuals were randomly assigned to an experimental group receiving intensified service (216 unemployed) and a control group (194 unem- ployed). The experimental group received on average 7.5 hours of assis- tance for 3 months (consisting of interviews with placement officers), com- pared with an average of 1.5 hours for the control group Data collected included individuals' labor market status (short-term contracts versus reg- ular jobs or unemployment) and average monthly income gains at the end of the 3-month period. Results showed a clear positive effect from intensi- fied job-search assistance. 48 percent of the experimental group members were employed after 3 months versus 34 percent of the control group The difference was also significant for employment type: 92 percent of the employed in the experimental group had obtained regular employment versus 68 percent for the control group The data also showed a positive program impact on average monthly earnings. * Training program in Norway. 8 Unlike the previous example, the evaluated program was an existing training program Random assignment was restricted to 18 employment offices facing excess demand for services. Employment offices selected candidates on the basis of criteria normally used and then randomly assigned them to an experimental or a control group. One year after the program's end, evaluators asked experimental and control group members about their labor market situation (employ- ment or unemployment). The training program had no significant impact on the employment rate However, strong resistance among administrators resulted in selection bias at the eligibility-control step. i9 In one employ- ment office, 10 training places were available, and the administrators found only 10 unemployed among 80 eligible for the program The random assignment procedure thus did not work Most employment offices also provided the control group with comparable services, which caused a sub- stitution bias. Despite their weaknesses, these examples suggest interesting methodological lessons. First, they differ from U.S. examples in size: the samples are small, and the number of observation periods is limited Consequently, costs are also rela- tively low Second, they confirm the importance of cooperation from program administrators. In two of the four European experiments, the administration 430 I Labor and the Welfare State showed some hostility to random assignment. Experimental design seems more readily accepted in a new program that will be developed on a larger scale if out- comes are good, as in the Swedish experiment. Another study, evaluating personnel increases in the Swedish Public Employment Service, offers an interesting European example of cost-benefit analysis.20 In 1987 the Swedish government gave the National Labor Market Board extra funds to add 250 placement officers and counselors to the Public Employment Service, boosting their numbers by about 5 percent. The first step in assessing the efficiency of this staff increase was an impact evaluation. Estimates of impact were based on before-and-after comparisons of three employment offices where the staff grew by 15 percent to 25 percent (program offices), and three offices where the number of staff either remained unchanged or expanded only slightly (comparison offices). The three program offices were matched with comparison offices serving a labor market with similar characteris- tics. Negotiations between the National Labor Market Board, the Ministry of Labor, and evaluators were used to assign offices to the two groups. The study focused on two types of impact: the duration of unemployment for job-seeking clients and the duration of job vacancies. The results showed no effect from expanded staff on the average length of unemployment but revealed significant effects on the duration of vacancies, which dropped by an average of about 1 day. Evaluators also estimated the real costs of resources such as wages, training, computers, and incidental material associated with the rise in person- nel. Evaluators then compared those costs with the market value of the output (MVO) arising from the accelerated vacancy filling. That value was measured by firms' willingness to pay for more goods and services, estimated on the basis of average wage costs. The difference between benefits (MVO) and costs was found to be negative. With no effect on the average duration of unemployment, the staff increase also failed to produce effects on public finance (if duration of unem- ployment falls, then costs for public assistance also drop, and tax payments rise). The staff increase clearly failed the cost-benefit test because it did not cover its costs to the government. The study suffered from important methodological weaknesses, howevcr, because the samplc sizc was small and the process of select- ing program offices and comparison offices was likely biased. This test was also performed in a period of economic expansion, when employment services seem to have less influence on labor markets The most interesting European microeconomic evaluations are based on panel data, which France systematically developed in the 1980s and used during the 1990s. Panels provide information about participants' backgrounds before Evaluation and Monitoring of Active Labor Market Policies 1 431 they begin a labor market program and rely on a quasi-experimental design to evaluate the program's net effect on employment rate and wages. The labor mar- ket exhibits some duration and path-dependence (hysteresis) effects: the proba- bility of getting a job usually decreases with the length of unemployment. Whatever the interpretation of this phenomenon (depreciation of human capi- tal, signaling, stigmatization), longitudinal data provide some information about the impact of programs on this path-dependency effect Panel data generally contribute to a better understanding of how employ- ment policies affect labor market dynamics. For example, the French CEREQ (Research Center on Employment and Qualifications) panel studied the inte- gration of young people into the labor market in relation to their educational level and qualifications. The French Ministry of Labor also developed other panels to evaluate labor market policies or drew them from more general labor market surveys. French experience, however, highlights some drawbacks. First, the reliability of results depends crucially on the period of observation, which can be quite short for a standard impact evaluation but must be longcr when the aim is to identify trajectories. Evaluators also need to consider attrition rates, which usu- ally rise with length of observation. Second, results can be difficult to interpret. Impact evaluations using longitudinal data take human capital and job-search models as a theoretical background, but they do not interpret results in accor- dance with those frameworks. Nevertheless, longitudinal data and the debates surrounding interpretation of results contribute to a better understanding of labor markets and policies. The French experience with youth programs illustrates the need for system- atic evaluation of active labor market policies. Until the end of the 1980s the only available information was based on interviews and postprogram survey data. Most studies calculated the employment rate of participants 3 months after the program. With that approach, the most efficient scheme (contrat d'adaptat:on, which provides on-the-job training) achieved an 80 percent employment rate, while the less-efficient scheme (the TUC-public temporary employment scheme) had a 30 percent employment rate. During the 1990s evaluators developed rigorous quasi-experimental designs using panel data. The results differ markedly from prior conclusions. If the impact on employment rates is the difference between participants and a matched control group, the only efficient program is the one showing a positive net impact. Some youth programs had a negative impact-participants had lower employment rates 3 months after the program than comparable young people. 432 1 Labor and the Welfare State This result may stem from the impact of program participation on job-search intensity (participants reduced their job search) and stigma effects (employers are reluctant to hire young people who participate in an employment program because they might have problems). The efficiency of a program should not be judged solely on the basis of inter- views and survey results In this example, a program's efficiency ranking changes if evaluators use rigorous methods. The evaluators were well aware that their pre- vious results were fragile, but the new results strongly influenced the govern- ment's assessment of youth programs They also clarified the need for regular rigorous evaluation What Do the European Evaluations Show? Empirical micro studies often provide ambivalent and even contradictory results. Nonetheless, recent surveys and analysis highlight converging outcomes 21 Empirical results from training and retraining are disappointing, despite their good reputation. Many studies find either very small positive effects or no effect at all, and some evaluations even point to a negative impact. Training is also very expensive. These results could reflect the fact that any positive effects of training on productivity and earnings occur only after a long delay. Studies cannot mea- sure such long-term effects because evaluators generally do not collect data over a long period (except in the case of some panel data), and long-term impacts are difficult to estimate because they require controlling for many variables. Temporary public jobs may exert some positive effects. In Germany, especially over the past decade in the territory of the former German Democratic Republic, public job creation is clearly effective in raising participants' probability of future employment. The impact is higher if temporary public jobs are combined with training. These programs seem to give better results for hard-to-place people. For the most qualified groups of unemployed, temporary public jobs may have a stig- matizing effect, which can reduce future earnings A French study that compared wages for participants in a temporary public job program with those of a matched control group found a negative impact on wages, at least for young pcoplc with some education. Job-search assistance through placement services and intensive counseling are the least-contested active labor market policy. The impact of job-search assis- tance on the probability of finding a job is always positive. Evaluators found one of the most impressive effects under the British Work Trials Initiative, which combined traditional job-search assistance with incentives for employers to take Evaluation and Monitoring of Active Labor Market Policies 1 433 on people who had been unemployed for more than 6 months During a 3-week trial period, participants continued to receive benefits plus compensation for work expenses. An evaluation showed increases in employment rates of 34-40 percent 6 months after the program ended.22 Several European countries have developed programs to encourage the unemployed to become self-employed, using benefits as a subsidy. Evaluations report good results where implementation is careful and enterprise projects are preassessed Recent studies suggest that "activation" of unemployment bene- fits-making them more work-oriented-yields positive results. Evaluation contributes to institutional reform. Several European countries have at least partially privatized public employment services. Such reforms appear to increase reemployment probability, even for hard-to-place unem- ployed. In Germany, a quasi-experimental evaluation compared long-term unemployed and very hard-to-place people (elderly, disabled, ex-addicts) reinte- grated through commercial or semipublic temporary work agencies versus pub- lic employment services (control group) The reintegration rate of the experi- mental group was significantly higher (27.4 percent versus 14.2 percent) Aggregate Impact and Macroeconomic Evaluations Aggregate impact studies aim to integrate all the direct and indirect effects of labor market policies on unemployment levels, employment, and wage bargain- ing Such evaluations, which are a distinct part of the European approach, do not require specific surveys or data collection because they generally use administra- tive information (financial data, number of participants) and general labor mar- ket statistics (employment, unemployment, wages). Despite some methodological limitations, European aggregate impact studies have produced interesting results, examples of which follow.23 Active labor market policies moderate wages. Active labor market programs tend to boost the employability of the unemployed, thus increasing competition in the labor market. This has a wage moderation effect. Only in Sweden did a study find a positive effect on wages (wage pressure effect) 24 Traditional Swedish labor market policy differs from that of other European countries in that expenditures are higher and programs are more job oriented (involving training and retraining and mobility allowances), with little hard-to-place targeting.25 In this context, active labor market policies will not lead to a competition effect and may well encourage workers to ask for wage increases. 434 1 Labor and the Welfare State Active labor market policies support labor market transitions. Most empirical studies based on a matching model find that active labor market policies have a positive impact on outflows from unemployment and improve the labor market's overall functioning. Nevertheless, such policies can produce negative effects. For example, several studies suggest that although targeted measures boost outflow from unemployment for the targeted group, outflow among other groups tends to decline. Policymakers should consider targeting cautiously and conduct com- prehensive evaluation studies to understand its effects. Most studies find that active labor market policies reduce unemployment, but the effects of specific programs vary. For example, subsidies create high displace- ment, substitution, and deadweight as well as exert a minor impact on aggregate employment levels. Such policies can nonetheless be efficient if they are targeted to the very long-term unemployed or to depressed areas such as the former German Democratic Republic. Temporary public jobs also risk displacement and substitution, but those effects are lower than for job subsidies if the temporary jobs are limited to nonmarket activities. Most studies find better results for train- ing, especially if it is short, targeted, and market oriented. Long-term impact may differ from short-term impact. Long-term effects will reflect the impact of wage bargaining, external competitiveness (including exchange rate effects), and budgets. Active labor market financing (through taxes, reductions in other public expenditures, and deficits) also influences macroeco- nomic performance. Overall, macroeconomic evaluations of labor market policies have two main advantages. First, data requirements are manageable because they rely on admin- istrative data (expenditures, number of participants) and general labor market statistics (unemployment, employment, wages). More sophisticated evaluation methods require information on outflow from unemployment toward employ- ment and inactivity. Second, although such studies might at first seem compli- cated to nonspecialists, estimation techniques are far simpler than those required for microeconomic quasi-experimental studies, and methodological problems are well known. The main drawbacks relate to the level of evaluation. First, aggregate impact evaluation does not reflect whether a country has very low employment expendi- tures and very few program participants. Furthermore, an evaluation of the over- all effects of reduced unemployment may mask negative outcomes for some indi- viduals, who may suffer from lower wages or bad working conditions, for instance. That is why macro and micro approaches are complementary, especially Evaluation and Monitoring of Active Labor Market Policies 435 when micro evaluations are nonscientific. In that case, aggregate impact analvsis will give some indication of the direct and indirect effects that are reducing the net impact of active labor market policies (windfall, substitution, deadweight). Simple surveys or interviews will help evaluators understand the consequences of program participation for individuals. Policy Recommendations Diversity is an important feature of European experience in monitoring and evaluating active labor market policies: there is no single model. European coun- tries are still developing approaches to monitoring and evaluation, scientific as well as nonscientific. This experience has relevance for East Asian countries, which might benefit from the following lessons learned in Europe: * Monitoring should be developed at a very early stage, when new polhcies are being implemented. This calls, first, for collecting basic data on programs and, second, for giving program administrators basic training in the impor- tance and methods of using data to improve program implementation. * Evaluation is necessary ifgovernments want to make good use of public funds. Policymakers can use many techniques to evaluate active labor market poli- cies It is naturally better to use scientific methods, but surveys or inter- views can provide interesting and useful information as long as they are not misinterpreted. At the microeconomic level, it is impossible to mea- sure the net impact of a program without a control group. * To measure net efficiency, evaluators miust use random assignment or quasi- experimental methodologies to estimate individual impact Random assignment is easier, but this procedure works best for small samples or the cost of eval- uation may be very high. Using random assignment to evaluate new pro- grams lowers resistance among administrators and participants, and it helps policymakers decide if a program is worth continuing and how to improve the program. For ongoing schemes, it is better to use quasi-experimental evaluation, although these methodologies are more technically complex and thus require expert assistance. The main limitations are likely to be data problems, underscoring the importance of data collection, especially some longitudinal (panel) data for future evaluation This requires regular sur- veys of a sample of participants and unemployed people. 436 Labor and the Welfare State * Macroecononic and aggregate impact evaluations are a uiseful complement to othzerfirms of evaluation. Macroeconomic evaluations are easy to perform with relatively minimal levels of expertise and basic data. Such evaluations can also allow international comparisons, especially when countries have similar economic and labor market characteristics East Asian nations might find such comparisons particularly useful. Notes I See Peter Robinson, "Active Labour Market Policies A Case of Evidence-Based Policy Making " O.%ford Revien of Economic Policy 16, no 1 (2000) 13-26, for more on this aspect 2 1 use a classic definition of "active labor market policies" based on Organization for Economic Cooperation and Development categories Active measures aim directly to return people to work (job-creation schemes, hiriog subsidies, job-search assistance) or maintain their employability (training) 3 P Auer and T Kruppe, "Monitoring of Labor Market Policy in EU Member States," in International Handbook of Labor Market Policy and Evaluation, ed G Schmid, J O'Reilley, and K Schomann (Cambndge, England Edward Elgar, 1998), pp 899-922 4 See, for instance, the French 5-year-emplo ment law of 1994 5 MISEP (Mutual Information Systems on Employment Policies), see section on Comparative Monitoring at the European level 6 The high degree of reactivity of Swedish active labor market policy has been studied in depth by Ohlsson He shows that the number of participants in temporary jobs in the public and nonmar- ket sector is closely related to economic cycles, following a countercvclical trend Countercclical responsiveness is made possible by good follow-up indicators of business cycles and labor market policies See H Ohlsson, "Job Creation Measures as Activist Fiscal Policy An Empirical Analysis of Policy Reaction Behavior," European j_ournal of Political Economny 8 (1992) 269-80, and H Ohlsson, "Lags in the Effects of Labor Market Policy An Empirical Analysis of Job Creation M{easures," Applied Economics 25 (1993) 343-48 7 Data from the European Employment Observatory, available at 8 See, for instance, a German study of policies to integrate youth into the labor market L Tronti, Benclhmarking Employment Performance and Labor Market Policies (European Commission, Employment Observatory, RESEARCH network, April 1998), pp 63-68, and Auer and Kruppe, op cit , on labor market training comparative monitoring 9 For details on the US experiments, see A Blorklund and H Regner, "Experimental Evaluation of European Labor Market Policy," in Schmid et al, op cit, pp 89-111, H Regner, "Choosing Among Alternative Non-Experimental Methods for Estimating the Impact of Training New Swedish Guidance," Working Paper No 8 (Swedish Institute for Social Research, Stockholm University, 1995) U S Department of Labor, "What's Working (and What's Not)" (draft, 1995) 10 Good methodological surveys can be found in R Fay, "Enhancing the Effectiveness of Active Labor Market Policies Evidence from Program Evaluation in OECD Countries," Labor and Market Social Policy Occasional Papers 18 (1996), Schmid et al , op cit , and J Martin, "What XVorks among Active Labor Market Policies. Evidence from OECD Countries' Experiences," OECD Econtomic Studies 30 (1999) 79-113 Evaluation and Monitoring of Active Labor Market Policies 1 437 11 J Gauti6, "Les evaluations d'ordre micro-economique impact sur les b6neficiaires et effets directs sur 1'emploi," in Lespolitnques de lemplo, cn Europe et aux Etats-Unis, ed J C Barbier, J Gautic (Cahiers du CEE Presses Universitaires de France, 1998) 12 They are also an important source of bias (the hypothesis that the control group is not affected by the program is not true) Many European authors discuss these problems See Gaune, op cit, L Beilmann and R Jackman, "Aggregate Impact Analysis," in Schmid et al , op cit, pp 143-60, and L Calmfors, "Does Active Labor Market Policy Increase Employment' Theoretical Considerations and Some Empirical Evidence from Sweden," seminar paper for the University of Stockholm, Institute for International Economic Studies, March 1995 13 The Etiropean Commission is encouraging development of cost-benefit analysis to support labor market policy analysis and coordination 14 I fiiund four in the European Union. See C Erhel, "L'Evaluation Macroeconomique des poli- tiques de 1'emploi," in Les polttiques de lemploz en Europe et aux Etats-U1tis (Cahiers du CEE Presses Universitaires de France, 1998), pp 257-76 15 In a survey of 95 evaluations undertaken in the Netherlands, De Koning found only 4 that used a control group See J De Koning, "T1raining for the Unemployed The Dutch Experience," paper for the TRANSLAM Project, Netherlands Economic Institute, Rotterdam, October 1997, p 27 16 The two other experiments are described in C Gorter and G Kalb, "Estimating the Effect of Counselling and Monitoring the Unemployed Using a Job Search Model," Journal of Human Resources 31, no 3 (1996) 590-610, and M White and J Lakey, The Restart Effect Does Active Labor Market Policy Reduce Unemplo)yment? (London Policy Studies Institute, 1992) 17 L Delander, "Studier kring den arbetsformedlande verksamhetcn," Statens OJf ntllga Utredningar 60 (1978) 183-248 18 0 Raaum and H Tort, Evaluernng av AMO-kurs Sysselsat:ntngseffekter og seleksjon till kurs, report 72 (Oslo SNF, 1993) 19 The estimated impact of training was not significant or very weak 20 This paragraph is based on L Delander and H Niklasson, "Cost Benefit Analysis," in Schmid et al, op cit, pp 183-87 The original reference is l Behrenz, Effekt- och effektivitetsanalys av 1987 drs personalJbrstarknmng till arbetsfi)rmedlingen (Lund, Sweden Department of Economics, 1993) 21 For recent surveys, see G Schmid, "Activating Labor Market Policy 'Flexicurity' through Transitional Labour Markets," paper prepared for the Foundation Saint Gobain Conference in Pans, November 8-9, 2000, and Robinson (2000) 22 M White, S Lissenburgh, and A Bryson, The Impact of PubhcJob Placing Programmes (London Policy Studies Institute, 1997) 23 For surveys, see L 13ellmann and R Jackman, "Aggregate Impact Analysis," in Schmid et al, op cit, pp 143-60, L Bellmann and R Jackman, "The Impact of Labor Market Policy on Wages, Employment, and Labor Market Mismatch," in Schmid et al , op cit , pp 725-43, and Erhel, op cit 24 Such a positive impact is not systematically found in the Swedish studies country results are contradictory 25 This description of the Swedish model holds for the 1970s and the first half of the 1980s, but the sharp rise in unemployment at the end of the 1980s weakened these Swedish trends CHAPTER 25 Employment Policies: The Case of Spain by Eduardo Gonzdlez Biedma ABSTRACT: Europe has faced decades of stubborn structural unemploy- ment since the oil crisis of 1973, and the European Union as well as every member state has proclaimed efforts to create jobs a policy priority. This chapter describes European efforts to combat unemployment, using Spain as a case study, and also traces unemployment in East Asia in the wake of the economic crisis. The chapter argues that crises can stimulate economies to shed inefficient industries and overly rigid labor regulations while creat- ing more viable businesses and jobs in the long run. The oil shock of 1973 and the Asian financial crisis of 1997-98, although very different in many respects, both spawned unusually high levels of unemployment accompanied by inflation and stagnation. The 1973 shock led to an unfamiliar phenomenon in Europe-lasting structural unemployment-whereas in East Asia traditionally low unemployment rates rose significantly with the crisis, although they were somewhat disguised by underemployment. The European Union (EU)-as well as every member country-has pro- claimed the struggle against unemployment its foremost policy goal. The best way to create employment, analysts now maintain, is to establish the right macro- economic environment, deregulate the economy, and foster flexibility by relax- ing labor regulations. Governments have also turned to active labor market poli- cies to bolster human capital and adjust labor to the needs of business, thereby unleashing the working potential of society and minimizing pockets of structural unemployment. This broad approach harks back to the Treaty of Rome, the EU's founding charter, but what is most significant is that Europe's employment goals depend on the efficient functioning of the common market. 439 440 1 Labor and the Welfare State This chapter examines remedies to unemployment pursued in Europe since the oil shock, highlighting the evolution of such policies in Spain and under- scoring those that have proved most successful. The chapter also examines recent unemployment in East Asia. The chapter argues that crises can create the condi- tions for renewed economic growth and employment if governments pursue flex- ible labor laws and careful investment rather than over-regulation and direct pub- lic subsidies Unemployment in Europe: A Pervasive New Phenomenon During the 1960s unemployment rates in Western Europe were insignificant The United Kingdom, the Federal Republic of Germany, France, and Switzerland welcomed large waves of immigrant labor, some from former European colonies and others from southern Europe. Policymakers maintained confidence in the "healing effects" of the European market and its freedom of movement for capi- tal, services, and workers. This full-employment dream lasted until the 1973 oil shock, which brought an abrupt halt to the era of smooth and stable economic growth. Concerted action by members of the Organization of Petroleum Exporting Countries, trig- gered by the Palestinian-Israeli conflict, produced sudden shortages in oil sup- plies and a fourfold increase in the price of crude oil.' Because energy was an essential resource for both work and capital, less energy meant lower productiv- ity.2 Europe had no mechanisms for addressing this disruption. Workers pre- ferred massive dismissals and plant closures to lower salaries, whereas capitalists did not accept the need to boost productivity, hoping instead that investments in productive capacity would magically restore prosperity.3 The crisis led to a rapid transformation in international trade and financial markets, exacerbated by growing international competition, provoking further crises in business cycles. These shocks helped expose long-term inefficiencies that European economies could no longer bear Employment and Social Policy in Europe after 1973 The 1973 crisis and a second crisis triggered by another oil shock in 1980-81 dra- matically altered views of full employment. Policymakers no longer viewed it as a self-adjusting phenomenon but as a major social issue that called for an explicit Employment Policies The Case of Spain 1 441 policy response. Public authorities described unemployment as the most glaring problem facing the European Community, which issued formal declarations, bills, projects, and blueprints designed to address it. The Social Action Program of 1974 represented the first formal step toward "the realization of full and better employment" in the European Community. The program included the following major goals: * Improving hiring services * Providing professional training, including creating a European center for professional training * Fostering equal opportunities for men and women * Promoting free international movement of workers * Establishing fair policies for dealing with workers from developing countries * Aiding people who have special difficulty obtaining employment, includ- ing youth, older people, and people with disabilities * Humanizing work and living conditions * Improving social security, including for those ineligible for regular programs * Promoting worker participation in company decisions * Protecting workers during company mergers and takeovers * Reducing work time by establishing a maximum workweek of 40 hours and mandating 4 weeks of paid vacation * Coordinating domestic regulation of layoffs The program sought to avoid or contain layoffs provoked by the crisis and suggested social protection for dismissed workers or those employed by compa- nies facing economic difficulty. Despite these provisions, the document ignored the core problem: a lack of jobs. The Second Social Action Program, triggered by the 1980-81 oil shock, fol- lowed a similar path, as did ensuing documents For example, the 1989 European Community Charter of Workers' Fundamental Social Rights established an employment "observatory," funded a program to create jobs for specific groups, and improved information exchange on job demand and supply. Those policies did not prove effective, unemployment remained stuck in the 9-10 percent range throughout the European Community. The 1994 Jacques Delors white paper constituted a turning point in employ- ment policy, stating categorically that employment was the absolute policy prior- ity The paper focused on two main remedies active employment policies- specifically professional education and special training for youth as well as 442 I Labor and the Welfare State enhanced hiring services-and efforts to foster economic growth to generate employment. This document underscored the need to stimulate small and medium-size businesses, which would in turn create jobs.4 Earlier programs and papers had assumed that employment policy was some- how unrelated to economic policy, and that social measures and labor legislation would achieve employment goals The white paper stressed the importance of maintaining social rights and creating high-quality jobs, but it also recognized that achieving full employment might well require restricting funding for social policies and weakening some labor standards. The Essen European Council of 1994 underscored the goal of "good" employment as well as full employment, implying that this might require pro- viding unemployed people with social benefits rather than simply creating poor- quality jobs. To expand employment, the council called for reducing indirect labor costs-namely, social security taxes.5 The European Council of 1997 advanced the important idea of mobilizing new technologies and the European network to create new industries and small and medium-size businesses, and thus long-term jobs, with US$10 billion in funding from the European Investment Bank over 3 years. Interestingly, before European Council authorities had reached these conclu- sions, European trade unions and employers' associations had signed agreements in 1986 and 1992 that contained clear policy prescriptions for creating employ- ment. These documents called for creating jobs by rapidly reducing interest rates, which would follow from appropriate macroeconomic policies. The Case of Spain Spain exemplifies the move toward macroeconomic stability and flexible labor laws as the cornerstone for economic growth and job creation. In Spain the 1973 oil shock coincided with a change in political regime: dictator General Francisco Franco died in 1975, and the country passed a new constitution in 1978, which in turn brought political troubles, legal reforms, and economic transformation. Spain was emerging from the economic and political isolation of the 1960s, although it did not yet belong to the European Community. Spanish unemployment rates have traditionally been higher than the European Community average, but after the oil crisis Spain began to experience unprece- dented unemployment stemming from sudden inflation, trade imbalances, and an Employment Policies The Case of Spain 1 443 obsolete industrial fabric based mostly on state-owned enterprise. Spain's legal and economic framework, encumbered with governmental controls, proved unprepared for open economic competition Economic indicators fell even more rapidly than in other European economies: inflation peaked at 24 percent in 1978 and unemployment rose as high as 22 percent. These problems have persisted, although unemployment rates have since dropped significantly. The government responded to the crisis by softening regulations affecting dismissals, worker mobility, payroll taxes, and compensation. These reforms par- alleled "emergency labor laws" throughout Europe. Unions sought to tighten employment regulations, including unemployment benefits, to minimize the impact of the crisis on workers. However, the Social Democratic Party, which ruled from 1982 to 1996, seldom met those demands, instead deregulating some aspects of employment law and stressing the role of collective bargaining. For example, the Estatuto de los Trabajadores, the main employment statute, allowed employment contracts to become more flexible. The government faced two general strikes and a falling out with its traditional partner, the General Worker's Union (UGT), which was particularly angry about statutes that allowed the creation of companies that provide temporary workers, as well as apprenticeship contracts that offer low-paid jobs. Ironically, unions sought to keep regulations enacted during the dictatorship while adding collective bargaining and the right to strike, which Franco had forbidden as too demanding for an aging and inefficient economic fabric. The newly revamped policies coincided with a new focus on active market policies, as the government spent considerable funds on professional education and subsidized full-time employment and new jobs. These measures enjoyed only modest success. Economic policy, meanwhile, focused on opening the econ- omy to international financial markets and trade, made easier as well as necessary after Spain joined the European Community. Most specialists agreed that fundamental changes in employment regulations were essential to improve labor market participation rates and fight unemploy- ment. However, pressure from unions and many progressive politicians had sapped the will to pursue deeper reforms. The result was that inflation, the colos- sal and inefficient public sector, budget deficits, high interest rates and taxes, an outdated and inadequate infrastructure, and weak entrepreneurship continued to present major hurdles to investment and the creation of new jobs. Thus, despite economic growth from 1986 to 1992, unemployment was surprisingly persistent, topping out at 22 percent in 1991 444 1 Labor and the Welfare State Increasing the Dynamism of the Economy Beginning in 1996, the conservative Popular Party government focused on fur- ther opening markets to competition and reducing the government role in the economy through privatization.6 Those measures, plus restrictive fiscal policies entailing "zero deficits" in public accounts, generated annual economic growth of 3.5 percent to 4 percent as well as sharp declines in unemployment. The legislature's Employment Program 2000 now alms to improve the busi- ness environment by easing taxes and interest rates, especially for service indus- tries (tourism is Spain's largest source of income) and information technology firms. The plan also endeavors to bolster research and development in informa- tion technology. Employment Program 2000 has further eased regulations on part-time jobs and calls for retraining workers to fill companies' needs Other aspects of the plan are designed to enable the country to assimilate immigrants and to reconcile family life and work by easing pressure on women's jobs. This program represents the state of the art in employment policy. ensuring a healthy economy through balanced public accounts, low interest rates, public investment oriented to economic growth, reduced direct public intervention in the economy, and changes in employment regulations that pose hurdles to invest- ment. However, the program has yet to fully address Spain's still thin safety net and pension system. East Asian Experience with Unemployment Employment challenges in East Asia are complex, given the region's diverse labor markets. Some 70 percent of Indonesians work in the informal sector, while 40 percent to 50 percent of the Thai labor force works in agriculture. Meanwhile 80 percent of the Republic of Korea's labor force is urban, segre- gated among firms of different size. Malaysia counts 1.7 million foreign work- ers in a labor force of 8 7 million.7 The East Asian crisis of 1997-98 stemmed mainly from a weakly regulated financial sector and an economy largely financed by short-term debt Yet the cri- sis bore a striking similarity to that in Europe in 1973, in that external factors sharply interrupted a period of economic growth and prosperity and created unexpected and unfamiliar social problems, particularly unemployment. For the five most affected East Asian countries, the crisis led to a sharp fall in demand, accompanied by price changes and dampened public spending 8 The result was a stunning decline in output: 15 percent in Indonesia, 7 percent in Thailand and Employment Policies The Case of Spain 445 Korea, 5 percent in Malaysia, and 0.5 percent in the Philippines The more industrialized the country, the tougher the consequences of the crisis The impacts on employment and wages were dramatic. Indonesia was the most affected country, some 5 million workers shifted from the formal to the informal sector, and salaries declined by 34 percent In Thailand, unemployment rose from 2 3 percent to 5.4 percent, and salaries fell by 6 percent, which eco- nomic recovery did not entirely redress. In Korea, unemployment reached a peak of 8.7 percent from 2.5 percent, and many people stopped looking for work. However, recovery was equally dramatic, and by October 1999 unemployment had declined to 4 6 percent. Malaysia and the Philippines were the least affected, Malaysia because it shifted job losses to immigrants, who accounted for as much as 20 percent of the work force Salaries declined by 12.5 percent, then partly recovered during the first half of 1999. In the Philippines, the unemployment rate of 8 percent to 9 percent changed little. Meanwhile work force participation rates changed little in Thailand and in the Philippines, in Malaysia and Korea they fell; and in Indonesia they grew because underemployed women entered the labor market to offset the loss of household income. Reduced employment opportunities obliged people to accept part-time and low-paid jobs,9 a phenomenon most pronounced in rural areas, which had the flexibility to absorb new workers New Approaches to Unemployment The three classic approaches to unemployment include boosting demand through public investment in infrastructure and community works, supporting unemployed people through unemployment benefits, and pursuing active labor market policies Europe's Employment Plan 2000 focuses on encouraging a more flexible work force through employment legislation, stimulating the economy by investing in information technology, and fostering the economic and political conditions for growth These approaches, pursued by member countries, have generally proved successful. Analysts have discredited public works as a way to cut unemployment, because that approach entails serious public debt and usually causes a rise in interest rates, taxes, or both, whereas the ratio of jobs created to public expenditures is relatively low Analysts view excessive public intervention in the economy as ineffective and unsustainable, it fails to create durable jobs and makes employment overly 446 I Labor and the Welfare State dependent on public sector initiative and management. Furthermore, although urban and rural public works programs offering low salaries may be helpful to very poor people, they do not attract blue-collar workers laid off during a crisis 10 This means that governments should rely on public works only when neces- sary and should apply strict measures of efficiency. However, Europe has suc- cumbed to the temptation to create artificial jobs. Such programs, which tend to require little capital and use many workers, have limited multiplier effects and foster continuous dependence on public funds, creating expensive pockets of inefficiency that are very hard to remove. Indonesia, the Philippines, and Thailand have experience along these lines. Assistance to the Unemployed Assisting the unemployed may seem essential, but avoiding fraud and misman- agement are difficult. Unemployment insurance tends to exclude people most in need: those in the informal sector and agriculture, as well as the self- employed and youth. To be eligible for benefits, people usually must have con- tributed to social security, but the people most acutely affected by a crisis tend to be those in the informal sector and those who have never held a job. Some people will opt to combine unemployment benefits with participation in the informal sector, so unemployment benefits could actually subsidize, and even promote and perpetuate, a form of labor relations that assumes a worker has other funds. To address these problems, unemployment benefits need to be geared to groups that truly need assistance, and special care must be taken to avoid fraud. Flexible unemployment benefit programs can make more sense than systems based on rigid rules. In East Asia, Korea alone now maintains a reliable unemployment insurance system. For other countries, this appears to be a good moment to establish flexible assistance schemes rather than formal insurance programs. Active Labor Market Policies Policies such as wage subsidies, professional training, job placement services, encouragement of jobs for youth and aged people, and support for self- employment, can be viewed as complements to other policies."1 In Europe, they are essentially considered social measures to help disadvantaged people and a vehicle for creating human capital. However, they are not effective in times of cri- sis when jobs are simply unavailable. The experience of Eastern European coun- tries shows that a well-educated population does not automatically bring low Employment Policies The Case of Spain 1 447 unemployment rates; overqualified workers who do not meet the needs of domes- tic businesses may have to emigrate in search of work. Organization for Economic Cooperation and Development (OECD) experience suggests that active labor market programs "tend to be small (on average their expenditures come to half of those spent for income support to the unemployed), and they entail high dead- weight loss. For example, for every five wage subsidies paid to employers, only one fills in a genuine vacancy-the rest are used by employers for jobs that they were already willing to fund."''2 The same comments apply to job placement agencies: they can do very little when the problem is a lack of jobs. Active labor market poli- cies thus work best in meeting real and routine needs to enhance the quality of the work force rather than in responding to crisis. The classic approaches to unemployment are necessary and effective but insufficient. Recent European experience underscores the importance of flexible employment systems, active efforts to attract new technologies, and efforts to enlarge markets and involve labor and employers in creating employment. Deregulating Aspects of Employment Regulations Deregulating aspects of employment regulations has proved successful in addressing unemployment, particularly during crises. European countries have sought to protect employee compensation and provide generous unemployment benefits while allowing layoffs, flexible forms of contracting, mergers, and nego- tiations on new working conditions. Without such measures, crises could have been far worse Governments essentially aimed at minimizing bankruptcies rather than blindly protecting worker expectations that proved unaffordable. Labor often demands stricter employment regulations to defend workers against layoffs. However, making employment regulations more rigid can create more unemployment and force many workers to shift to the informal sector. Concessions on working conditions, lower payroll taxes, and easing the hiring bf part-time or temporary workers may be efficient short-run solutions in fighting unemployment. 13 The former ideal of a full-time, lifetime job for everyone seems to be fading. The requirements of open economies, the characteristics of new kinds of busi- ness, and the explosion of the service economy all point to higher worker turnover and more temporary and part-time jobs. These should not be consid- ered underemployment. The new jobs will ensure higher salaries and some sta- bility in family income, even though they do not involve a permanent job for life. Employment regulations need to ease the growth of this new economy and meet its requirements insofar as they point in generally positive directions. 448 I Labor and the Welfare State Functional and Geographical Mobility Mobile workers can take advantage of employment opportunities. Government investments in education and training are essential in creating functional mobility. Open Relationships among Management, Unions, and Government Efforts to repress unions undermine basic human rights, are incompatible with modern democratic societies, and create unnecessary social tension and unrest. However, overregulation of labor markets can make the situation worse. Govern- ments should focus on bringing workers and employers together and balancing the interests of labor and management in specific industries.'4 Market Specialization and Enlargement Crises often produce at least one positive result. they can clean up an economy. Extinction of inefficient industries can exert an effect similar to that of a fire in an old forest the scene is nasty the next day, but in the long run the crisis allows stronger, greener, and more fruitful industries to grow. Crises can also help countries discover their areas of comparative advantage with the help of pres- sures and opportunities in the global economy, re-creating more stable and effi- cient employment. In Spain the oil shock prevented the government from further subsidizing coal and steel industries, which had experienced growing losses. The shutdown of several mines and factories forced their communities to live a tragedy. A few years later, however, most of those communities have shifted to the tertiary sec- tor or other types of manufacturing at no cost to the public budget, and they are now making much more money than before. The crisis clearly proved beneficial in the long run, although this was hard for people caught in the throes of tur- moil to realize. New Technologies and the Information Society The European Union has focused its latest employment plans on developing jobs in the information and services sectors. The role of government here is delicate: it must continue to provide essential assistance to people in need while pursuing active labor market policies that help spawn new businesses. Crises in Europe helped the economy shift strategically away from old industries that were no longer viable toward more up-to-date activities.1s Employment Policies The Case of Spain 1 449 Conclusion: Crises Can Foster Economic Renewal An unexpected conclusion is that crises are not entirely detrimental to employ- ment and economic growth. European experience highlights how crises can help economic systems mature Governments that allow economic forces to generate new ideas and resources can deal with crises and the resulting unemployment better and reasonably quickly The best option is always to create the macroeco- nomic conditions for growth Above all, governments should avoid overprotec- tive measures such as nationalization of companies and unnecessary or costly public subsidies whose sole purpose is to create new jobs, those actions can cause long-term misbehavior in the economy. Governments must alleviate the plight of unemployed people, basing benefits not solely on former contributions to social security but on real needs Systems should not discourage job search or subsidize the informal sector or under- employment. Ironically, flexible labor legislation is frequently effective in main- taining employment and protecting existing workers because it helps create new )obs and heal companies in trouble. Active labor market policies are useful in maintaining an economy's competitiveness, a key condition for growth and development. However, in the short run they are unlikely to relieve unemploy- ment. Overall, union involvement, social dialogue, and social peace are essential ingredients for boosting employment Notes I See Geoffrey Heal and Graciela Chichilisky, Oil and the International Economy (Oxford, England Clarendon, 1991), p 74, Raymond Vernon, ed, Thle Oil Criss (New York W W Norton, 1976), and Romano Prod: and Alberto Clo, "Europe," and Robert S Pyndyck and Julio Rotemberg, "Energy Shocks and the Macroeconomy," in Ol Shock Pohlcy, Response, and Implementation, ed Alvin L Alm and RobertJ Einer (Cambridge, Mass Ballinger, 1984) 2 R A Gonzalez and R Nils Folsom, "Responding to the Oil Shock The US Economy since 1973," The Freeman, The Foundation for Economic Education (February 1989) 3 See Pyndyck and Rotemberg, p 110-11 4 Jacques Delors, European Social Policy, Way Fornwardfor the Union, A White Paper, COM (94) 333, July 1994 5 European Commission, Emnployment Rates Report 1998, Employment Performance in the Mernber States, p 16 This report underscored that "taxation has important effects on the functioning of the labor market," since "taxes on labor are often highlighted as one of the main culprits of high unemployment in Europe, as they increase labor costs and ma)' also affect the composition of labor supply and demand " Yet "over the last 15 years the development of taxation systems (taxes and social security contributions) shows a fiscal bias unfavorable to employment-creation in most Member States " On average in the European Union between 1980 and 1996 the burden of taxes 450 I Labor and the Welfare State and charges on labor has increased steadily (from 35 percent to almost 43 percent), whereas it has decreased for other factors of production, mainly capital (from 42 percent to 36 percent) and has remained stable for consumption (close to 14 percent) Taxation has important effects on the functioning of the labor market Taxes on labor are often highlighted as one of the main culprits of high unemployment in Europe, because they increase labor costs and may also affect the composition of labor supply and demand. The incidence of taxes can be magnified if the degree of competition is also low in product markets (since taxes can be easily shifted forward on product prices) and may also depend on the kind of wage-negotiating mechanisms prevail- ing in labor markets. 6 See J I. Perez Infante, "Reformas laborales y creaci6n de empleo en la economia espafnola en el contexto de la um6n monetaria," in Euroy empleo, ed J Aragon Medina (Madnd, 1998), p 150 7. As reported in World Bank, "A Way Forward," prepared for the Bangkok Social Meeting, 1999 . 8. World Bank, Social Issues Arisingfrom the East Asia Economic Crisis A Work in Progress, draft, 1999, p 6 9. A Cox Edwards and C Manning, The Economics of Employment Protection, and Unemployment Insurance Schemes Reflections on Policy Options for Thailand, Malaysia, the Philippines, and Indonesia, World Bank report, October 5, 1999, p. 2. 10 Other active policies, such as wage subsidies, loans for small business start-up, youth training, and in-service retraining for workers whose skills no longer fit their employers' requirements, may be relevant 11. For a discussion of the impact of such policies, see Gordon Betcherman, Amit Dar, Amy Luinstra, and Makoto Ogawa, "Active Labor Market Policies. Policy Issues for East Asia," World Bank, Social Protection Unit, December 1999 The authors question such approaches on the basis of OECD country experience They conclude that, "Ultimately ALMPs [active labor mar- ket policies] are judged by their performance in improving the employability and earnings of workers and the evaluations indicate that the investments made often have little or no impact on these outcomes." However, they also recognize other helpful outcomes of those policies, such as their social impact and integration of economically marginalized and at-risk groups 12 See Cox and Manning, op cit , p 3, who also refer to two major recent research products OECD, "The Adjustment Potential of the Labor Market," part 11 of the OECD3obs Study Evtdence and Explanations (Paris OECD, 1994), and G Saint-Paul, "The Political Economy of Employment Protection," Economics Working Paper No. 355, Universitat Pompeu Fabra, National Center for Research Science, 1999. 13 See the research by Duncan C